Skip to main content

Advances, Systems and Applications

Intelligent acceptance systems for distribution automation terminals: an overview of edge computing technologies and applications


The investigation into intelligent acceptance systems for distribution automation terminals has spanned over a decade, furnishing indispensable assistance to the power industry. The integration of cutting-edge edge computing technologies into these systems has presented efficacious, low-latency, and energy-efficient remedies. This paper provides a comprehensive review and synthesis of research achievements in the field of intelligent acceptance systems for distribution automation terminals over the past few years. Firstly, this paper introduces the definition, composition, functions, and significance of distribution automation terminals, analyzes the advantages of employing edge computing in this domain, and elaborates on the design and implementation of intelligent acceptance systems based on edge computing technology. Additionally, this paper examines the technical challenges, security, and privacy issues associated with the application of edge computing in intelligent acceptance systems and proposes practical solutions. Finally, this paper summarizes the contributions and significance of this paper and provides an outlook on future research directions. It is evident from the review that the integration of edge computing has effectively alleviated these challenges, but new issues await resolution.


With the continuous development and modernization of the power industry, the intelligent acceptance system for distribution automation terminals plays a critical role in ensuring the reliability and security of power supply [1]. The intelligent acceptance system for distribution automation terminals is a system that utilizes advanced technologies and algorithms to automate the inspection, testing, and evaluation of distribution terminal equipment. Traditional acceptance of distribution terminals is typically performed manually, relying on human intervention for the coordination and acceptance between field terminals and distribution control centers [2]. Field personnel manually simulate analog quantities and switch positions according to signal definitions, communicate with dispatch center personnel via telephone on a per-signal basis, and rely on dispatch center personnel to observe system responses and analyze and judge the acceptability of each signal change.

Therefore, traditional manual methods suffer from issues such as lengthy coordination times, a lack of standardization in the acceptance testing process, a lack of rigor and precision in acceptance management, significant burdens on main station personnel during peak periods, and limited accuracy. To overcome the limitations of traditional manual methods, the introduction of an intelligent acceptance system for distribution automation terminals is crucial. This system automates various testing and inspection tasks, enhancing efficiency and accuracy. It can rapidly identify potential issues and defects, providing timely feedback. Automation reduces manual intervention, cutting labor costs, and ensuring consistent acceptance processes. By automating multiple testing and inspection tasks, the intelligent acceptance system for distribution automation terminals significantly shortens the acceptance duration, promptly identifies potential issues and defects, and provides timely feedback. Furthermore, it enhances accuracy by employing advanced algorithms and sensors to precisely evaluate equipment performance, minimizing human influence. Consequently, the intelligent acceptance system significantly enhances the quality and reliability of distribution terminal equipment.

The intelligent acceptance system of distribution automation terminals faces challenges like delays [3], network loads [4], data security [5], and system robustness [6]. Researchers and engineers are increasingly adopting edge computing technology to address these challenges. The intelligent acceptance system of distribution automation terminals typically involves processing and analyzing a large amount of data, and moving computational tasks from traditional centralized servers to mobile edge nodes near the terminal devices can significantly reduce data transmission delays. This enables real-time data processing and decision-making, thereby improving the efficiency and responsiveness of the acceptance process.

Traditional intelligent acceptance systems usually require transmitting a large amount of data to remote servers for processing and analysis. This can increase the network transmission load and potentially impact the performance of other network services. By introducing mobile edge computing, the intelligent acceptance system of distribution automation terminals can perform data processing and analysis locally, thereby alleviating the network load and enhancing the overall system’s reliability and stability [7]. The intelligent acceptance system of distribution automation terminals involves a vast amount of sensitive data, such as power equipment status and power quality. Storing and processing data at the edge nodes can reduce the risks associated with data transmission, minimizing the possibilities of data tampering or leakage. Edge computing offers a more secure and controllable data processing environment, contributing to safeguarding data security and privacy. This system often operates in complex environments with unstable network conditions. By conducting data processing and decision-making at the edge nodes, the system can better cope with network interruptions or unstable connections, ensuring the normal operation of the system. This distributed edge computing architecture provides higher system robustness and reliability.

Edge computing technology shifts computational and data processing tasks from centralized servers to edge nodes near terminal devices [8]. This distributed architecture offers low-latency data processing, reduces network load, and improves system reliability [9]. Leveraging edge computing technology, the intelligent acceptance system of distribution automation terminals can better address challenges related to large data volumes, real-time requirements, and security [10,11,12,13]. These advantages make mobile edge computing an ideal choice for achieving efficient, reliable, and secure intelligent acceptance systems. These intelligent grid solutions provide reliability in power distribution and transmission for developing countries, such as maintaining large-scale power infrastructure without excessive costs [14]. However, there is currently a relatively limited amount of literature that provides a comprehensive overview of edge computing technologies applied to intelligent acceptance systems for distribution automation terminals. Although the application of edge computing technology in the power industry is continuously growing, the research on its specific use in intelligent acceptance systems for distribution automation terminals is still in its early stages. Existing research primarily focuses on the infrastructure, control strategies, data acquisition, and processing of distribution automation systems, and the potential of edge computing technology in intelligent acceptance systems has not been fully explored and investigated. Therefore, the aim of this paper is to address this research gap and provide a comprehensive review of the latest advancements and applications of edge computing technology in intelligent acceptance systems for distribution automation terminals, along with the challenges and prospects that lie ahead.

The contributions of this paper are as follows: Firstly, it fills the research gap in the literature regarding the use of edge computing technology in intelligent acceptance systems for distribution automation terminals. By providing a comprehensive review of the latest advancements and applications, it offers a profound understanding and insights into this field. Secondly, from various perspectives and through concrete case studies and performance evaluations, it reveals the practical potential of edge computing technology in intelligent acceptance systems. These insights shed light on its real-world applications and benefits. Thirdly, the paper addresses the technical challenges, security, and privacy issues faced by edge computing technology in intelligent acceptance systems. It also explores possible solutions and future research directions, offering valuable guidance for both academia and the industry. Lastly, the in-depth review of the latest developments in edge computing technology for intelligent acceptance systems in distribution automation terminals holds practical significance for professionals and researchers in the power industry. It provides valuable technological references and guidance for the intelligent upgrading and development of power systems.

The remainder of this paper is structured and organized as follows. In Overview of Distribution Automation Terminals (DAT) section, we provide a comprehensive introduction to distribution automation terminals. Edge computing for DAT’s intelligent acceptance system section elaborates on the application of edge computing technology in intelligent acceptance systems for distribution automation terminals. Our focus on the design and implementation of intelligent acceptance systems based on edge computing technology is presented in Design and implementation of edge computing-based intelligent acceptance system section. In Challenges and Limitations of Edge Computing in DAT's Intelligent Acceptance System section, we analyze the technical challenges, security, and privacy issues faced by edge computing technology in intelligent acceptance systems, and propose possible solutions and future research directions. Finally, we conclude the paper and provide an outlook on future research directions.

Overview of distribution automation terminals (DAT)

In this section, we provide an overview of distribution automation terminals, a critical component in power distribution systems. We define distribution automation terminals and highlight its functions, emphasizing its importance in ensuring efficient and reliable power distribution.


An automatic process refers to a task that is executed in an automated fashion, leading to enhanced operational efficiency. The notion of distribution automation was initially introduced in the 1970s [15]. The Institute of Electrical and Electronics Engineers (IEEE) has provided a definition for Distribution Automation Systems (DAS) as systems that empower electric utilities to remotely monitor, coordinate, and operate distribution components in real time. [16].With the rapid economic development, there is an increasing demand from users for power supply quality and reliability. Distribution automation systems utilize modern computer, communication, and network technologies to achieve automation in medium-voltage distribution networks, effectively improving power supply quality and production management efficiency. Distribution automation terminals transmit information to the substation or master station of the distribution automation system, while receiving control commands from the substation or master station to remotely operate distribution switches, thereby enabling real-time monitoring, fault detection, fault isolation, and network reconfiguration of the distribution network. In recent years, distribution automation has gained significant attention from national power grids and research and production units.

The distribution automation system consists of three parts: the master station, communication network, and terminal devices [17]. Distribution automation terminals are mainly used for monitoring and controlling ring main units, reclosers, pole-mounted sectional switches, distribution transformers, and other components in medium-voltage distribution networks. They communicate with the distribution automation master station to provide the necessary data for distribution operation, control, and management. The performance and reliability of these terminals directly influence the effectiveness of the entire system [18].


Terminal units are grouped into three types: distribution terminal unit (DTU), feeder terminal unit (FTU), and transformer terminal unit (TTU). They are utilized to oversee operational data linked to feeders, distribution switching stations, and distribution transformers, correspondingly [19].

The first type is the monitoring terminal for switchgear, public, and user distribution substations, commonly referred to as the DTU terminal. Its basic structure is similar to that of traditional remote data terminals, mainly sharing structural similarities, without the need for additional protective devices. This approach saves space and maximizes the utilization of limited resources. Unlike traditional remote data terminals, the DTU terminal features editable logic control functionality, allowing for remote unit monitoring and operation independently of the master station. Additionally, fault detection functionality is added to the traditional terminal technology, enabling the issuance of commands and automatic isolation of the area with faults when the main unit experiences a failure, ensuring power stability [20]. The DTU terminal typically does not have a backup power supply, which simplifies its structure. Furthermore, due to its protective functionality, there is no need to design operational control loops, greatly improving the structural flexibility of the terminal equipment, making it easier to maintain and manage. The schematic representation of the decentralized distributed DTU structure is shown in the Fig. 1.

Fig. 1
figure 1

The structure of a decentralized distributed Terminal Unit (DTU)

The second type is the monitoring terminal for sectionalizing switches along distribution feeders, known as the feeder terminal unit (FTU) [21]. The feeder terminal unit is classified into two types: one type is suspended on outdoor overhead line poles, known as pole-mounted feeder terminal switches. These devices are typically made from corrosion-resistant materials and are housed in rainproof, moisture-proof, and dust-proof enclosures due to the harsh outdoor environment they operate in. The other type is used in ring main units, and their structure and the number of feeder terminals used may differ depending on the specific environmental conditions. Additionally, the installation methods vary between the two cases. If the outdoor ring main unit has conditions suitable for installing feeder terminal units within the unit, there is no need for specialized enclosures. Conversely, protective enclosures with the required functionality are used. However, if space constraints exist within the ring main unit for accommodating the monitoring and control equipment, there is no need to use specialized enclosures, which saves space. The configuration of the pole-mounted feeder terminal unit is shown in the Fig. 2.

Fig. 2
figure 2

The composition of a pole-mounted Feeder Terminal Unit (FTU)

The third type is the monitoring terminal for distribution transformers, commonly known as the transformer terminal unit (TTU). The TTU is an effective device used in the substation automation system to monitor the operational status of transformers. It is a crucial component of the substation automation system, installed as a remote terminal at the bottom of the system to monitor real-time operating parameters of distribution transformers [22]. Its primary function is to record and detect the load conditions of distribution transformers. The TTU can promptly upload the collected data and report the condition of the distribution system based on the actual situation. The structure of the transformer terminal unit is relatively simple and compact, without the need for a battery backup. If the TTU is installed outdoors, a protective enclosure made of corrosion-resistant materials is required. The basic structure of the TTU is depicted in the Fig. 3.

