Large-scale distributed photovoltaic integration and collaboration method and system
By coordinating distributed energy resources through self-optimization and online duality algorithms, the stability and flexibility issues of virtual power plants in large-scale distributed energy management are solved, achieving efficient energy consumption and grid flexibility services.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- STATE GRID JIANGSU ELECTRIC POWER CO LTD RESEARCH INSTITUTE
- Filing Date
- 2025-11-17
- Publication Date
- 2026-04-21
AI Technical Summary
Existing virtual power plant technology struggles to efficiently manage large-scale, geographically dispersed, and diverse distributed energy resources, leading to challenges in grid stability and flexibility, and high curtailment rates.
By employing a self-optimizing approach, and coordinating the operation of distributed energy resources through event-driven control and online primal dual algorithms, a bottom-up virtual power plant system is formed, achieving efficient access and collaborative control.
It maintains near-optimal operation in complex and dynamic environments, combining simplicity and feasibility, reducing curtailment rates, increasing renewable energy absorption rates, and ensuring reliable power supply.
Smart Images
Figure CN121906601A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to a method and system for large-scale distributed photovoltaic integration and coordination, belonging to the fields of virtual power plant technology, distributed photovoltaic management, and grid flexibility services. Background Technology
[0002] The global energy transition is driving the widespread integration of renewable energy into the power grid, fundamentally reshaping the global power system. As of 2023, renewable energy accounted for more than 30% of global electricity generation, and this proportion is projected to exceed 50% by 2030. However, the inherent volatility of renewable energy sources such as wind and solar power poses significant challenges to grid stability and reliability. Traditional power systems, designed around centralized generation units, struggle to provide the flexibility required for high renewable energy penetration, resulting in curtailment rates as high as 20% in some regions.
[0003] Distributed energy resources offer a viable solution to this flexibility challenge. These user-side resources include distributed generation equipment, energy storage systems, electric vehicles (EVs), and controllable loads, potentially providing hundreds of gigawatts of flexibility capacity globally. Virtual power plants aggregate these diverse distributed energy resources into a coordinated whole, providing key grid flexibility services such as peak shaving and frequency regulation. However, existing virtual power plant technologies still face many challenges in efficiently managing large-scale, geographically dispersed, diverse, and complex distributed energy resources operating under complex conditions. Summary of the Invention
[0004] The purpose of this invention is to overcome the shortcomings of existing technologies and provide a method and system for large-scale distributed photovoltaic integration and coordination, which can achieve efficient access and aggregation of distributed energy resources over a wide area; at the same time, it adopts a self-optimizing method to coordinate the operation of distributed energy resources, enabling them to adapt to real-time operating conditions and evolve towards the optimal state. Ultimately, a bottom-up virtual power plant system is formed, realizing efficient access and coordinated control of diverse distributed energy resources.
[0005] To achieve the above objectives, the present invention is implemented using the following technical solution: In a first aspect, the present invention provides a method for large-scale distributed photovoltaic integration and coordination, comprising: Operational data from multiple distributed energy resources are collected at the terminal device layer; Operational data is transmitted to the data domain via the edge layer, where information sharing is achieved based on a publish-subscribe model. In the data domain, based on the operational data, an event-driven control framework is used to detect whether the system state deviates from the optimal domain of multiple indicators, and an event signal is generated when a deviation occurs. In response to the event signal, an operation strategy for distributed energy resources is generated in the cloud using a self-optimization method, wherein the self-optimization method employs an online primal-dual algorithm. The operation strategy is distributed to the terminal device layer through the data domain and edge layer to control the operation of distributed energy resources.
[0006] Furthermore, the collection of operational data from multiple distributed energy resources through the terminal device layer includes: Power, status, and environmental data of distributed energy resources are collected at a sub-second sampling rate through the terminal device layer. The terminal equipment layer includes distributed photovoltaics, energy storage systems, electric vehicle charging piles, variable frequency air conditioners, lighting loads, and environmental sensors.