Fig. 3
figure 3

Basic structure diagram of transformer terminal unit (TTU)


The distribution automation terminal primarily consists of five components: the central monitoring unit, the human-machine interface module, the operation control loop, the communication terminal, and the power module [23], as depicted in Fig. 4.

Distribution automation terminals are centered around the central monitoring unit, which performs critical functions including the acquisition of analog and digital input signals, fault detection, calculation of vital operational parameters like voltage, current, and active power, generation of control signals, and facilitating remote communication. These distribution automation terminals currently employ a platform-based and modular design for their output and communication interfaces. This design offers scalability and configurability, enabling tailoring to specific requirements.

Fig. 4
figure 4

The basic components of a distribution automation terminal

The communication terminal, also known as a communication adapter, establishes a connection between the monitoring unit and the distribution network automation communication medium. It can be connected via the monitoring unit’s Ethernet interface or RS232 serial interface. Different types of communication terminals are available, including fiber optic terminals, wireless terminals, modems (for analog channels), and carrier terminals, depending on the communication equipment channel type. The operation control circuit can be utilized in FTU (Feeder Terminal Unit) and requires manual control buttons. This device aids operators in understanding the switch status through circuit displays based on the switch positions.

The power module supplies various DC power sources required by the distribution terminal circuit. For DTU (Distribution Terminal Unit), the external power source is typically obtained from the AC 220 V utility power within the switchgear. In the event of a utility power failure, an uninterruptible power supply (UPS) is utilized as a backup power source. For TTU (Transformer Terminal Unit), the power input is derived from the low-voltage side output of the distribution transformer. In the case of FTU, since there is generally no dedicated AC 220 V power supply along the distribution lines, voltage transformers are commonly used to provide voltage measurement signals and power the FTU simultaneously. The FTU power supply should be equipped with a storage battery to ensure uninterrupted power supply to the terminal’s circuit during line power outages and to provide power for switch operations.


The functional requirements of distribution automation terminals vary depending on the monitoring objects and application scenarios. Therefore, in practical engineering, it is necessary to make choices regarding the following functions based on specific application requirements [24]. The primary functions of the distribution automation terminal are illustrated in Fig. 5.

Fig. 5
figure 5

The new functional structure diagram of the distribution automation system

The Supervisory Control and Data Acquisition (SCADA) function represents the “three remotes” (telemetry, telecontrol, and telecommunication) of the traditional Remote Terminal Unit. The distribution terminal should be able to measure electrical quantities reflecting the system’s imbalance under normal operating conditions, such as voltage, current, active power, reactive power, apparent power, and power factor. It should also be capable of accessing direct current input to monitor the voltage and supply current of the backup battery. Telecommunication mainly involves accessing signals from auxiliary contacts of distribution switches, energy storage unit’s normal operation signal, and so on. Telecontrol includes outputs for closing and tripping distribution switches, as well as switch status outputs [25,26,27].


In recent years, distribution automation terminals have become increasingly important in distribution automation systems. Their significance can be summarized as follows:

Firstly, distribution automation terminals continuously monitor the state, parameters, and performance of the power grid through built-in sensors and measurement devices. They collect a vast amount of data, including key indicators such as voltage, current, power, and frequency. Secondly, distribution automation terminals are equipped with built-in protection devices that can promptly detect faults, short circuits, overloads, and other abnormal conditions within the power system [28]. The fault detection and protection functions of distribution automation terminals are critical components for achieving reliable power supply. Additionally, distribution automation terminals possess remote control and operation capabilities, allowing them to communicate with the higher-level control center or other terminal devices for remote control and operation of grid equipment. This flexibility in remote control enables operators to respond quickly to changes in the power system’s requirements, carry out load adjustments, switch lines, optimize equipment, and achieve efficient operation and optimized management of the power system [29]. Lastly, through the intelligent features of distribution automation terminals, the power system can integrate distributed energy resources, optimize power load management, and engage with the electricity market. The automation capabilities of distribution automation terminals reduce manual intervention, enhance the system’s level of automation and operational efficiency, and extend the lifespan of existing distribution system infrastructure.

The overview of distribution automation terminal presented in this section sheds light on its significance in modern power distribution systems. Understanding the definition, functions, and types of distribution automation terminals provides a foundation for exploring the application of edge computing in distribution automation terminals’ intelligent acceptance system, as discussed in the subsequent sections.

Edge computing for DAT’s intelligent acceptance system

In this section, we delve into the concept of edge computing and its advantages in the context of distribution automation terminals’ intelligent acceptance system. By combining the power of edge computing with the functionalities of distribution automation terminals, we aim to enhance the system’s efficiency, responsiveness, and decision-making capabilities.

Edge computing

In recent years, driven by advancements in technologies like the Internet of Things (IoT), cloud computing, and big data, data volume has surged. Traditional cloud computing models necessitate uploading extensive data to cloud servers. However, due to the gap between cloud servers and terminal devices, challenges like transmission speed, energy use, latency, network interference, and data security become hard to circumvent. Despite the powerful computing capability of cloud computing, which can address the challenges of extensive calculations and device battery consumption, the advancement of smart terminals, new network applications, and evolving user demands for seamless experiences have raised requirements for data transmission speed, low latency, and service quality. Consequently, cloud computing struggles to meet the needs of many technologies and scenarios [30,31,32]. To address the issues of latency and energy consumption caused by the distance between cloud data centers and terminal devices, scholars have proposed shifting cloud functionality to the network edge. Mobile Edge Computing (MEC) is a novel network architecture and computing paradigm that offers information technology services and computational abilities at the edge of mobile networks, in close proximity to terminal mobile devices [33,34,35,36].

In 2014, the concept of MEC was defined as “a new platform that provides IT service environment and cloud computing capabilities at the edge of mobile networks.“ by ETSI (European Telecommunications Standards Institute). In 2016, ETSI expanded MEC to Multi-Access Edge Computing, encompassing multiple access paths extending beyond mobile communication networks to other access networks such as Wi-Fi and wired connections [37]. MEC does not replace cloud computing but rather serves as its extension [38]. Unlike cloud computing, MEC offloads the computational tasks of terminal devices to edge servers closer to the devices. These edge servers can provide computing and content caching functionalities. Edge servers, distributed at the network edge and sometimes referred to as computing or edge nodes, relieve the computational load on terminal devices, curbing interactions with central cloud data centers and notably diminishing message exchange wait times. With a defined storage and computing capacity, these edge servers, positioned closer to terminal devices, enable computation-intensive or latency-sensitive mobile devices to shift their computing tasks for execution at the edge [39, 40].

Figure 6 depicts the fundamental three-layer architecture of MEC. It consists of a three-layer structure: the cloud layer, the edge layer, and the terminal layer. In MEC, mobile terminal devices cannot communicate directly with servers; they need to communicate with MEC servers through base stations or wireless access points in the terminal layer. MEC servers are deployed in the edge layer, closer to the terminal devices, and can provide computing and caching services, thus mitigating the latency and energy consumption issues caused by having all terminal device tasks request services from a remote cloud.

Fig. 6
figure 6

The basic three-layer architecture of MEC

Intelligent acceptance system

The intelligent acceptance system is a key technology in distribution automation terminals, used to ensure the correct installation, configuration, and proper functioning of distribution automation terminals. This system utilizes advanced sensors, communication, and data analysis technologies to comprehensively monitor and assess distribution automation equipment, providing an automated acceptance process. The main aspects of the intelligent acceptance system are as follows:

  1. 1.

    Equipment installation and configuration: The intelligent acceptance system can be used for the installation and configuration of distribution automation equipment. It verifies the correct connection and configuration of the equipment, ensuring its proper functioning. Through automated testing and diagnostics, the intelligent acceptance system can quickly detect and rectify issues in installation and configuration, reducing human errors.

  2. 2.

    Function validation and performance assessment: The intelligent acceptance system can validate the functionality and assess the performance of distribution automation equipment. It can perform various tests and operations, such as sending control commands and detecting device responses. Through real-time data collection and analysis, the intelligent acceptance system can evaluate the performance indicators of the equipment, ensuring compliance with design and specification requirements.

  3. 3.

    Fault diagnosis and early warning: The intelligent acceptance system can conduct fault diagnosis and early warning. It monitors the operational status and parameters of the equipment and identifies potential faults and abnormal conditions by comparing them with pre-defined normal operation modes and behavioral rules. Through timely alerts and alarms, the intelligent acceptance system assists operation and maintenance personnel in quickly identifying and resolving problems, minimizing the impact of faults on the power system.

  4. 4.

    Data analysis and optimization: The intelligent acceptance system can perform data analysis and optimization. It can collect and analyze real-time data, identify potential issues and bottlenecks in the power system using algorithms and models, and provide corresponding optimization recommendations. The data analysis capabilities of the intelligent acceptance system can enhance the efficiency, reliability, and security of the power system.

  5. 5.

    Operation and maintenance management and remote monitoring: The intelligent acceptance system can be integrated with higher-level control centers or other management systems to enable remote operation, monitoring, and management of equipment. Through remote access and control, the intelligent acceptance system enhances the efficiency of maintenance personnel, allowing them to promptly address issues and conduct remote maintenance.

In practical application scenarios, the intelligent acceptance system verifies the correct connection of all cables, ensures sensor calibration accuracy, and checks if communication interfaces are configured according to specifications. If any inconsistencies are detected, the system generates an automated report, highlighting the specific issues. After the installation of remote control devices in a distribution substation, the intelligent acceptance system tests their functionality by sending control commands to open and close circuit breakers. It then monitors the equipment’s response, including response times and sequence of operations. By evaluating these parameters against design specifications, it ensures that circuit breakers perform as expected in real-world situations, such as breaker operations. A utility company employs the intelligent acceptance system to manage and monitor a network of telemetry equipment distributed across various substations. Through a centralized control center, operators can remotely access and control equipment at different substations. For instance, they can remotely close circuit breakers to restore power after temporary faults without dispatching on-site personnel. This reduces downtime and enhances the overall operational efficiency [41].

In summary, the intelligent acceptance system plays a significant role in distribution automation. Through automated testing, diagnostics, and data analysis, it provides comprehensive equipment acceptance and performance assessment, supporting equipment installation, configuration, and fault handling. Applying the intelligent acceptance system can bolster the power system’s reliability, efficiency, and security, offering vital support to distribution automation systems’ operation and management.

Application of edge computing in intelligent acceptance system

In recent times, edge computing has assumed a pivotal role in the intelligent acceptance system of distribution automation terminals (DAT). This section will delve into the utilization of edge computing within the intelligent acceptance system of DAT.

Firstly, edge computing provides real-time data processing capabilities to the intelligent acceptance system. As early as 2017, Nastic et al. [42] proposed a real-time data processing platform applied in the context of the Internet of Things (IoT). By performing local data processing on edge devices, the system can rapidly analyze data and make timely decisions. Secondly, edge computing enables secure local data storage and caching for the intelligent acceptance system. For instance, paper [31] proposed a blockchain-based data security storage method in an edge computing scenario. Utilizing this capability, the power system can securely store and cache relevant data on edge devices, reducing the demand for data transmission to the cloud. This facilitates quick access to historical data, real-time analysis, and comparisons, thereby enhancing the efficiency and flexibility of the acceptance system. Furthermore, edge computing empowers the intelligent acceptance system with offline operation and autonomy, allowing the system to operate independently even when disconnected from the cloud [10]. Lastly, edge computing delivers edge analysis and machine learning capabilities to the intelligent acceptance system. Sophisticated analysis and machine learning algorithms can be deployed on edge devices for real-time data analysis [11], anomaly detection [12], and fault prediction [13].

Edge computing holds significant application value in the intelligent acceptance system of distribution automation terminals. It offers capabilities such as real-time data processing, local data storage and caching, offline operation and autonomy, edge analysis and machine learning, and bandwidth optimization. These applications bring higher efficiency, flexibility, and reliability to the intelligent acceptance system, providing robust support for the acceptance work of distribution automation terminals.

Limitations of traditional intelligent acceptance system

Although the intelligent acceptance system played a pivotal role in distribution automation terminals, the limitations of traditional intelligent acceptance systems have restricted their widespread adoption and application in several aspects.

One problem with traditional intelligent acceptance systems is data processing latency. Relying on cloud-based data processing and analysis leads to elongated data transmission and cloud processing durations, thereby constraining the system’s responsiveness [43]. This latency can be further increased in scenarios that require large-scale data processing and complex algorithm execution. Such delays may prevent acceptance personnel from timely accessing critical real-time data and analysis results, adversely affecting the accurate assessment and decision-making regarding the status of power equipment [44].

Traditional intelligent acceptance systems typically rely on cloud resources for data storage, computation, and analysis [45, 46]. Handling large amounts of data transmission and storage requires high bandwidth and large-capacity cloud resources, which increases costs and introduces complex network management. For real-time acceptance requirements, the time delay in cloud processing may not meet the system’s demands, thereby affecting the accuracy and efficiency of acceptance. Furthermore, traditional intelligent acceptance systems that depend on cloud resources face challenges in ensuring data privacy and security requirements, especially considering the sensitive nature of the multidimensional data in power systems.

Traditional intelligent acceptance systems have high bandwidth requirements. In large-scale distribution automation terminals, a tremendous amount of data is generated, particularly when it involves high-frequency sampling and sensor data. Traditional systems need to transmit this data to the cloud for processing and storage, which places high demands on network bandwidth [47]. However, in network environments with limited bandwidth, such as remote areas or mobile network environments, this high bandwidth requirement may result in data transmission delays and instability.

Undoubtedly, traditional intelligent acceptance systems have various limitations, and these limitations are also the advantages offered by intelligent acceptance systems that employ edge computing technology.

Benefits of using edge computing

Based on edge computing, intelligent acceptance systems have the following advantages compared to traditional intelligent acceptance systems:

Improved acceptance efficiency: Edge computing technology significantly enhances the efficiency of intelligent acceptance systems. By performing real-time data processing and analysis on edge devices, the system reduces the time delay of transmitting data to the cloud for processing [48]. This enables acceptance personnel to promptly access critical real-time data and analysis results, facilitating quick assessment of the status of power equipment and accurate decision-making.

Reduced dependence on the cloud: Edge computing-based intelligent acceptance systems decrease reliance on cloud resources. By processing and analyzing data locally on the devices, the system reduces the need for data transmission and computation with the cloud. This not only lowers communication latency with the cloud but also reduces dependence on cloud resources [49]. By enabling local computation and decision-making at the edge, intelligent acceptance systems enhance system independence and reliability while reducing requirements for network connectivity to the cloud and associated costs.

Optimized bandwidth usage: Edge computing technology optimizes the bandwidth usage of intelligent acceptance systems. By performing real-time data processing and analysis on edge devices, the system can filter and aggregate data locally, transmitting only relevant information or insightful results to the cloud. This bandwidth optimization reduces data transmission volume and lowers the demand for network bandwidth, thereby improving system performance and scalability. By effectively utilizing limited network resources, intelligent acceptance systems can operate more efficiently and reduce operational costs.

Challenges and limitations of using edge computing

Despite the advantages of edge computing technology in the intelligent acceptance system of distribution automation terminals (DAT), its application also faces several challenges and limitations, including:

Limited computing resources: Edge devices often feature restricted processing power, memory, and storage capacity when compared to cloud servers. This limitation could curtail the extent and intricacy of data processing and analysis attainable on edge devices. Therefore, it is essential to optimize algorithms and develop efficient resource management techniques to allocate available resources properly, ensuring the effective utilization of limited computing resources. Allocating computing and network resources amid diverse or fluctuating circumstances poses a demanding endeavor [50]. To tackle this obstacle, numerous scholars have put forth assorted approaches. For instance, a Joint Communication and Computation (JCC) resource allocation method has been recommended. This method aims to fulfill resource requisites on Mobile Edge Computing (MEC) servers, aligning with user specifications, given that the necessary resources are accessible on MEC servers [51].

Network resource reliability: Edge computing relies on network connections to transfer data between edge devices and central systems. However, in certain deployment environments, network connections may be unstable, intermittent, or subject to delays. Due to limited edge computing server resources, congestion may occur when there is a large number of requests and an increase in data traffic. The primary reason for congestion is the limitation of server resources. This is an important issue in edge computing networks that can be addressed through the following methods: (1) Traffic buffering: When network capacity is at maximum and resources are in use, data packets are stored and queued until the buffer capacity is reached [52]; (2) Intelligent resource allocation: In this technique, limited resources are used to support incoming requests one by one [53].

Data security and privacy: Edge computing involves data processing and storage on edge devices, which elevates the significance of data security and privacy protection. Edge devices are susceptible to physical security risks and the potential for data leakage [54]. Using different encryption techniques can reduce such risks, but it may increase processing time and decrease application performance [55].

System management and maintenance: As the intelligent acceptance system of distribution automation terminals involves the collaborative work of multiple edge devices and central systems, system management and maintenance can become complex. Ensuring the stable operation of edge devices, software updates, and troubleshooting are challenging tasks.

These challenges and limitations need to be addressed through appropriate technical solutions and effective system design to overcome them, ensuring the effective application and optimal performance of edge computing in the intelligent acceptance system of distribution automation terminals.

The exploration of edge computing for DAT’s intelligent acceptance system highlights its potential to revolutionize traditional methods and enable real-time data processing and analysis. The advantages of edge computing in this domain set the stage for discussing the design and implementation of an intelligent acceptance system powered by edge computing in the next section.

Design and implementation of edge computing-based intelligent acceptance system

In this section, we introduce the architecture and components of an edge computing-based intelligent acceptance system for distribution automation terminals. We discuss the selection of edge devices and sensors, data collection, processing, and the application of machine learning algorithms to optimize the acceptance process.

Architecture and components

The edge computing-based intelligent acceptance system for distribution automation consists of three layers: cloud, edge, and device. Each layer performs specific functions, working together to ensure the system’s efficient and stable operation, as shown in Fig. 7.

Fig. 7
figure 7

Intelligent Acceptation System

Cloud layer

The cloud layer serves as the core management and data processing center of the intelligent acceptance system. It is responsible for the following functions:

Data Storage and Management: The cloud layer handles the storage and management of large-scale data, including acceptance data, historical data, and system configuration information. By utilizing high-performance databases and storage systems, the cloud layer efficiently stores and retrieves data, providing support for subsequent data analysis and decision-making [56].

Data Analysis and Algorithms: The cloud layer utilizes powerful computing capabilities and machine learning algorithms to analyze and process collected data. It can execute complex data analysis and mining algorithms, achieving intelligent identification and prediction of power equipment status, performance indicators, and abnormal situations [57].

Decision Support and Optimization: Based on data analysis results, the cloud layer provides decision support and optimization suggestions. The system can generate detailed acceptance reports, fault diagnosis reports, and equipment performance evaluation reports, assisting operations personnel in making accurate decisions [58].

Edge layer

The edge layer serves as an intermediary between the cloud layer and the device layer, encompassing the following functionalities:

Data Preprocessing and Filtering: The edge layer preprocesses and filters raw data from the device layer to minimize data transmission volume and alleviate cloud processing loads. It can perform simple data cleaning, denoising, and sampling operations, enhancing data quality and availability [59].

Real-time Data Analysis and Decision-making: The edge layer has a certain computing and analytical capability, allowing real-time analysis and decision-making on preprocessed data. It can execute rapid data analysis algorithms and rule engines to achieve real-time monitoring, fault detection, and warning functions for power equipment.

Data Caching and Buffering: The edge layer can locally cache a portion of data to provide fast data access and response capabilities. It can intelligently manage data caching and buffering based on system requirements and resource constraints, reducing frequent access and transmission to the cloud layer [60].

Device layer

The device layer comprises the actual equipment of distribution automation terminals, including sensors, measurement instruments, and controllers. It possesses the following functions:

Data Collection and Transmission: The device layer is responsible for real-time data collection from power equipment, including parameters such as current, voltage, power, and temperature. It transmits the data to the edge layer or cloud layer for processing and analysis through sensors and measurement instruments [61].

Control and Execution: The device layer executes control operations on power equipment based on system instructions and control strategies. It can achieve automated switch control, protection actions, and fault handling functions, ensuring the safe operation and fault recovery capability of the power system.

Device Status Monitoring and Maintenance: The device layer can monitor the real-time status and operational parameters of power equipment, detecting the health status and fault conditions of the equipment. Through self-checking and self-diagnostic functions, it provides real-time feedback and anomaly reports on equipment status, assisting maintenance personnel in equipment maintenance and fault troubleshooting.

Selection of edge devices and sensors

Before selecting edge devices and sensors, a thorough analysis and definition of the intelligent acceptance system’s requirements are necessary. This includes functional requirements, data collection needs, computing capability requirements, communication requirements, as well as reliability and security requirements. By clarifying the system’s needs, it can better guide the selection of edge devices and sensors.

Selection of edge devices

The selection of edge devices should be evaluated based on the system requirements and available resources, considering the following factors:

Computing Capability: Edge devices should have sufficient computing power to handle real-time data processing, analysis, and decision-making tasks. Depending on the system’s complexity and data volume, appropriate hardware and software components such as processors, memory, and operating systems need to be chosen. High-performance edge devices can better meet the demands for complex algorithms and real-time data processing [62].

Network Connectivity: Due to the rapid growth of IoT devices, IoT faces several communication-related challenges in edge computing networks. One key issue in these networks is how to effectively choose IoT edge computing devices when multiple edge nodes are available to transmit information. Edge devices should support stable network connections to facilitate data exchange and communication with the cloud layer and other edge devices. When selecting edge devices, factors such as network interfaces, transmission rates, and protocol compatibility need to be considered. A higher network connectivity performance ensures timely data transmission and communication reliability. To overcome this problem, Shafig et al. [63] introduced a novel framework called SoftSystem that utilizes the soft set technique to suggest relevant IIoT devices.