[0007] Furthermore, the transmission of operational data to the data domain via the edge layer includes: The smart energy gateway performs local aggregation and protocol conversion of operational data, wherein the smart energy gateway is based on the MQTT protocol. The transformed runtime data is published to the data domain, which broadcasts data using a publish-subscribe model based on the MQTT / MQTT-SN protocol.
[0008] Furthermore, the step of detecting whether the system state deviates from the optimal domain of multiple indicators through an event-driven control framework includes: Define a multi-indicator optimization domain, including virtual power plant system-level indicators and distributed energy resource-level indicators; Monitor the system status in real time and calculate the gap between the current state and the optimal domain; When the gap exceeds a set threshold, an event signal is generated, wherein the event signal is used to trigger a self-optimization process.
[0009] Furthermore, the method for generating a distributed energy resource operation strategy using a self-optimizing approach in the cloud includes: An online primal-dual algorithm based on virtual queues is adopted, which uses virtual queues and Lyapunov optimization techniques to handle the time-related constraints of distributed energy resources; The optimization problem is decomposed into independent subproblems for each distributed energy resource using the primal dual algorithm; Online optimization is achieved through alternating iterations of primary and dual variables, generating an operational strategy.
[0010] Furthermore, the control of distributed energy resources includes at least one of the following operating modes: The renewable energy consumption model absorbs excess electricity during peak renewable energy output periods by coordinating energy storage systems with flexible loads. Off-grid islanded operation mode: When the grid is interrupted, it switches to off-grid operation mode to prioritize power supply to important loads and manage energy storage resources; The grid interaction mode provides peak-shaving support services based on grid demand.
[0011] Furthermore, the terminal device layer includes multiple atomic nodes, each atomic node corresponding to a distributed energy resource subsystem or functional module; The atomic node includes an autonomous control processor and application software, and accesses the data domain using a standardized four-layer data element interface structure based on MQTT / MQTT-SN topics. The four-layer data element interface structure includes an application layer, a behavior abstraction layer, a receiving layer, and a classification layer.
[0012] Furthermore, the data domain supports online expansion, online fault tolerance, and online maintenance mechanisms; Online extension is achieved through a decoupled communication mechanism, allowing new modules or subsystems to be integrated by configuring data field interfaces and setting content encoding; Online fault tolerance is achieved through redundancy and integration logic, and system integrity is maintained by filtering data. Online maintenance supports concurrent maintenance and testing without interrupting normal operation.
[0013] Furthermore, the edge layer is connected to the terminal device layer via a Wi-Fi mesh network with a maximum bandwidth of 150 Mbps; the edge layer is connected to the cloud layer via Ethernet with a maximum bandwidth of 100 Mbps; the cloud layer server has a storage capacity of 8 TB; the smart energy gateway of the edge layer has a storage capacity of 8 GB; and the terminal device sampling rate of the terminal device layer is 400 milliseconds.
[0014] Secondly, the present invention provides a large-scale distributed photovoltaic integration and coordination system, comprising: The data acquisition module, deployed at the terminal device layer, is used to collect operational data from multiple distributed energy resources; A data transmission module, deployed at the edge layer, is used to transmit the running data to a data domain, wherein the data domain achieves information sharing based on a publish-subscribe model; The event detection module, deployed in the cloud, is used to detect whether the system state deviates from the optimal domain of multiple indicators based on the running data and through an event-driven control framework, and to generate an event signal when the deviation occurs. The strategy generation module, deployed in the cloud, is used to generate operating strategies for distributed energy resources in response to the event signal, using a self-optimization method, wherein the self-optimization method employs an online primal-dual algorithm. The control execution module is used to distribute the operation strategy to the terminal device layer through the data domain and edge layer to control the operation of the distributed energy resources.