Reliability and Stability: Edge devices should have high reliability and stability to operate continuously in uninterrupted operating environments. Choosing high-quality and low-failure-rate edge devices ensures system stability and reliability. Additionally, the temperature range and environmental adaptability of edge devices need to be considered to cope with various harsh working conditions.

Manageability: Select edge devices that are easy to manage and maintain for system configuration, monitoring, and troubleshooting. Edge devices should have features such as remote management, software update support, and fault diagnosis capabilities, facilitating remote management and maintenance operations. This improves system manageability and maintenance efficiency.

Selection of sensors

The selection of sensors is a critical step based on data collection needs and the system’s monitoring objectives [64]. The following factors need to be considered:

Measurement Parameters: Select appropriate types of sensors based on the power equipment parameters that need to be measured by the system. For example, current sensors, voltage sensors, temperature sensors, etc. Depending on different monitoring objectives, choose sensors with corresponding measurement parameters to meet the monitoring requirements for power equipment status.

Accuracy and Sensitivity: Choose sensors with sufficient accuracy and sensitivity to ensure data quality and accuracy. The resolution, error range, and sensitivity of sensors directly impact the accuracy and reliability of data collection.

Data Acquisition Rate: Depending on the system’s real-time requirements, select sensors that can provide fast data acquisition. Consider factors such as the sensor’s sampling rate and response time to ensure the sensor meets the system’s requirements for real-time data.

Reliability and Durability: Choose sensors with high reliability and durability, capable of operating for an extended period in harsh environments. Consider factors such as the sensor’s protection level, anti-interference capability, and lifespan to ensure stable operation of the sensor under various working conditions.

By comprehensively considering the characteristics of edge devices and sensors and the system’s requirements, the most suitable combination can be selected to achieve accurate monitoring and data collection of power equipment in the intelligent acceptance system. Furthermore, considering system scalability and cost-effectiveness, standardized devices and sensors can be considered to achieve better interoperability and maintainability.

Data collection and processing

Data acquisition is a crucial step in the intelligent acceptance system, used to obtain real-time data from distribution automation terminal devices. Edge computing provides the intelligent acceptance system with real-time data processing capabilities, enabling data acquisition to take place at the edge devices. Edge devices can be equipped with various types of sensors, such as current sensors, voltage sensors, temperature sensors, etc., to collect different parameters from power equipment.

Edge computing has distinct advantages in data collection and processing. Firstly, it pushes data processing closer to the data source, reducing data transmission latency [48]. This means that data can be processed closer to where it is generated, resulting in faster response times. Secondly, edge devices can perform real-time data processing and analysis tasks, extracting valuable information according to system requirements, without the need to send all data to central servers for processing [49]. This distributed data processing approach enhances system efficiency.

As the volume of power data continues to grow, advanced data acquisition and processing methods are crucial for the intelligent acceptance system. MEC is an architectural technology that offers a computing paradigm and IoT services from centralized servers to distributed network edges. In the context of MEC, the management of data streams is handled by proximate edge nodes. This approach leads to a significant augmentation in computational capacity and a reduction in data transmission latency [65]. To illustrate, Zhou and colleagues [66] leveraged deep learning techniques to validate the accuracy of sensor data. They established a MEC-driven framework for local data processing, ensuring consistent computation and adaptable data conveyance.

Machine learning algorithms

Deep learning has gained significant attention in recent years, thanks to its exceptional capabilities [67]. These algorithms are known for their robustness, generality, scalability, and adaptive learning, all of which have played a pivotal role in advancing modern sustainable power systems [4, 68].

Numerous studies have demonstrated the outstanding performance of deep learning algorithms in the power system. For instance, in [69], researchers proved the effectiveness of deep learning algorithms in tackling complex power system problems. By learning patterns and features from large-scale datasets, deep learning can extract valuable information, including power load forecasting, power equipment fault diagnosis, and power system security assessment, among others. The application of deep learning algorithms in these tasks has yielded remarkable results, playing a vital role in enhancing power system operational efficiency and reducing the risk of faults.

Machine learning algorithms are playing an increasingly vital role in various fields, particularly in IoT domain [70]. Machine learning algorithms can also be applied to fault diagnosis in intelligent acceptance systems. Faults in distribution systems can lead to power outages, and issues such as short circuits, overloads, and human errors can result in significant losses. Fault detection is an analytical process that enables the rapid identification of the root cause of a problem based on power operation data and configuration data when subsequent issues arise. This facilitates the prompt resolution and restoration of normal power system operation. In [71], a method combining wavelet transform and DNN (Deep Neural Network) is proposed to provide fault types, fault phases, and fault locations in microgrid systems. Compared to traditional methods, the proposed approach demonstrates more accurate prediction results. In [72], an adaptive Convolutional Neural Network (CNN) approach for fault diagnosis is proposed, aiming to locate faults in distribution networks. This method boasts advantages such as short computation time and high accuracy/speed in fault selection. Additionally, in [73], the authors introduce a CNN algorithm for obtaining fault types and locations within distribution systems. Simulation results demonstrate the precise performance of CNN in comparison to other techniques like Support Vector Machine (SVM). Additionally, in [74], a method for fault recognition in the voltage sampling module of distribution terminals is introduced. This method employs a combination of Generative Adversarial Network (GAN) and CNN. The GAN model is employed to generate patterns and learn from developed samples, subsequently enhancing the CNN’s fault detection accuracy significantly.

Integration with existing DAT infrastructure

The existing distribution automation infrastructure is a critical component in the power system, comprising a series of components and functional modules designed to achieve monitoring, control, and management of the distribution system [75, 76]. The distribution automation infrastructure includes various vital equipment and systems, such as intelligent electronic devices [77], monitoring devices [78], data acquisition systems [79], and communication networks [80].

In modern power systems, various distribution automation infrastructures play crucial roles. With the increasing power demand, integration of renewable energy, and challenges in grid security, the distribution system becomes more complex. The combination of edge computing with distribution automation infrastructure provides the power system with more intelligent, automated, and sustainable operation and management capabilities. This integration not only enhances system efficiency and reliability but also provides essential support for the development and sustainable growth of the power system.

Integrating edge computing technology with the existing distribution automation infrastructure can further enhance system intelligence and performance. In modern power systems, deploying edge devices such as edge servers or edge nodes at the edge of the distribution network allows real-time data collection and processing, tightly integrated with sensors and monitoring devices. Furthermore, harnessing the computational capabilities of edge devices allows for the offloading of certain data processing and analysis tasks from the cloud to these edge devices [81, 82]. This reduces data transmission latency to the cloud, achieving real-time data processing and decision-making, thereby enhancing system response speed and efficiency. The distribution system can conduct data filtering and preprocessing on edge devices, reducing data transmission to the cloud. Additionally, edge devices can have storage space for data storage and caching, reducing reliance on cloud storage resources and improving data access speed and efficiency.

Implementation details and challenges

To combine edge computing with distribution automation terminals and integrate it into the intelligent acceptance system, a comprehensive system architecture must be designed, incorporating edge computing as part of the distribution automation terminal infrastructure. This architecture should clearly define the data flow and communication methods between edge devices and cloud systems, ensuring secure and reliable data transmission. Additionally, the deployment locations and quantities of edge devices should be planned based on the distribution system’s requirements and topology. Furthermore, to seamlessly integrate edge devices with existing sensors and monitoring equipment for real-time data collection and transmission, sensor networks and communication technologies can be utilized to facilitate efficient and security data exchange and communication between edge devices and various nodes within the distribution system [83]. This enables real-time monitoring and data acquisition of power equipment status. An effective collaborative mechanism between edge devices and cloud systems should be established, allowing edge devices to transmit processed data to the cloud for further analysis and decision-making, while the cloud system can send instructions and configuration parameters to edge devices. This collaborative effort between edge and cloud allows for advanced data processing and system control. Lastly, when integrating edge computing technologies, security and privacy preservation are key considerations. To ensure data security and prevent unauthorized access as well as data leakage, it is crucial to implement suitable encryption and authentication mechanisms. An effective solution, for example, is to adopt the data security framework proposed by Chadwick et al. [84].

Integrating edge computing technology with existing distribution automation terminal infrastructure involves work in system architecture design, data collection and transmission, introducing edge computing capabilities, data storage and caching, edge-cloud collaboration, as well as security and privacy protection. Such integration can enhance the intelligence level of distribution automation systems and improve system reliability, real-time performance, and overall efficiency.

The design and implementation of the edge computing-based intelligent acceptance system for distribution automation terminals demonstrate its potential to enhance system efficiency and decision-making capabilities. The integration of edge computing with the existing distribution automation terminal infrastructure and the provided case studies contribute significantly to the advancement of intelligent acceptance systems in power distribution.

Challenges and limitations of edge computing in DAT’s intelligent acceptance system

This section addresses the challenges and limitations that edge computing may encounter when applied to distribution automation terminals’ intelligent acceptance system. We discuss technical constraints, reliability issues, power consumption, and security and privacy concerns that warrant attention for successful implementation.

Technical challenges and limitations

Latency and bandwidth constraints

When applying edge computing to intelligent acceptance systems, there are still challenges and limitations that need to be addressed. One of these challenges is data processing latency and bandwidth. Due to the relatively limited computational and storage capabilities of edge devices, edge computing nodes may not be able to process and analyze large amounts of data as quickly as the cloud [85]. This results in data processing latency issues, affecting the real-time capability and responsiveness of the system. Another related limitation is bandwidth constraints. Edge devices typically communicate with the cloud or other edge nodes through limited network bandwidth. When a large amount of data needs to be transferred between edge devices and the cloud, bandwidth constraints can become a bottleneck, causing data transfer speeds to slow down [86]. This can lead to increased data processing latency, potentially affecting real-time capabilities and system performance. Such constraints and limitations can have adverse effects on the reliability and performance of intelligent acceptance systems, particularly with regard to real-time data processing and analysis, which are essential components of these systems. Therefore, the presence of processing latency and bandwidth constraints may prevent acceptance personnel from obtaining real-time data and accurate analysis results in a timely manner.

Reliability and availability issues

The extensive use of distribution automation terminals not only enhances the operational management level of distribution networks but also improves power supply quality. However, practical applications often encounter issues such as low online rates, frequent crashes, and data security concerns, seriously affecting the normal operation of distribution automation systems. Addressing these issues has become a top priority for power supply enterprises [87].

When applying edge computing to intelligent acceptance systems, reliability and availability are other important aspects that need to be considered and addressed [88]. Reliability refers to the stability and dependability of the system when facing faults, errors, or exceptional situations. Since edge computing involves multiple distributed nodes and devices, communication and coordination between nodes may encounter instability and errors. This may lead to issues such as data transmission loss, processing errors, or system crashes, thereby affecting the reliability of the intelligent acceptance system. Availability refers to the system’s ability to continuously provide services. Edge computing nodes may be affected by factors such as power failures, network interruptions, or device malfunctions, causing nodes to function improperly or fail to provide the required services. This can result in the unavailability of the intelligent acceptance system, preventing acceptance personnel from accessing data and making timely decisions.