[0015] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: This invention provides a method and system for large-scale distributed photovoltaic integration and coordination. It adopts a self-optimization method and coordinates the operation of distributed energy resources through event-driven control and online primal dual algorithms, driving the system to continuously evolve towards the optimal domain of multiple indicators. It achieves near-optimal operation in complex dynamic environments while maintaining the simplicity and feasibility of the operation logic. Attached Figure Description
[0016] Figure 1 This is a conceptual and framework diagram of SVPP provided in the embodiments of the present invention; Figure 2 This is a system architecture diagram of SVPP provided in an embodiment of the present invention; Figure 3 This is a diagram of the SVPP system architecture provided in an embodiment of the present invention; Figure 4 This is an aerial photograph of the demonstration park provided in an embodiment of the present invention; Figure 5 This is a diagram of the power distribution network structure of a demonstration park provided in an embodiment of the present invention; Figure 6 This is a graph showing the average daily photovoltaic power generation and electricity consumption in the park, provided in an embodiment of the present invention. Figure 7 This is a flowchart of the self-optimization operation of SVPP provided in the embodiments of the present invention; Figure 8 These are the strategies and effects of the renewable energy model provided in the embodiments of the present invention; Figure 9 This is a diagram illustrating the strategy and effects of the off-network island mode provided in an embodiment of the present invention; Figure 10 This is a diagram illustrating the operation strategy of the power grid interaction mode provided in this embodiment of the invention.
[0017] Figure 11 This is a field deployment diagram of the distributed resource terminal provided in an embodiment of the present invention; Detailed Implementation
[0018] The present invention will be further described below with reference to the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and should not be used to limit the scope of protection of the present invention.
[0019] Example 1: This example introduces a method for large-scale distributed photovoltaic integration and coordination, including: Operational data from multiple distributed energy resources are collected at the terminal device layer; Operational data is transmitted to the data domain via the edge layer, where information sharing is achieved based on a publish-subscribe model. In the data domain, based on the operational data, an event-driven control framework is used to detect whether the system state deviates from the optimal domain of multiple indicators, and an event signal is generated when a deviation occurs. In response to the event signal, an operation strategy for distributed energy resources is generated in the cloud using a self-optimization method, wherein the self-optimization method employs an online primal-dual algorithm. The operation strategy is distributed to the terminal device layer through the data domain and edge layer to control the operation of distributed energy resources.
[0020] The large-scale distributed photovoltaic integration and coordination method provided in this embodiment involves the following steps in its application process: S1. Establish a conceptual framework for a self-optimizing virtual power plant (SVPP), such as... Figure 1 As shown, the framework includes three core levels: "optimization", "striving for optimization", and "self". S2. Construct an SVPP system architecture based on Autonomous Distributed Systems (ADS) to achieve efficient integration and dynamic aggregation of distributed energy resources. Figure 2 and Figure 3 These are the system architecture and system architecture construction diagram for SVPP; S3. A self-optimizing method is adopted, and the operation of distributed energy resources is coordinated through event-driven control and online primal dual algorithm; S4. Based on the system architecture and operation method described above, deploy SVPP in practical application scenarios to solve the problems of renewable energy consumption, reliable off-grid operation, and grid interaction capability.
[0021] Specifically: S11. The "optimal" level uses multi-level indicators to define the ideal operating range, including virtual power plant system-level indicators and distributed energy resource-level indicators. Virtual power plant system-level indicators are determined by the overall operational needs of the virtual power plant, including overall economic benefits, aggregated renewable energy consumption, operational reliability, and the service capabilities provided by the virtual power plant to the power system. Distributed energy resource-level indicators are determined by the operational needs and characteristics of individual distributed energy resources, ensuring that distributed energy resources operate within feasible power ranges, meet user electricity demands, and consider user-side power losses.
[0022] S12. The "optimization" layer uses an event-driven control framework. When the system state deviates from the optimization domain, it is defined as an event. Based on real-time data, strategies are quickly generated to bring the system closer to the ideal operating state. The system measures the "gap" between the current state and the optimization domain. As long as the gap is within a set range, the operation is considered effective. As the operation progresses, SVPP continuously optimizes the strategy, systematically eliminating events and gradually evolving towards the ideal operating state.