Power consumption and heat dissipation

During urbanization in developing nations, power consumption demand has surged alongside a substantial increase in the count of IoT devices across manufacturing and usage sectors [89]. The proliferation of IoT devices and escalating data processing needs intensify the energy demands on the power system. Edge devices’ computation and communication actions can lead to elevated energy consumption, a concern that escalates as the system expands. Moreover, during intensive computing, edge devices may generate a considerable amount of heat, necessitating effective heat dissipation measures to maintain stable device operation. Energy consumption issues can potentially impact the long-term sustainability and operational costs of intelligent acceptance systems [90]. High energy consumption implies more frequent power supply or larger battery capacity requirements, significantly increasing the operational costs of the system. Additionally, higher energy consumption can have adverse effects on the environment, which is a critical concern that could potentially be addressed by providing sustainable and renewable energy sources [91]. The heat dissipation issue arises from the intensive computing of edge devices. During extensive data processing and the execution of intricate algorithms, edge devices have the potential to generate considerable heat. Prolonged exposure to elevated temperatures can result in device overheating, consequently diminishing the system’s performance and reliability [92]. Therefore, effective heat dissipation measures are crucial factors in ensuring the proper functioning of the devices.

These challenges and limitations necessitate a series of mitigation measures when applying edge computing to intelligent acceptance systems. Firstly, exploring optimized data processing algorithms, offloading certain tasks to the cloud for processing, or employing higher-performance edge devices can alleviate data processing latency and address bandwidth limitations. Secondly, ensuring system reliability and availability is of paramount importance, achievable through measures such as device redundancy and backup configurations, fault detection, and automatic switchover. Additionally, regarding power consumption and heat dissipation concerns, consideration can be given to the use of more energy-efficient hardware devices and the adoption of renewable energy sources to mitigate the environmental impact of energy consumption.

Security and privacy concerns

Data confidentiality and integrity

Intelligent acceptance systems handle substantial volumes of sensitive data, encompassing power equipment status, monitoring data, and analysis outcomes. Thus, safeguarding data privacy and integrity becomes pivotal to uphold user rights and sustain system credibility [93]. Data privacy refers to the assurance that sensitive data is not accessible, acquired, or tampered with by unauthorized individuals or entities during processing and transmission. Edge computing involves multiple distributed nodes and devices, and data transmission and processing between different nodes may face security threats and risks, such as data leaks, theft, or tampering. Thus, to protect data privacy, it’s essential to employ measures like data encryption, access control, and identity authentication [94]. Data integrity refers to the assurance that data is not accidentally or maliciously altered, damaged, or lost during transmission and processing. In the edge computing environment, data transmission and processing involving multiple nodes and devices may encounter situations such as transmission errors, packet loss, or node failures, which can affect data integrity. Therefore, measures such as data verification, error detection, and fault tolerance mechanisms need to be implemented to ensure data integrity.

Data confidentiality and integrity are crucial security aspects within intelligent acceptance systems. In the context of edge computing, these aspects encounter specific challenges, including complex management and resource constraints. Nonetheless, these challenges can be surmounted by implementing suitable technologies and strategies, thereby ensuring data security and bolstering system reliability [95]. As technology continues to evolve, we can anticipate witnessing further innovative solutions in intelligent acceptance systems that address these challenges while concurrently satisfying data privacy and integrity requisites.

Authentication and access control

Identity authentication and access control play a critical role in intelligent acceptance systems to ensure that only authorized users or devices can access and operate system resources and data. Appropriate identity authentication and access control mechanisms can prevent unauthorized access and potential malicious activities, thereby safeguarding system security and data integrity [96]. However, when edge computing is applied to intelligent acceptance systems, although identity authentication and access control are important measures to ensure data security and system trustworthiness, edge computing also brings some challenges and limitations. Firstly, the distributed nature of the edge computing environment increases the complexity of identity authentication and access control. Due to the diversity of edge devices and nodes, there exist different identity authentication mechanisms and access control policies. Therefore, how to unify the management and coordination of identity authentication and access control mechanisms distributed across edge nodes becomes a challenge. Secondly, resource limitations of edge nodes may impact the implementation of identity authentication and access control. Edge devices often possess constrained computing power and storage capacity, potentially limiting the use of intricate identity authentication and access control algorithms. In the edge computing environment, lightweight authentication and authorization schemes need to be designed to achieve efficient identity authentication and access control in resource-constrained environments.

Compliance with regulations and standards

Intelligent acceptance systems handle sensitive data and operational information of power equipment, thus requiring adherence to relevant regulations and legal standards to ensure data security, privacy, and legality. Compliance requirements encompass a wide range of areas, including data protection, privacy preservation, and network security. Depending on the regulations and standards of different regions and industries, intelligent acceptance systems may need to comply with a series of regulations and legal requirements. To adhere to these legal regulations, it’s crucial to adopt suitable security measures, encompassing data encryption, access control, and identity authentication. This ensures the preservation of data confidentiality, integrity, and availability.

In the edge computing environment, implementing compliance and legality requirements poses certain challenges. Firstly, the distribution and heterogeneity of edge nodes increase the complexity of management. Given the number and diversity of edge nodes, it is essential to ensure that all nodes comply with legal requirements, including security configurations, access control, and log recording, among others. Secondly, the changing and updating of legal regulations also present challenges for the edge environment. As relevant regulations and standards evolve and update, intelligent acceptance systems need to promptly adjust and update their security measures to meet the latest requirements. Regular compliance assessments and reviews are essential to maintain consistent adherence to regulatory and legal mandates in the system.

Potential solutions and future research directions

Edge-cloud collaboration

Edge computing is an architecture that performs distributed processing and storage at the data source, offering numerous advantages in data processing and computation [97]. Solutions that include edge computing can typically reduce communication latency, improve network scalability, and enhance information accessibility, thus enabling more agile and efficient business development. Edge nodes are deployed at the edge side, and various terminal devices access the platform through the edge side, placing higher demands on edge-side resources. The edge-cloud computing architecture consists of numerous edge servers and terminals that require unified management through the edge cloud to support edge applications.

Edge-cloud synergy is a potential solution to address the challenges faced by edge computing in intelligent acceptance systems. By enabling coordinated work between edge devices and cloud resources, their respective strengths can be fully utilized. As an example, edge devices can handle real-time data processing and analysis tasks, while cloud resources can offer enhanced computing capabilities and expanded storage capacity [31, 98]. Through optimizing data flow and task allocation to achieve collaboration between the edge and the cloud, the performance and efficiency of intelligent acceptance systems can be improved.

To enhance intelligent acceptance systems, optimizing their performance is crucial. Achieving this involves close collaboration between edge computing and cloud computing to cater to varying application scenarios. This synergy entails resource coordination, application management, data integration, and intelligent operations. Resource synergy combines edge nodes’ computing, storage, network, virtualization, and other infrastructure resources to facilitate network services. Application management synergy refers to the deployment and runtime environment of network applications provided by edge nodes, managing and scheduling multiple application lifecycles on the nodes. Data synergy involves edge nodes taking charge of data collection at the edge, performing initial data processing and analysis based on predefined models, and subsequently uploading the results to the cloud. Intelligent synergy encompasses the execution of inference using intelligent models on edge nodes. This enables distributed intelligence to be achieved [99].

In the context of the power industry, edge-cloud synergy technology, as shown in Fig. 8, conducts decision-making on edge-side data through value mining, data computation, and data application from bottom to top, uploading it to the cloud center. The cloud center controls the edge-side data with instructions, enabling collaborative bidirectional interaction between user edge-side data and the power grid cloud center.

Fig. 8
figure 8

Technical architecture of edge-cloud collaboration

Demand response business is an essential service that facilitates the interaction between the power supply side and the user side. It involves user-side participation in response to pricing or incentives, leading to changes in consumption patterns. This participation plays a crucial role in reducing peak loads on the power grid and ensuring safe and stable operations. The adjustment of electricity prices or incentive mechanisms by the power supply side relies on relevant data from supply-demand interactions. Therefore, the application of edge-cloud synergy technology holds significant significance for optimizing demand-side resources. Electricity analysis primarily involves the analysis of various aspects of electricity consumption, faults, repair efficiency, peak power consumption, among others, encompassing scenarios associated with multiple data analyses. Fault warning focuses on predicting equipment failures, power consumption loads, and potential electricity usage conditions. Electricity scheduling assesses the power grid’s safety and operational status based on data or monitoring information provided by various information collection devices, combined with power grid operational parameters, and subsequently makes adjustments to the system. The data application of edge-cloud synergy encompasses various aspects of data analysis, prediction, and control for equipment and systems.

Hybrid edge-cloud architecture

When developing energy-saving solutions, various approaches can be taken to store and process energy footprints. Cloud computing platforms are the most widely used paradigms for this purpose [100, 101]. Despite the slight delays in implementing real-time energy efficiency applications, cloud computing technology is garnering increasing attention [102, 103]. Meanwhile, edge computing has emerged as a more favorable solution to address the challenges encountered in cloud computing, particularly as an alternative or supplementary method for processing energy data at the Internet of Things (IoT) edge to mitigate latency issues [104, 105]. Nevertheless, edge computing still demands additional electrical power for independent utilization in order to fulfill the high computational requirements of AI-based energy-saving solutions. Consequently, the hybrid edge-cloud architecture is presently considered the most optimal trend for implementing intelligent acceptance systems in distribution automation [106, 107]. Figure 9 illustrates the hybrid edge-cloud architecture.

Fig. 9
figure 9

Hybrid edge-cloud architecture

Additionally, in recent years, the analysis, detection, and visualization of abnormal energy consumption and behavior in distribution automation systems have received increasing attention. This is because such research is crucial in developing efficient energy-saving systems [108]. For example, in [109], supervised and unsupervised anomaly detection methods were introduced, while in [110], an energy anomaly detection approach based on using micro-moments and deep learning models was proposed and compared with various traditional machine learning models. Although these technologies can achieve high accuracy in anomaly detection, their main challenge lies in the computational resources required for implementation and operation. Consequently, researchers have explored various architectures, including cloud-based, edge-based, and edge-cloud hybrid solutions [110].

Although most energy-saving solutions in distribution automation smart acceptance systems currently utilize cloud computing for data collection, preprocessing, and analysis of energy data, edge computing has garnered increasing interest. Nevertheless, edge computing still requires more electrical power for individual use to meet the high computational demands of AI-based energy-saving solutions. At the same time, the hybrid edge-cloud architecture emerges as a promising approach in current energy-efficient systems implementation. It provides flexible control over energy usage in distribution automation smart acceptance systems, minimizes cloud hosting costs, and enhances privacy protection. For instance, in [111], authors propose a novel energy-efficient system based on a hybrid edge-cloud computing architecture. In the edge-cloud collaboration, high concurrency in offloading deep learning tasks can overload edge servers and lead to unacceptable latency. Optimal offloading decisions are challenging in dynamic, heterogeneous edge-cloud environments. To tackle these concerns, Xu et al. [112] introduced GPOV (Game theory-based Partitioning and Offloading with CNN, a dynamic offloading approach that fuses game theory and CNN. The CNN partitioning allows for more efficient resource utilization and reduces latency through parallelism.

Blockchain-based security and privacy protection

In recent times, blockchain distributed technology has surfaced as a remedy for fortifying the architecture of enterprise systems and hierarchical network operations. This technology introduces attributes like integrity, transparency, privacy, and security [113]. Through the creation and implementation of distributed applications (DApps), accessing information becomes more convenient, and blockchain-backed APIs amplify the effectiveness and dependability of designing distributed information systems [114,115,116]. In the context of smart grids and distribution environments, blockchain can be utilized to protect intelligent acceptance systems, encrypt information, and record node execution details and events. The storage based on blockchain relies entirely on its immutability, preserving the ledger for transparent information investigations [117]. Furthermore, blockchain, which supports a distributed ledger environment, provides a chronologically ordered structure to record transaction execution events and distribution fluctuations in the smart grid nodes. To facilitate execution and delivery, two different channels are deployed in alliance public/private networks. However, the independent execution of distribution tasks within the smart grid is accomplished using chaincode scripts, orchestrating various operations through DApps for scheduling, control, management, and organization. This decentralized distributed technology bolsters the security and privacy of the ledger, allowing the creation of a transparent ecosystem that resists tampering and forgery [118].

The assessment of challenges and limitations in the application of edge computing to DAT’s intelligent acceptance system calls for proactive measures to mitigate latency, enhance reliability, and address security and privacy concerns. The potential solutions and future research directions proposed in this section offer promising avenues for overcoming these obstacles.

Conclusion and future work

This paper presents a comprehensive review of the latest advancements and applications of edge computing technologies in intelligent acceptance systems for distribution automation terminals. Initially, we introduce the definition and functions of distribution automation terminals, emphasizing their significance in power system operation and management. Subsequently, we delve into the definition and advantages of edge computing, providing examples and explanations of its applications in intelligent acceptance systems, particularly highlighting its advantages in data analysis and decision support. Furthermore, we focus on the design and implementation of intelligent acceptance systems based on edge computing technologies, including system architecture, device and sensor selection, data acquisition and processing, and the application of machine learning algorithms. When discussing technical challenges, security, and privacy issues, we identify the challenges faced by edge computing in intelligent acceptance systems and propose potential solutions. In particular, we emphasize the importance of edge-cloud collaboration, hybrid edge-cloud architectures, innovative edge devices and sensors, advanced machine learning algorithms, and blockchain-based security and privacy protection in future research.

Intelligent acceptance systems have extensive application prospects and significant development potential, especially in the field of distribution automation and power. In the future, they will evolve in several key directions:

Firstly, intelligent acceptance systems will collaborate more with other intelligent systems, such as smart distribution grids and energy management systems. This collaboration will enable broader data sharing and cooperative control, thereby enhancing the overall efficiency and reliability of the power system. Secondly, with the continuous advancement of artificial intelligence and machine learning technologies, intelligent acceptance systems will become more intelligent. They will have the capability to conduct in-depth data analysis, automatically identify potential issues, and provide more intelligent fault detection and diagnostics, ultimately reducing maintenance time and costs. Sustainability and energy efficiency are critical topics in the future. Intelligent acceptance systems will adopt more energy-efficient hardware and green energy sources to reduce energy consumption and minimize their environmental impact. The application of blockchain technology will ensure data security and integrity, particularly in data sharing and exchange. This will help establish trustworthy data records and auditing mechanisms. Finally, as the level of system intelligence increases, network security and data privacy protection will become crucial focus areas. Future systems will employ stronger security measures to defend against potential threats. These directions will collectively drive the widespread application and continuous development of intelligent acceptance systems in the field of power and beyond.

In the development of intelligent acceptance systems, edge computing technologies will continue to play a pivotal role. To propel further advancements in this field, several key research directions merit exploration:

Enhancing the Performance and Efficiency of Edge Computing: With the continuous increase in data volume and complexity, improving the performance and efficiency of edge computing is a critical concern. Researchers can explore novel edge device and sensor technologies and optimize edge computing algorithms and models to enhance system responsiveness and processing capabilities.

Strengthening Edge-Cloud Collaboration: Edge-cloud collaboration is crucial to enhancing the performance and reliability of intelligent acceptance systems. Future research can explore improved collaboration mechanisms and data transfer schemes to facilitate efficient cooperation and data sharing between edge devices and the cloud.

Enhancing Security and Privacy Protection: Intelligent acceptance systems involve vast amounts of sensitive data, making security and privacy paramount. Researchers can further explore security solutions based on blockchain and encryption technologies to ensure data confidentiality and integrity.

Developing Autonomous Decision-Making Capabilities for Intelligent Acceptance Systems: Future research can explore the application of autonomous decision-making technologies such as reinforcement learning in intelligent acceptance systems, enabling the system to learn from experience, optimize acceptance strategies, and adapt automatically to different devices and environments.

Leveraging Emerging Technologies: With continuous technological advancements, intelligent acceptance systems can leverage emerging technologies like the IoT, big data analysis, and AI to further enhance their performance and functionality.

The application of edge computing technologies in intelligent acceptance systems holds tremendous potential. By improving the performance of edge computing, enhancing security protection, strengthening edge-cloud collaboration, and advancing autonomous decision-making capabilities, we are poised to achieve more precise, efficient, and reliable intelligent acceptance systems, providing robust technical support for the acceptance work of distribution automation terminals.

Availability of data and materials

The data that support the findings of this study are available from the corresponding author upon reasonable request.


  1. Xuan GL, Wang W, Liu Y, Lang B (2020) The reliability evaluation of power supply distribution system based on distribution automation. J Phys Conf Ser 1585(1):012034

    Google Scholar 

  2. Yang CH, Vyatkin V, Pang C (2013) Model-driven development of control software for distributed automation: a survey and an approach. IEEE Trans Syst Man Cybern Syst 44(3):292–305

    Google Scholar 

  3. Wollschlaeger M, Sauter T, Jasperneite J (2017) The future of industrial communication: automation networks in the era of the internet of things and industry 4.0. IEEE Industr Electron Mag 11(1):17–27

    Google Scholar 

  4. Jafari M, Kavousi-Fard A, Dabbaghjamanesh M, Karimi M (2022) A survey on deep learning role in distribution automation system: a new collaborative Learning-to-Learning (L2L) concept. IEEE Access. 10:81220–81238

  5. Suo S, Huang K, Kuang X, Cao Y, Chen L, Tao W (2021) Communication Security Design of Distribution Automation System with Multiple Protection. Proc. ICCECE, Guangzhou, pp 750–754

    Google Scholar 

  6. Ukil A, Deck B, Shah VH (2012) Current-only directional overcurrent protection for distribution automation: challenges and solutions. IEEE Trans Smart Grid 3(4):1687–1694

    Google Scholar 

  7. Zhao ZC, Zhao R, Xia JJ, Lei XF, Li D, Yuen C, Li S, Fan (2019) A novel framework of three-hierarchical offloading optimization for MEC in industrial IoT networks. IEEE Trans Industr Inf 16(8):5424–5434

    Google Scholar 

  8. Zhou XH, Bilal M, Dou RH, Rodrigues JJ, Zhao QZ, Dai JG, Xu XL (2023) Edge computation offloading with content caching in 6G-enabled IoV. IEEE Trans Intell Transp Syst.

    Article  Google Scholar 

  9. Xu XL, Yang CY, Bilal M, Li WM, Wang HH (2022) Computation offloading for energy and delay trade-offs with traffic flow prediction in edge computing-enabled IoV. IEEE Trans Intell Transp Syst.

    Article  Google Scholar 

  10. Liu Y, Yang C, Jiang L, Xie SL, Zhang Y (2019) Intelligent edge computing for iot-based energy management in smart cities. IEEE Netw 33(2):111–117

    Google Scholar 

  11. Hosseini MP, Tran TX, Pompili D, Elisevich K, Soltanian-Zadeh H (2020) Multimodal data analysis of epileptic EEG and rs-fMRI via deep learning and edge computing. Artif Intell Med 104:101813

    Google Scholar 

  12. Hussain B, Du QH, Imran A, Imran MA (2020) Artificial intelligence-powered mobile edge computing-based anomaly detection in cellular networks. IEEE Trans Industr Inf 16(8):4986–4996

    Google Scholar 

  13. Bellavista P, Penna RD, Foschini L, Scotece D (2020) Machine learning for predictive diagnostics at the edge: an IIoT practical example. Proc. ICC, Dublin, pp 1–7

    Google Scholar 

  14. Jamil F, Iqbal N, Imran S, Ahmad, Kim D (2021) Peer-to-peer energy trading mechanism based on blockchain and machine learning for sustainable electrical power supply in smart grid. IEEE Access 9:39193–39217

    Google Scholar 

  15. Pahwa A (2005) Planning and Analysis Tools to evaluate distribution automation implementation and benefits. IEEE Power Eng Soc Gen Meet 3:2622–2623

    Google Scholar 

  16. Gruenemeyer D (1991) Distribution automation: How should it be evaluated? Proc. 1991 Rural Electric Power Conference, Dearborn, p C3/1-C310

    Google Scholar 

  17. Mohagheghi S, Mousavi M, Stoupis J, Wang Z (2009) (2009) “Modeling distribution automation system components using IEC 61850.” Proc. IEEE Power and Energy Society General Meeting, Calgary, AB, Canada, pp 1–6

    Google Scholar 

  18. Jusoh WNSEW, Ghani MRA, Hanafiah MAM, Raman SH (2014) “Remote Terminal Unit (RTU) hardware design and development for distribution automation system,” in Proc. ISGT ASIA, Kuala Lumpur, p 572–576

  19. Zhou XS, Xiang PD, Ma YJ, Gao ZQ, Wu YJ, Yin JL, Xu XN (2016) An overview on distribution automation system. Proc. CCDC, Yinchuan, pp 3667–3671

    Google Scholar 

  20. Zhang XC, Tang Y, Liu Q, Liu GF, Ning X, Chen JK (2021) A fault analysis method based on association rule mining for distribution terminal unit. Appl Sci 11(11):5221

    Google Scholar 

  21. Han GZ, Xu BY, Suonan J (2012) IEC 61850-based feeder terminal unit modeling and mapping to IEC 60870-5-104. IEEE Trans Power Del 27(4):2046–2053

    Google Scholar 

  22. Kilishi IM, Muhammad KB, Bello A, Alhasan I (2014) Hardware design of transformer terminal unit (TTU) based on substation automation. IOSR J Electr Electron Eng 9(3):51–66

    Google Scholar 

  23. Liu J (1998) “Management of distribution automation system (DAS)“. Power System Technology-Beijing 22:27–30

    Google Scholar 

  24. Nordell DE (2008) “Communication systems for Distribution Automation,” in Proc. 2008 IEEE/PES Transmission and Distribution Conference and Exposition, Chicago, IL, USA, pp. 1–14

  25. Koizumi S, Okumura M, Yanase T (2005) Application and development of distribution automation system in TEPCO. Proc. IEEE Power Engineering Society General Meeting, San Francisco, CA, USA, pp 2429–2435

    Google Scholar 

  26. Kropp T (2006) System threats and vulnerabilities– an EMS and SCADA security system overview. IEEE Power Energ Mag 31(3):46–50