[0023] S13. At the "self" level, a collaborative mechanism is built to support the autonomous operation of different distributed energy resources, enabling each unit to respond to system-level demands based on its own capabilities. The core of this feature lies in the innovative system architecture of SVPP, which, through the construction of a collaborative mechanism, transmits system-level demands to individual distributed energy resources, allowing each unit to respond according to its own capabilities.
[0024] S21. The SVPP system architecture based on Autonomous Distributed Systems (ADS) comprises two core components: atomic nodes and a data domain. The atomic nodes are the basic units of the system; each distributed energy resource subsystem and functional module is designed as an independent atomic node with equivalent capabilities, including an Autonomous Control Processor (ACP) and application software (APL). The data domain is the information interaction space, where content encoding is used to determine data relevance, enabling efficient and real-time information sharing between nodes.
[0025] S22. Each distributed energy resource subsystem and functional module is transformed into an independent atomic node, equipped with autonomous management functions, an autonomous control processor (ACP) with self-management and decision-making capabilities, and application software (APL) that performs specific tasks related to the atomic node's function in SVPP.
[0026] Atomic nodes employ a standardized four-layer data element interface structure based on Message Queuing Telemetry Transport (MQTT) / Sensor Network Message Queuing Telemetry Transport (MQTT-SN) topics for accessing the data domain. This structure includes an application layer, a behavior abstraction layer, a receiving layer, and a classification layer. Atomic node behaviors are categorized into three types: "Measurement," "Control," and "Escalation," each containing corresponding request and response components. This design ensures the scalability and efficiency of the entire system interface, supporting seamless integration of different nodes.
[0027] S23. The data domain is built using the MQTT / MQTT-SN protocol and employs a publish-subscribe broadcast mechanism for data distribution. The data domain incorporates online expansion, online fault tolerance, and online maintenance mechanisms, supporting the integration of new modules or subsystems by configuring data field interfaces and setting content encodings, achieving plug-and-play functionality. Online expansion is achieved through a decoupled communication mechanism, allowing new modules or subsystems to integrate by configuring data field interfaces and setting appropriate content encodings; online fault tolerance is achieved through redundancy and integration logic, maintaining system integrity by filtering data; online maintenance supports concurrent maintenance and testing without interrupting normal operation.
[0028] S24. SVPP constructs a "cloud-edge-device" collaborative platform, comprising a cloud layer, an edge layer, and a terminal device layer. The cloud layer employs advanced analytics and self-optimizing methods, responsible for data processing, storage, and distributed energy resource operation strategy formulation. The edge layer, implemented through a smart energy gateway, acts as an edge device based on the MQTT protocol, responsible for local data aggregation and protocol conversion. The terminal device layer covers all distributed energy resources connected to SVPP, supporting sub-second data acquisition and feedback. To achieve interconnection between layers, SVPP constructs a robust communication network: the terminal device layer and the edge layer are connected via a Wi-Fi mesh network with a maximum bandwidth of 150 Mbps. This network possesses self-organizing, self-healing characteristics and mobile broadband capabilities, effectively addressing interference and wiring limitations in complex power distribution network environments, ensuring high stability, high transmission rates, and high anti-interference capabilities. The edge layer and the cloud layer are connected via Ethernet with a maximum bandwidth of 100 Mbps, ensuring reliable and high-speed communication between distributed edge devices and cloud infrastructure.
[0029] In terms of data processing, the sampling rate of terminal devices at the terminal device layer is typically 400 milliseconds, supporting near real-time monitoring and control. Each smart energy gateway at the edge layer has a storage capacity of 8GB, with local data caching and processing capabilities. The cloud layer server has a storage capacity of 8 TB, supporting comprehensive data analysis and long-term storage of system information.
[0030] To verify the effectiveness of the invention, a demonstration project was deployed in a business park. Figure 4 The aerial view of the demonstration park, which includes six buildings, was shown. Figure 5 The demonstration park showcases its power distribution network structure, which is supplied by the upper-level power grid through 10 kV distribution lines. After being stepped down by transformers, the power is distributed to the lines within the park. The low-voltage distribution network is divided into 0.4 kV sections I, II, and III busbars, supporting flexible operation modes such as grid connection and off-grid operation.
[0031] Figure 6 The bar chart shows the average daily photovoltaic power generation and electricity consumption in the park from February to April 2024.
[0032] S31. The self-optimizing operation method adopts an event-driven paradigm and an online primal-dual algorithm based on a virtual queue to achieve "time decoupling" and "space decoupling". Through a virtual queue and Lyapunov optimization techniques, it handles the time-dependent constraints of distributed energy resources with time-varying dependencies; through the primal-dual algorithm, the optimization problem is decomposed into independent sub-problems for each distributed energy resource; online optimization is achieved through alternating iterations of primal and dual variables.
[0033] S32. SVPP addresses practical challenges through three operating modes: renewable energy integration mode, off-grid islanding mode, and grid interaction mode.
[0034] Renewable energy consumption models improve the renewable energy consumption rate by coordinating energy storage systems with flexible loads to absorb excess electricity during peak renewable energy output periods. SVPP adjusts the operating status of distributed energy resources based on real-time fluctuations, increasing the local consumption of renewable energy and reducing curtailment rates. Figure 7 The self-optimization process of SVPP is demonstrated.
[0035] Taking February 16, 2024 as an example: When the photovoltaic power generation exceeds the load demand at noon, SVPP absorbs the excess photovoltaic output by adjusting distributed energy resources, smooths the load curve of the park, and improves the renewable energy absorption rate. The specific effect is shown in Figure 8.
[0036] The off-grid islanding operation mode ensures that the SVPP can switch to off-grid operation during grid outages, prioritizing power supply to critical loads and managing energy storage resources. The grid interaction mode provides flexible services based on grid demand, such as peak-shaving support. This mode ensures that the SVPP can switch to off-grid operation during grid outages. The SVPP prioritizes power supply to critical loads, manages energy storage resources, and maintains power supply for critical services by adjusting the operation of distributed energy resources. After grid restoration, the SVPP coordinates a smooth transition to normal operation.
[0037] Taking April 10, 2024 as an example: When the power grid is interrupted, the SVPP switches to off-grid mode and starts the preset strategy; when the energy storage capacity is close to the preset threshold, the SVPP implements power rationing for low-priority loads to ensure power supply for important loads; when the power grid is restored, the SVPP gradually restores the power supply to the loads and charges the energy storage system, realizing an efficient transition of the park to normal grid-connected operation. The specific effect is shown in Figure 9.
[0038] In the grid interaction mode, the SVPP provides flexibility services (such as peak shaving) based on grid demand. When the grid issues a support request, the SVPP meets the grid demand by coordinating distributed energy resources.
[0039] Taking April 15, 2024 as an example: When responding to the grid peak shaving request, SVPP adjusts distributed energy resources to reduce the load during the specified peak period, as shown in Figure 10.
[0040] S41. To verify the effectiveness of the self-optimization concept and evaluate the feasibility of SVPP in real-world scenarios, this invention designs a demonstration project. The demonstration project is located in a commercial park in southern China, which integrates various distributed energy resources, including rooftop photovoltaic systems, electric vehicle charging stations, energy storage systems, and controllable loads such as variable frequency air conditioning and lighting systems. Figure 11 The on-site deployment of distributed resource terminals within the demonstration park was showcased.
[0041] S42. The system integrates a total of 4,542 distributed energy resource terminals, covering various types of devices. To achieve self-organizing aggregation of these distributed energy resource terminals, the project deployed 165 edge computing smart energy gateways, interconnected via a Wi-Fi mesh network, serving as the communication and control hub for distributed energy resources. Table 1 lists the types and quantities of distributed energy resource terminals connected to SVPP.
[0042] Table 1. Types and Quantities of Distributed Energy Resources (DERs) Connected to SVPP type quantity Distributed photovoltaic 4 Energy storage system 2 Electric vehicle charging stations 3 Inverter air conditioner 393 Lighting load 656 Environmental sensors 19 Other load terminals 3465 total 4542 S43. Based on the established SVPP architecture and operating model, the self-optimization method was applied to address key challenges in the park's energy system. Through renewable energy consumption mode, off-grid islanding operation mode, and grid interaction mode, SVPP achieved photovoltaic power consumption of 28,427.7 kWh, reduced carbon dioxide emissions by 23,651.8 kg, and generated economic benefits of 18,739.5 yuan for the park. Table 2 summarizes the operating performance indicators from February to April 2024.
[0043] Table 2 SVPP Performance Indicators (February to April 2024) type numerical values Photovoltaic power consumption (kWh) 28427.7 Reduce carbon dioxide emissions (kg) 23651.8 Economic benefits generated (yuan) 18739.5 In summary, this embodiment of SVPP proposes a method for integrating and managing large-scale distributed energy resources, combining an Autonomous Distributed System (ADS) architecture with a self-optimizing operation method. In a practical SVPP project in a commercial park in southern China, the feasibility of addressing issues related to renewable energy consumption, power supply reliability, and grid flexibility was verified by coordinating the distributed energy resources of six buildings and the park's energy storage system. Future research will further refine the simulation platform, test SVPP strategies in multi-park collaborative scenarios, and integrate more grid service functions to further verify its scalability and adaptability in complex grid environments.
[0044] Example 2: This example provides a large-scale distributed photovoltaic integration and coordination system, including: The data acquisition module, deployed at the terminal device layer, is used to collect operational data from multiple distributed energy resources; A data transmission module, deployed at the edge layer, is used to transmit the running data to a data domain, wherein the data domain achieves information sharing based on a publish-subscribe model; The event detection module, deployed in the cloud, is used to detect whether the system state deviates from the optimal domain of multiple indicators based on the running data and through an event-driven control framework, and to generate an event signal when the deviation occurs. The strategy generation module, deployed in the cloud, is used to generate operating strategies for distributed energy resources in response to the event signal, using a self-optimization method, wherein the self-optimization method employs an online primal-dual algorithm. The control execution module is used to distribute the operation strategy to the terminal device layer through the data domain and edge layer to control the operation of the distributed energy resources.
[0045] The specific functions of each module described above are explained in the relevant content of the method in Embodiment 1, and will not be repeated here.
[0046] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.
[0047] Those skilled in the art will understand that embodiments of this disclosure can be provided as methods, systems, or computer program products. Therefore, this disclosure can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this disclosure can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0048] This disclosure is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this disclosure. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create a machine for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.
[0049] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.
[0050] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.
[0051] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this disclosure and not to limit its protection scope. Although this disclosure has been described in detail with reference to the above embodiments, those skilled in the art should understand that after reading this disclosure, they can still make various changes, modifications or equivalent substitutions to the specific implementation of the invention, but these changes, modifications or equivalent substitutions are all within the protection scope of the pending claims.
Claims
1. A method for large-scale distributed photovoltaic integration and coordination, characterized in that, include: Operational data from multiple distributed energy resources are collected at the terminal device layer; Operational data is transmitted to the data domain via the edge layer, where information sharing is achieved based on a publish-subscribe model. In the data domain, based on the operational data, an event-driven control framework is used to detect whether the system state deviates from the optimal domain of multiple indicators, and an event signal is generated when a deviation occurs. In response to the event signal, an operation strategy for distributed energy resources is generated in the cloud using a self-optimization method, wherein the self-optimization method employs an online primal-dual algorithm. The operation strategy is distributed to the terminal device layer through the data domain and edge layer to control the operation of distributed energy resources.
2. The method for large-scale distributed photovoltaic integration and coordination according to claim 1, characterized in that, The process of collecting operational data from multiple distributed energy resources through the terminal device layer includes: Power, status, and environmental data of distributed energy resources are collected at a sub-second sampling rate through the terminal device layer. The terminal equipment layer includes distributed photovoltaics, energy storage systems, electric vehicle charging piles, variable frequency air conditioners, lighting loads, and environmental sensors.
3. The method for large-scale distributed photovoltaic integration and coordination according to claim 1, characterized in that, The transmission of operational data to the data domain via the edge layer includes: The smart energy gateway performs local aggregation and protocol conversion of operational data, wherein the smart energy gateway is based on the MQTT protocol. The transformed runtime data is published to the data domain, which broadcasts data using a publish-subscribe model based on the MQTT / MQTT-SN protocol.
4. The method for large-scale distributed photovoltaic integration and coordination according to claim 1, characterized in that, The method of detecting whether the system state deviates from the optimal domain of multiple indicators through an event-driven control framework includes: Define a multi-indicator optimization domain, including virtual power plant system-level indicators and distributed energy resource-level indicators; Monitor the system status in real time and calculate the gap between the current state and the optimal domain; When the gap exceeds a set threshold, an event signal is generated, wherein the event signal is used to trigger a self-optimization process.
5. The method for large-scale distributed photovoltaic integration and coordination according to claim 1, characterized in that, The operational strategy for generating distributed energy resources using a self-optimizing method in the cloud includes: An online primal-dual algorithm based on virtual queues is adopted, which uses virtual queues and Lyapunov optimization techniques to handle the time-related constraints of distributed energy resources; The optimization problem is decomposed into independent subproblems for each distributed energy resource using the primal dual algorithm; Online optimization is achieved through alternating iterations of primary and dual variables, generating an operational strategy.
6. The method for large-scale distributed photovoltaic integration and coordination according to claim 1, characterized in that, The operation of the controlled distributed energy resources includes at least one of the following operating modes: The renewable energy consumption model absorbs excess electricity during peak renewable energy output periods by coordinating energy storage systems with flexible loads. Off-grid islanded operation mode: When the grid is interrupted, it switches to off-grid operation mode to prioritize power supply to important loads and manage energy storage resources; The grid interaction mode provides peak-shaving support services based on grid demand.
7. The method for large-scale distributed photovoltaic integration and coordination according to claim 1, characterized in that, The terminal device layer includes multiple atomic nodes, each atomic node corresponding to a distributed energy resource subsystem or functional module; The atomic node includes an autonomous control processor and application software, and accesses the data domain using a standardized four-layer data element interface structure based on MQTT / MQTT-SN topics. The four-layer data element interface structure includes an application layer, a behavior abstraction layer, a receiving layer, and a classification layer.
8. The method for large-scale distributed photovoltaic integration and coordination according to claim 1, characterized in that, The data domain supports online expansion, online fault tolerance, and online maintenance mechanisms; Online extension is achieved through a decoupled communication mechanism, allowing new modules or subsystems to be integrated by configuring data field interfaces and setting content encoding; Online fault tolerance is achieved through redundancy and integration logic, and system integrity is maintained by filtering data. Online maintenance supports concurrent maintenance and testing without interrupting normal operation.
9. The method for large-scale distributed photovoltaic integration and coordination according to claim 1, characterized in that, The edge layer is connected to the terminal device layer via a Wi-Fi mesh network with a maximum bandwidth of 150 Mbps; the edge layer is connected to the cloud layer via Ethernet with a maximum bandwidth of 100 Mbps; the cloud layer server has a storage capacity of 8 TB; the smart energy gateway of the edge layer has a storage capacity of 8 GB; and the terminal device sampling rate of the terminal device layer is 400 milliseconds.
10. A large-scale distributed photovoltaic integration and coordination system, characterized in that, include: The data acquisition module, deployed at the terminal device layer, is used to collect operational data from multiple distributed energy resources; A data transmission module, deployed at the edge layer, is used to transmit the running data to a data domain, wherein the data domain achieves information sharing based on a publish-subscribe model; The event detection module, deployed in the cloud, is used to detect whether the system state deviates from the optimal domain of multiple indicators based on the running data and through an event-driven control framework, and to generate an event signal when the deviation occurs. The strategy generation module, deployed in the cloud, is used to generate operating strategies for distributed energy resources in response to the event signal, using a self-optimization method, wherein the self-optimization method employs an online primal-dual algorithm. The control execution module is used to distribute the operation strategy to the terminal device layer through the data domain and edge layer to control the operation of the distributed energy resources.