    Google Scholar 

  27. Creery A, Byres EJ (2005) Industrial cybersecurity for power system and SCADA networks. Proc. Record of Conference Papers Industry Applications Society 52nd Annual Petroleum and Chemical Industry Conference, Denver, pp 303–309

    Google Scholar 

  28. Ciontea CI, Iov F (2021) A study of load imbalance influence on power quality assessment for distribution networks. Electricity 2(1):77–90

    Google Scholar 

  29. Hou Z, Sun X, Ju ZL, Li Z, Niu QB, Sl L (2021) Design of the monitoring system for power distribution area by clean energy intelligent integration terminals. IOP Conf Ser Earth Environ Sci 804(3):032051

    Google Scholar 

  30. Zhao HL, Deng SG, Liu ZJ, Yin JW, Dustdar S (2022) Distributed redundant placement for microservice-based applications at the edge. IEEE Trans Serv Comput 15(3):1732–1745

    Google Scholar 

  31. Yuan L, He Q, Chen FF, Zhang J, Qi LY, Xu XL, Xiang Y, Yang Y (2021) CSEdge: enabling collaborative edge storage for multi-access edge computing based on blockchain. IEEE Trans Parallel Distrib Syst 33(8):1873–1887

    Google Scholar 

  32. Zhang YW, Wang KB, He Q, Chen FF, Deng SG, Zheng ZB, Yang Y (2019) Covering-based web service quality prediction via neighborhood-aware matrix factorization. IEEE Trans Serv Comput 14(5):1333–1344

    Google Scholar 

  33. Deng SG, Xiang ZZ, Taheri J, Khoshkholghi MA, Yin J, Zomaya AY, Dustdar S (2021) Optimal application deployment in resource constrained distributed edges. IEEE Trans Mob Comput 20(5):1907–1923

    Google Scholar 

  34. Wu JT, Zhang JY, Zhang YW, Wen YP (2023) Constraint-aware and multi-objective optimization for micro-service composition in mobile edge computing. Software: Pract Experience.

  35. Zhang YW, Pan J, Qi LY, He Q (2021) Privacy-preserving quality prediction for edge-based IoT services. Futur Gener Comput Syst 114:336–348

    Google Scholar 

  36. Cui GM, He Q, Chen FF, Jin H, Yang Y (2020) Trading off between user coverage and network robustness for edge server placement. IEEE Trans Cloud Comput 10(3):2178–2189

    Google Scholar 

  37. Sabella D, Vaillant A, Kuure P, Rauschenbach U, Giust F (2016) Mobile-edge computing architecture: the role of MEC in the internet of things. IEEE Consum Electron Mag 5(4):84–91

    Google Scholar 

  38. Shirazi SN, Gouglidis A, Farshad A, Hutchison D (2017) The extended cloud: review and analysis of mobile edge computing and fog from a security and resilience perspective. IEEE J Sel Areas Commun 35(11):2586–2595

    Google Scholar 

  39. Lin L, Liao XF, Jin H, Li P (2019) Computation Offloading Toward Edge Computing. Proc IEEE 107(8):1584–1607

    Google Scholar 

  40. Mach P, Becvar Z (2017) Mobile edge computing: a survey on architecture and computation offloading. IEEE Commun Surv Tutor 19(3):1628–1656

    Google Scholar 

  41. Kumar A, Abhishek K, Ghalib MR, Shanka A, Cheng X (2022) Intrusion detection and prevention system for an IoT environment. Digit Commun Netw 8(4):540–551

    Google Scholar 

  42. Nastic S, Rausch T, Scekic O, Dustdar S, Gusev M, Koteska B, Prodan R (2017) A serverless real-time data analytics platform for edge computing. IEEE Internet Comput 21(4):64–71

    Google Scholar 

  43. Rodrigues TG, Suto K, Nishiyama H, Kato N (2017) Hybrid method for minimizing service delay in edge cloud computing through VM migration and transmission power control. IEEE Trans Comput 66(5):810–819

    MathSciNet  Google Scholar 

  44. Muyizere D, Letting LK, Munyazikwiye BB (2022) Effects of communication signal delay on the power grid: a review. Electronics 11(6):874

    Google Scholar 

  45. Xu C, Zhao F, Wang ZC, Lin XN, He S, Shao C (2013) Design of cloud computing architecture for power system analysis. Proc. TENCON 2013, Xi’an, pp 1–4

    Google Scholar 

  46. Fang BL, Yin X, Tan Y, Li CB, Gao YP, Cao YJ, Li JL (2016) The contributions of cloud technologies to smart grid. Renew Sustain Energy Rev 59:1326–1331

    Google Scholar 

  47. Azodolmolky S, Wieder P, Yahyapour R (2013) Cloud computing networking: challenges and opportunities for innovations. IEEE Commun Mag 51(7):54–62

    Google Scholar 

  48. Mao YY, You CS, Zhang J, Huang KB, Letaief KB (2017) A survey on mobile edge computing: the communication perspective. IEEE Commun Surv Tutor 19(4):2322–2358

    Google Scholar 

  49. Zhang K, Leng SP, He YJ, Maharjan S, Zhang Y (2018) Mobile edge computing and networking for green and low-latency internet of things. IEEE Commun Mag 56(5):39–45

    Google Scholar 

  50. Cui GM, He Q, Xia XY, Chen FF, Dong F, Jin H, Yang Y (2023) Ol-eua: online user allocation for noma-based mobile edge computing. IEEE Trans Mob Comput 22(4):2295–2306

    Google Scholar 

  51. Cui GM, He Q, Xia XY, Chen FF, Gu T, Jin H, Yang Y (2023) Demand response in NOMA-based mobile edge computing: a two-phase game-theoretical approach. IEEE Trans Mob Comput 22(3):1449–1463

    Google Scholar 

  52. Nasimi M, Habibi MA, Han B, Schotten HD (2018) “Edge-Assisted Congestion Control Mechanism for 5G Network Using Software-Defined Networking”, in Proc. ISWCS, Lisbon, Portugal, pp. 1–5.

  53. Wang JD, Zhao L, Liu J, Kato N (2021) Smart resource allocation for mobile edge computing: a deep reinforcement learning approach. IEEE Trans Emerg Top Comput 9(3):1529–1541

    Google Scholar 

  54. Zheng Y, Li Z, Xu XL, Zhao QZ (2022) Dynamic defenses in cyber security: techniques, methods and challenges. Digit Commun Netw 8(4):422–435

    Google Scholar 

  55. Ahmed E, Rehmani MH (2017) Mobile edge computing: opportunities, solutions, and challenges. Futur Gener Comput Syst 70:59–63

    Google Scholar 

  56. Wang C, Ren K, Lou WJ, Li J (2010) Toward publicly auditable secure cloud data storage services. IEEE Netw 24(4):19–24

    Google Scholar 

  57. Agrawal D, Das S, Abbadi AE (2011) Big data and cloud computing: current state and future opportunities. Proc. EDBT/ICDT ‘11, New York, pp 530–533

    Google Scholar 

  58. Demirkan H, Delen D (2013) Leveraging the capabilities of service-oriented decision support systems: putting analytics and big data in cloud. Decis Support Syst 55(1):412–421

    Google Scholar 

  59. Lv ZH, Qiao L, Verma S, Kavita (2021) AI-enabled IoT-edge data analytics for connected living. ACM Trans Internet Technol 21(4):1–20

    Google Scholar 

  60. Xia XY, Chen FF, He Q, Grundy J, Abdelrazek M, Jin H (2021) Online collaborative data caching in edge computing. IEEE Trans Parallel Distrib Syst 32(2):281–294

    Google Scholar 

  61. Plageras AP, Psannis KE, Stergiou C, Wang H, Gupta BB (2018) Efficient IoT-based sensor BIG data collection–processing and analysis in smart buildings. Futur Gener Comput Syst 82:349–357

    Google Scholar 

  62. Chowdhery A, Levorato M, Burago I, Baidya S (2018) “Urban iot edge analytics,” Fog Computing in the Internet of Things: Intelligence at the Edge. 6:101–120

  63. Shafiq M, Tian Z, Bashir AK, Cengiz K, Tahir A (2020) SoftSystem: smart edge computing device selection method for IoT based on soft set technique, Wirel Commun Mob Comput 2020:1–10

  64. Elijah O, Rahman TA, Orikumhi I, Leow CY, Hindia MN (2018) An overview of internet of things (IoT) and data analytics in agriculture: benefits and challenges. IEEE Internet Things J 5(5):3758–3773

    Google Scholar 

  65. Xia XY, Zhou Y, Li J, Yu RY (2019) Quality-aware sparse data collection in MEC-enhanced mobile crowdsensing systems. IEEE Trans Comput Soc Syst 6(5):1051–1062

    Google Scholar 

  66. Zhou ZY, Liao HJ, Gu B, Huq KMS, Mumtaz S, Rodriguez J (2018) “Robust mob crowd sensing: when deep learn meets edge computing. IEEE Netw 32(4):54–60

    Google Scholar 

  67. Wu JT, Yang GJ, Yang H, Zhu YH, Li ZH, Lei L, Zhao CJ (2020) Extracting apple tree crown information from remote imagery using deep learning. Comput Electron Agric 174:105504

    Google Scholar 

  68. Xu ZY, Zhu DW, Chen JH, Yu BH (2022) Splitting and placement of data-intensive applications with machine learning for power system in cloud computing. Digit Commun Netw 8(4):476–484

    Google Scholar 

  69. Ozcanli AK, Yaprakdal F, Baysal M (2020) Deep learning methods and applications for electrical power systems: a comprehensive review. Int J Energy Res 44(9):7136–7157

    Google Scholar 

  70. Wu JT, Yang GJ, Yang XD, Yu B, Han L, Zhu YH (2019) Automatic counting of in situ rice seedlings from UAV images based on a deep fully convolutional neural network. Remote Sens 11(6):691

    Google Scholar 

  71. James JQ, Hou Y, Lam AYS, Li VOK (2019) Intelligent fault detection scheme for microgrids with wavelet-based deep neural networks. IEEE Trans Smart Grid 10(2):1694–1703

    Google Scholar 

  72. Liang J, Jing T, Niu H, Wang J (2020) Two-terminal fault location method of distribution network based on adaptive convolution neural network. IEEE Access 8:54035–54043

    Google Scholar 

  73. Guo MF, Zeng XD, Chen DY, Yang NC (2018) Deep-learning-based earth fault detection using continuous wavelet transform and convolutional neural network in resonant grounding distribution systems. IEEE Sens J 18(3):1291–1300

    Google Scholar 

  74. Liu JH, Qu FM, Hong XW, Zhang HG (2019) A small-sample wind turbine fault detection method with synthetic fault data using generative adversarial nets. IEEE Trans Industr Inf 15(7):3877–3888

    Google Scholar 

  75. His PH, Chen SL (1998) Distribution automation communication infrastructure. IEEE Trans Power Del 13(3):728–734

    Google Scholar 

  76. Li H, Rosenwald GW, Jung J, Liu CC (2005) Strategic power infrastructure defense. Proc IEEE 93(5):918–933

    Google Scholar 

  77. Zavoda F (2008) The key role of intelligent electronic devices (IED) in advanced Distribution Automation (ADA). Proc. 2008 China International Conference on Electricity Distribution, Guangzhou, pp 1–7

    Google Scholar 

  78. Sharmila AM, Raj SS (2014) A smart distribution automation using supervisory control and data acquisition with advanced metering infrastructure and GPRS technology. Int J Eng Res Gen Sci 2(5):191–196

    Google Scholar 

  79. Nunoo S, Ofei AK (2010) Distribution automation (DA) using supervisory control and data acquisition (SCADA) with advanced metering infrastructure (AMI). Proc. IEEE Conference on Innovative Technologies for an Efficient and Reliable Electricity Supply, Waltham, pp 454–458

    Google Scholar 

  80. Kume S, Rizzi AA (2001) “A high-performance network infrastructure and protocols for distributed automation”, in Proc. ICRA, Seoul, pp. 3121–3126.

  81. Wang T, Lu YC, Cao ZH, Shu L, Zheng X, Liu AF, Xie MD (2019) When sensor-cloud meets mobile edge computing. Sensors 19(23):5324

    Google Scholar 

  82. Xu XL, Liu ZJ, Bilal M, Vimal S, Song HB (2022) Computation offloading and service caching for intelligent transportation systems with digital twin. IEEE Trans Intell Transp Syst 23(11):20757–20772

    Google Scholar 

  83. Wang F, Li GS, Wang YL, Rafique W, Khosravi MR, Liu GF, Liu YW, Qi* LY (2022) Privacy-aware traffic flow prediction based on multi-party sensor data with zero trust in Smart City. ACM Trans Internet Technol.

    Article  Google Scholar 

  84. Chadwick DW, Fan WJ, Costantino G, Lemos RD, Cerbo FD, Herwono I, Manea M, Mori P, Sajjad A, Wang XS (2020) A cloud-edge based data security architecture for sharing and analysing cyber threat information. Futur Gener Comput Syst 102:710–722

    Google Scholar 

  85. Tian H, Xu XL, Lin TY, Cheng Y, Qian C, Ren L, Bilal M (2022) DIMA: distributed cooperative microservice caching for internet of things in edge computing by deep reinforcement learning. World Wide Web 25(5):1769–1792

    Google Scholar 

  86. Li Y, Xia SC, Zheng MY, Cao B, Liu QL (2022) Lyapunov optimization-based trade-off policy for mobile cloud offloading in heterogeneous wireless networks. IEEE Trans Cloud Comput 10(1):491–505

    Google Scholar 

  87. Gupta RP, Srivastava SC (2004) Technology development and implementation for power distribution automation system. Water Energy Intern 61(4):40–47

    Google Scholar 

  88. Momoh JA (2017) Electric power distribution, automation, protection, and control. CRC Press

    Google Scholar 

  89. Khan AA, Shaikh AA, Shaikh ZA, Laghari AA, Karim S (2022) IPM-Model: AI and metaheuristic-enabled face recognition using image partial matching for multimedia forensics investigation with genetic algorithm. Multimed Tools Appl 81(17):23533–23549

    Google Scholar 

  90. Zhou XK, Yang X, Ma JH, Wang KI (2022) Energy-efficient smart routing based on link correlation mining for wireless edge computing in IoT. IEEE Internet Things J 9(16):14988–14997

    Google Scholar 

  91. Kim SK, Huh JH (2018) “A study on the improvement of smart grid security performance and blockchain smart grid perspective. Energies 11(8):1973

    Google Scholar 

  92. Deng D, Li XW, Menon V, Piran MJ, Chen H, Jan MA (2022) Learning-based joint UAV trajectory and power allocation optimization for secure IoT networks. Digit Commun Netw 8(4):415–421

    Google Scholar 

  93. Lim IH, Hong S, Choi MS, Lee SJ, Kim TW, Lee SW, Ha BN (2010) Security protocols against cyber attacks in the distribution automation system. IEEE Trans Power Deliv 25(1):448–455

    Google Scholar 

  94. Qi LY, Yang YH, Zhou XK, Rafique W, Ma JH (2022) Fast anomaly identification based on multiaspect data streams for intelligent intrusion detection toward secure industry 4.0. IEEE Trans Industr Inf 18(9):6503–6511

    Google Scholar 

  95. Almagrabi AO, Bashir AK (2022) A classification-based privacy-preserving decision-making for secure data sharing in internet of things assisted applications. Digit Commun Netw 8(4):436–445

    Google Scholar 

  96. Zhou XK, Liang W, Li WM, Yan K, Shimizu S, Wang KI (2022) Hierarchical adversarial attacks against graph-neural-network-based IoT network intrusion detection system. IEEE Internet Things J 9(12):9310–9319

    Google Scholar 

  97. Yang CY, Xu XL, Zhou XK, Qi LY (2022) Deep Q network-driven task offloading for efficient multimedia data analysis in edge computing-assisted IoV. ACM Trans Multimedia Comput Commun Appl 18(2s):124

    Google Scholar 

  98. Yao JC, Zhang SY, Yao Y et al (2022) Edge-cloud polarization and collaboration: a comprehensive survey for Ai. IEEE Trans Knowl Data Eng 35(7):6866–6886

    Google Scholar 

  99. Xue M, W HM, Peng G, Wolter K (2022) DDPQN: an efficient DNN offloading strategy in local-edge-cloud collaborative environments. IEEE Trans Serv Comput 15(2):640–655

    Google Scholar 

  100. Sayed A, Alsalemi A, Himeur Y, Bensaali F, Amira A (2021) Endorsing energy efficiency through accurate appliance-level power monitoring, automation and data visualization. Proc. NISS, Singapore, pp 603–617

    Google Scholar 

  101. Himeur Y, Alsalemi A, Bensaali F, Amira A (2021) Detection of appliance-level abnormal energy consumption in buildings using autoencoders and micro-moments. Proc. International Conference On Big Data and Internet of Things. Springer International Publishing, Cham, pp 179–193

    Google Scholar 

  102. Metallidou CK, Psannis KE, Egyptiadou EA (2020) Energy efficiency in smart buildings: IoT approaches. IEEE Access 8:63679–63699

    Google Scholar 

  103. Himeur Y, Alsalemi A, Al-Kababji A, Bensaali F, Amira A, Sardianos C, Dimitrakopoulos G, Varlamis I (2021) A survey of recommender systems for energy efficiency in buildings: principles, challenges and prospects. Inform Fusion 72:1–21

    Google Scholar 

  104. Sardianos C, Varlamis I, Chronis C, Dimitrakopoulos G, Himeur Y, Alsalemi A, Bensaali F, Amira A (2020) Data analytics, automations, and micro-moment based recommendations for energy efficiency. Proc. BigDataService, Oxford, UK, pp 96–103

    Google Scholar 

  105. Alsalemi A, Himeur Y, Bensaali F, Amira A, Sardianos C, Varlamis I, Dimitrakopoulos G (2020) Achieving domestic energy efficiency using micro-moments and intelligent recommendations. IEEE Access 8:15047–15055

    Google Scholar 

  106. Reiter A, Prünster B, Zefferer T (2017) Hybrid mobile edge computing: unleashing the full potential of edge computing in mobile device use cases. Proc. CCGRID, Madrid, pp 935–944

    Google Scholar 

  107. Sittón-Candanedo I, Alonso RS, García Ó, Gil AB, Rodríguez-González S (2019) A review on edge computing in smart energy by means of a systematic mapping study. Electronics 9(1):48

    Google Scholar 

  108. Himeur Y, Ghanem K, Alsalemi A, Bensaali F, Amira A (2021) Artificial intelligence based anomaly detection of energy consumption in buildings: a review, current trends and new perspectives. Appl Energy 287:116601

    Google Scholar 

  109. Himeur Y, Alsalemi A, Bensaali F, Amira A (2021) Smart power consumption abnormality detection in buildings using micromoments and improved k-nearest neighbors. Int J Intell Syst 36(6):2865–2894

  110. Himeur Y, Alsalemi A, Bensaali F, Amira A (2020) A novel approach for detecting anomalous energy consumption based on micro-moments and deep neural networks. Cogn Comput 12(6):1381–1401

    Google Scholar 

  111. Himeur Y, Alsalemi A, Bensaali F, Amira A (2021) The emergence of hybrid edge-cloud computing for energy efficiency in buildings. Proc. SAI Intelligent Systems Conference. Springer International Publishing, Cham, pp 70–83

    Google Scholar 

  112. Xu XL, Tang SZ, Qi LY, Zhou XK, Dai F, Dou WC (2023) CNN partitioning and offloading for vehicular edge networks in Web3. IEEE Commun Mag.

    Article  Google Scholar 

  113. Singh P, Masud M, Hossain MS, Kaur A (2021) Blockchain and homomorphic encryption-based privacy-preserving data aggregation model in smart grid. Comput Electr Eng 93:107209

    Google Scholar 

  114. Khan AA, Shaikh AA, Laghari AA (2023) IoT with multimedia investigation: a secure process of digital forensics chain-of-custody using blockchain hyperledger sawtooth. Arab J Sci Eng 48(8):10173–10188

  115. Shaikh ZA, Khan AA, Teng L, Wagan AA, Laghari AA (2022) BIoMT modular infrastructure: the recent challenges, issues, and limitations in blockchain hyperledger-enabled e-healthcare application. Wire Commun Mobile Comput 2022:1–14

    Google Scholar 

  116. Xu XL, Gu J, Yan HZ, Liu WT, Qi LY, Zhou XK (2023) Reputation-aware supplier assessment for blockchain-enabled supply chain in industry 4.0. IEEE Trans Industr Inf 19(4):5485–5494

    Google Scholar 

  117. Khan AA, Laghari AA, Shafiq M, Awan SA, Gu Z (2022) Vehicle to everything (V2X) and edge computing: a secure lifecycle for UAV-assisted vehicle network and offloading with blockchain. Drones 6(12):377

    Google Scholar 

  118. Khan AA, Laghari AA, Rashid M, Li H, Javed AR, Gadekallu TR (2023) Artificial intelligence and blockchain technology for secure smart grid and power distribution automation: a state-of-the-art review. Sustain Energy Technol Assess 57:103282

    Google Scholar 

Download references


This work was supported by the Guangxi Power Grid Technology Project (No. 041400KK52220002).

Author information

Authors and Affiliations



Mingzhen Liang conceived the paper, designed its structure, and drafted the initial manuscript. Mingzeng Zhu collected reference materials and organized the literature review. Hefeng Li and Ying Lu supervised and guided the paper’s completion, overseeing content validation and proofreading to ensure accuracy. Min Pang was responsible for preparing the final manuscript and overall coordination of the work. All authors discussed the results and contributed to the preparation of the final manuscript. All authors reviewed the manuscript.

Corresponding author

Correspondence to Mingzhen Liang.

Ethics declarations

Ethics approval and consent to participate

The research has consent for Ethical Approval.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Zhu, M., Liang, M., Li, H. et al. Intelligent acceptance systems for distribution automation terminals: an overview of edge computing technologies and applications. J Cloud Comp 12, 149 (2023).

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: