A regional energy distributed control method based on edge computing

Abstract

The present invention discloses a regional energy distributed control method based on edge computing, which relates to the technical field of distributed control, including collecting edge node regional energy data and establishing an edge node energy distribution model; performing discrete multi-objective cascade optimization scheduling on the edge node energy distribution model and decoupling hierarchical optimization; generating a two-order optimization scheduling based on columns and constraints, and the two-order optimization scheduling includes iterative operation of a first-order scheduling algorithm and a second-order scheduling algorithm. The global optimization goal is guaranteed by coordinating with other sub-models through power interaction; it effectively responds to the challenges of a large number of devices and users and a large amount of calculation in the regional energy Internet, which not only simplifies the scheduling calculation, but also maintains the local autonomy of each sub-model while ensuring global optimization. The two-order optimization scheduling strategy is adopted to achieve a balance between global optimization and local optimization, which can cope with the uncertainty factors in the system and further improve the stability and optimization effect.

Description

Technical Field

[0001] The present invention relates to the technical field of distributed control, and in particular to a regional energy distributed control method based on edge computing. Background Art

[0002] With the continuous growth of global energy demand and the increasing prevalence of renewable energy sources (such as solar and wind power), traditional centralized energy management models are facing increasing challenges. In particular, with the increasing number of distributed energy systems (such as household photovoltaic power generation, wind farms, and energy storage devices), efficient and intelligent management and scheduling of these distributed energy resources has become a core issue facing modern energy management systems. Against this backdrop, edge computing, as a new computing architecture, has begun to gain widespread application in the energy sector. Edge computing not only shortens data transmission paths and reduces latency, but also enables local real-time data processing and decision-making, thereby improving system response speed and computational efficiency. Integrating edge computing, regional energy distributed control methods are becoming an effective solution, enabling intelligent scheduling and management of distributed energy resources locally, reducing reliance on central servers and enhancing system flexibility and scalability.

[0003] Currently, a Chinese invention patent application numbered CN202411711099.6 discloses a distributed energy regional autonomous control system and method with multi-level collaborative control. These systems improve the system's real-time performance and response speed; enhance the grid's ability to absorb clean energy, maximizing its utilization; reduce the data processing burden and protect user privacy through edge computing; and possess rapid emergency response capabilities, enabling rapid adjustment of load distribution in the event of a grid failure or emergency, ensuring the safety and stability of grid operation. However, existing technologies, due to the large number of devices and users, require a massive amount of scheduling calculations, and it is difficult to efficiently solve the problem globally while also taking into account local autonomy. This makes it impossible to effectively address uncertainties, resulting in limited optimization results. Summary of the Invention

[0004] The technical problem solved by the present invention is that in the existing technology, due to the large number of devices and users, the scheduling calculation is huge, and it is difficult to efficiently solve the problem globally while taking into account local autonomy. It is impossible to effectively deal with uncertainty factors and the optimization effect is not obvious.

[0005] To solve the above technical problems, the present invention provides the following technical solutions: a regional energy distributed control method based on edge computing, comprising the following steps:

[0006] Step S1: Collect regional energy data of edge nodes and establish an energy distribution model of edge nodes;

[0007] Step S2: Perform discrete multi-objective cascade optimization scheduling on the edge node energy distribution model and decouple hierarchical optimization;

[0008] Step S3: Generate a two-stage optimization schedule based on the columns and constraints, wherein the two-stage optimization schedule includes iterative execution of a first-stage scheduling algorithm and a second-stage scheduling algorithm.

[0009] Preferably, collecting edge node regional energy data includes:

[0010] The energy data of distributed energy devices is collected in real time through edge computing nodes deployed in the region. The energy data includes operating status, energy consumption data and environmental parameters, forming a multi-dimensional energy data stream.

[0011] The edge computing nodes include intelligent gateways and sensor networks, the distributed energy equipment includes photovoltaics, energy storage and gas turbines, the operating states include startup, shutdown and stagnation, the energy consumption data include temperature, humidity, pressure, current, voltage, power, energy output and energy consumption, and the environmental parameters include temperature, humidity, air pressure and light intensity;

[0012] The multi-dimensional energy data stream is converged to the edge computing node through the intelligent gateway, and is marked, synchronized and preprocessed in time series in the edge computing node. The preprocessing includes denoising and normalization.

[0013] Preferably, establishing an edge node energy distribution model based on the energy data includes:

[0014] The geographical location partitions and functional levels of edge nodes are integrated through the Geographic Information System (GIS), and the multi-energy flow coupling characteristics are hierarchically described in combination with the geographical location partitions and functional levels of edge nodes. The geographical location partitions include city centers, suburbs, and edge areas, the functional levels include core nodes and peripheral nodes, and the multi-energy flow coupling characteristics include first temperature, second temperature, electricity, and gas.

[0015] The hierarchical description includes:

[0016] Divide the current edge node energy distribution model into several levels;

[0017] Utilize virtual power plant technology to aggregate distributed energy resources into a virtual entity, forming a dispatchable virtual power plant;

[0018] Based on the integrated functions of the virtual power plant, the energy distribution model of the edge node is intelligently scheduled through an optimization algorithm. The goals of intelligent scheduling include maximizing energy utilization efficiency, minimizing scheduling costs, meeting the needs of each node, and optimizing energy flow. The optimized edge node energy distribution model is deployed in the actual regional energy Internet, which includes all edge nodes, core nodes, energy stations, and energy storage equipment.

[0019] Preferably, step S2 includes:

[0020] Setting optimization goals of the edge node energy distribution model includes reducing energy loss of the edge node energy distribution model, improving energy utilization of the edge node energy distribution model, and reducing the cost of the edge node energy distribution model;

[0021] Discretization of scheduling decisions includes discretizing all energy data for each time period by category, and for each category of energy data, dividing the continuous data into several discrete intervals according to predetermined intervals. The categories include operating status, energy consumption data, and environmental parameters.

[0022] Preferably, performing decoupling hierarchical optimization on the edge node energy distribution model includes:

[0023] The regional edge node energy distribution model is decoupled into several virtual energy distribution sub-models, and each virtual energy distribution sub-model is managed in a unified manner. The unified management includes managing the power interaction between virtual energy distribution sub-models through a dispatchable virtual power plant. The internal scheduling planning of each virtual energy distribution sub-model is autonomously optimized by the local controller, and the target cascade method is used to coordinate the edge node energy distribution model and the virtual energy distribution sub-model through penalty functions and consistency parameters.

[0024] Preferably, the coordination process specifically includes:

[0025] The edge node energy distribution model is used as the top-level management node to coordinate the global energy supply and demand balance and interactively optimize the multi-dimensional energy data flow;

[0026] The virtual energy distribution sub-model is used as the lower-level execution node, and the scheduling strategy is adjusted according to the local device constraints, and the interactive power is fed back to the upper-level management node. The local device constraints include preset energy storage device charging and discharging restrictions and the dynamic characteristics of the regional energy Internet. The dynamic characteristics include dynamic feature quantities obtained by neural network feature extraction of preprocessed multi-dimensional energy data streams.

[0027] Preferably, step S3 includes:

[0028] The first-order scheduling algorithm includes initializing parameters of the optimized edge node energy distribution model, the parameters including a first penalty function factor, a second penalty function factor, a first convergence criterion, a second convergence criterion, an iteration counter k=1, a first penalty function iteration multiplier, and a second penalty function iteration multiplier;

[0029] The first penalty function factor is used to control the penalty function factor of the electric energy interaction information error;

[0030] The second penalty function factor is used to control the penalty function factor of the scheduling error of the virtual energy distribution sub-model;

[0031] The first convergence criterion is the convergence error threshold of the power interaction between the virtual energy distribution sub-models;

[0032] The second convergence criterion is the convergence error threshold of the power interaction between the virtual energy distribution sub-model and the edge node energy distribution model;

[0033] The first penalty function iteration multiplier is used to update the multiplication factor of the first penalty function factor;

[0034] The second penalty function iteration multiplier is used to update the multiplication factor of the second penalty function factor.

[0035] The first interaction power variable and the second interaction power variable are updated iteratively until a convergence condition is satisfied, wherein the convergence condition is that the iteration reaches a preset iteration threshold.

[0036] Preferably, the first-order scheduling algorithm includes:

[0037] Step S301: respectively obtaining first power interaction information and second power interaction information between each virtual energy distribution sub-model;

[0038] The first electric energy interaction information is the electric energy interaction unit quantity from the edge node energy distribution model to the virtual energy distribution sub-model;

[0039] The second electric energy interaction information is the electric energy interaction unit quantity from the virtual energy distribution sub-model to the edge node energy distribution model;

[0040] Step S302: Solve the optimization problem of all virtual energy distribution sub-models based on the optimization algorithm to obtain the first power scheduling and the second power scheduling between the virtual energy distribution sub-models;

[0041] The first power scheduling is the power output by the virtual energy distribution sub-model to the edge node energy distribution model;

[0042] The second electric energy dispatch is the electric energy output by the virtual energy distribution sub-model to other virtual energy distribution sub-models;

[0043] Step S303: Update the penalty function factor according to the optimization result. The updating process includes:

[0044] First penalty function factor + 1 = first penalty function iteration multiplier × first penalty function factor;

[0045] Second penalty function factor + 1 = second penalty function iteration multiplier × second penalty function factor;

[0046] Step S304: Update the iteration counter k=k+1, and return to step S302 until the convergence criterion is met;

[0047] The convergence criterion is when the power exchange error between the virtual energy distribution sub-models meets the convergence conditions, and the convergence conditions include:

[0048] |First power interaction information-first power scheduling|<first convergence criterion;

[0049] |Second power interaction information-second power scheduling|<second convergence criterion;

[0050] Step S305: When the convergence criteria are met at the same time, the iterative process ends and the optimization results are output. The optimization results include the power scheduling plan and power exchange strategy of each virtual energy distribution sub-model.

[0051] Preferably, the second-order scheduling algorithm includes:

[0052] The column constraint generation algorithm CCG is adopted to correct the decision deviation in the first-order scheduling algorithm through interactive iteration with the first-order scheduling algorithm.

[0053] Preferably, the second-order scheduling algorithm specifically includes:

[0054] Step S311: Initialize the lower bound of the subproblem to negative infinity, the upper bound of the subproblem to positive infinity, set the iteration counter of the subproblem r=1, and set the third convergence criterion;

[0055] Step S312: The convergence condition is set as: subproblem upper bound - subproblem lower bound ≥ the third convergence criterion;

[0056] Step S313: iteratively optimize the optimization results of the first-order scheduling algorithm, update the dual variables and objective constraints in the calculation process of the first-order scheduling algorithm, and update the upper bound and lower bound of the subproblem;

[0057] Step S314: Update r=r+1;

[0058] Step S315: Optimize the currently updated subproblem as the optimization object, recursively optimize the upper bound and lower bound of the subproblem of the subproblem, and obtain the upper bound and lower bound that meet the third convergence criterion;

[0059] Step S316: Stop iteration when convergence condition is met;

[0060] All data volumes of the regional energy distributed control method based on edge computing are within unit time.

[0061] Beneficial effects of the present invention: This method decouples the edge node energy distribution model into multiple small integrated energy sub-models for analysis and optimization, and coordinates with other sub-models through power interaction to ensure the global optimization goal; through the regional energy distributed control method based on edge computing, it can effectively cope with the challenges of numerous devices and users and large computational load in the regional energy Internet. By decoupling the edge node energy distribution model from the virtual energy distribution sub-model and utilizing discrete multi-objective cascade optimization scheduling, it not only simplifies the scheduling calculation, but also maintains the local autonomy of each sub-model while ensuring global optimization. A two-order optimization scheduling strategy is adopted to achieve a balance between global optimization and local optimization through the collaborative work of the first-order scheduling algorithm and the second-order scheduling algorithm, which can cope with the uncertainty factors in the system and further improve the stability and optimization effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 A schematic diagram of the basic flow of a regional energy distributed control method based on edge computing provided for one embodiment of the present invention. DETAILED DESCRIPTION

[0063] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings. It is obvious that the described embodiments are only part of the embodiments of the present invention, but not all of the embodiments.

[0064] Reference Figure 1 , as an embodiment of the present invention, provides a regional energy distributed control method based on edge computing, comprising the following steps:

[0065] Step S1: Collect regional energy data of edge nodes and establish an energy distribution model of edge nodes;

[0066] Step S2: Perform discrete multi-objective cascade optimization scheduling on the edge node energy distribution model and decouple hierarchical optimization;

[0067] Step S3: Generate a two-stage optimization schedule based on the columns and constraints. The two-stage optimization schedule includes iterative operations of the first-stage scheduling algorithm and the second-stage scheduling algorithm.

[0068] Collecting edge node regional energy data includes:

[0069] The energy data of distributed energy devices is collected in real time through edge computing nodes deployed in the region. The energy data includes operating status, energy consumption data, and environmental parameters, forming a multi-dimensional energy data stream.

[0070] Edge computing nodes include smart gateways and sensor networks. Distributed energy devices include photovoltaics, energy storage, and gas turbines. Operational states include startup, shutdown, and stagnation. Energy consumption data includes temperature, humidity, pressure, current, voltage, power, energy output, and energy consumption. Environmental parameters include temperature, humidity, air pressure, and light intensity.

[0071] Multi-dimensional energy data streams are aggregated to edge computing nodes through intelligent gateways, where they are marked, synchronized, and preprocessed in time series. Preprocessing includes denoising and normalization.

[0072] Establishing an edge node energy distribution model based on energy data includes:

[0073] The geographic information system (GIS) is used to integrate the geographical location partitioning and functional hierarchy of edge nodes. The multi-energy flow coupling characteristics are hierarchically described by combining the geographical location partitioning and functional hierarchy of edge nodes. The geographical location partitioning includes urban centers, suburbs, and edge areas. The functional hierarchy includes core nodes and peripheral nodes. Core nodes are used for large-scale coordination and control the overall operation of the main edge computing node network. Peripheral nodes are used for local scheduling. The multi-energy flow coupling characteristics include the first temperature, the second temperature, electricity, and gas.

[0074] The hierarchical description includes:

[0075] The current edge node energy distribution model is divided into several levels. For example, the first temperature (heating) and the second temperature (cooling) are important parameters of the thermal system. Especially in the central area of ​​the city, the management and scheduling of temperature flows require centralized coordination at the core node to schedule the coupling between heating and cooling energy devices to improve the energy efficiency of the system.

[0076] Core nodes need to coordinate the coupling between electricity and gas. For example, when natural gas supply is insufficient, energy storage devices and power systems can provide supplementary energy. End nodes balance electricity and gas demand by controlling the startup or shutdown of local equipment.

[0077] Using virtual power plant technology to aggregate distributed energy resources into a virtual entity to form a dispatchable virtual power plant, which simplifies the scheduling task by aggregating multiple distributed resources into a unified control unit;

[0078] Virtual power plants can balance demand and supply, reduce energy waste, and improve the stability of the overall system by integrating different types of multi-energy flow coupling characteristics.

[0079] Based on the integrated functions of the virtual power plant, the energy distribution model of the edge node is intelligently scheduled through an optimization algorithm. The goals of intelligent scheduling include maximizing energy utilization efficiency, minimizing scheduling costs, meeting the needs of each node, and optimizing energy flow. The optimized edge node energy distribution model is deployed in the actual regional energy Internet, which includes all edge nodes, core nodes, energy stations, and energy storage equipment.

[0080] The regional energy Internet dispatching system optimized based on collected data and virtual power plant technology can effectively improve the utilization efficiency of energy resources, reduce system operating costs, and ensure the stability and reliability of energy supply.

[0081] Step S2 includes:

[0082] This method uses virtualization technology to decouple the power coupling relationship between the global energy scheduling platform and each sub-model. The global energy scheduling platform does not need to obtain the equipment scheduling information within each sub-model, but instead coordinates each sub-model as a whole. Each virtual energy distribution sub-model only focuses on the power interaction with the global energy scheduling platform, without paying attention to the internal scheduling details of other sub-models.

[0083] Setting optimization goals of the edge node energy distribution model includes reducing energy loss of the edge node energy distribution model, improving energy utilization of the edge node energy distribution model, and reducing the cost of the edge node energy distribution model;

[0084] Due to the discrete nature of equipment and energy forms, the discretization processing of scheduling decisions includes: discretizing all energy data in each time period by category. For each category of energy data, the continuous data is divided into several discrete intervals according to predetermined intervals. The categories include operating status, energy consumption data and environmental parameters.

[0085] Decoupling hierarchical optimization of the edge node energy distribution model includes:

[0086] The regional edge node energy distribution model is decoupled into several virtual energy distribution sub-models, and each virtual energy distribution sub-model is managed in a unified manner. The unified management includes managing the power interaction between virtual energy distribution sub-models through a dispatchable virtual power plant. The internal scheduling planning of each virtual energy distribution sub-model is autonomously optimized by the local controller. The target cascade method is used to coordinate the edge node energy distribution model and the virtual energy distribution sub-model through the penalty function and consistency parameters to achieve a balance between global optimization and local autonomy.

[0087] The coordination process specifically includes:

[0088] The edge node energy distribution model is used as the top-level management node to coordinate the global energy supply and demand balance and interactively optimize the multi-dimensional energy data flow;

[0089] The virtual energy distribution sub-model is used as the lower-level execution node. The scheduling strategy is adjusted according to the local device constraints, and the interactive power is fed back to the upper-level management node. The local device constraints include the preset charging and discharging restrictions of the energy storage equipment and the dynamic characteristics of the regional energy Internet. The dynamic characteristics include the dynamic feature quantities obtained by neural network feature extraction of the pre-processed multi-dimensional energy data stream.

[0090] Step S3 includes:

[0091] The first-order scheduling algorithm includes initializing parameters of the optimized edge node energy distribution model, the parameters including a first penalty function factor, a second penalty function factor, a first convergence criterion, a second convergence criterion, an iteration counter k=1, a first penalty function iteration multiplier, and a second penalty function iteration multiplier;

[0092] The first penalty function factor is used to control the penalty function factor of the electric energy interaction information error;

[0093] The second penalty function factor is used to control the penalty function factor of the scheduling error of the virtual energy distribution sub-model;

[0094] The first convergence criterion is the convergence error threshold of the power interaction between the virtual energy distribution sub-models;

[0095] The second convergence criterion is the convergence error threshold of the power interaction between the virtual energy distribution sub-model and the edge node energy distribution model;

[0096] The first penalty function iteration multiplier is used to update the multiplication factor of the first penalty function factor;

[0097] The second penalty function iteration multiplier is used to update the multiplication factor of the second penalty function factor.

[0098] The first interaction power variable and the second interaction power variable are updated iteratively until a convergence condition is satisfied, where the convergence condition is that the iteration reaches a preset iteration threshold.

[0099] The first-order scheduling algorithm ensures the robustness of the system in uncertain scenarios.

[0100] The first-order scheduling algorithm includes:

[0101] Step S301: respectively obtaining first power interaction information and second power interaction information between each virtual energy distribution sub-model;

[0102] The first power interaction information is the power interaction unit quantity from the edge node energy distribution model to the virtual energy distribution sub-model;

[0103] The second electric energy interaction information is the electric energy interaction unit quantity from the virtual energy distribution sub-model to the edge node energy distribution model;

[0104] Step S302: Solve the optimization problem of all virtual energy distribution sub-models based on the optimization algorithm to obtain the first power scheduling and the second power scheduling between the virtual energy distribution sub-models;

[0105] The first power scheduling is the power output by the virtual energy distribution sub-model to the edge node energy distribution model;

[0106] The second electric energy dispatch is the electric energy output by the virtual energy distribution sub-model to other virtual energy distribution sub-models;

[0107] Step S303: Update the penalty function factor according to the optimization result. The updating process includes:

[0108] First penalty function factor + 1 = first penalty function iteration multiplier × first penalty function factor;

[0109] Second penalty function factor + 1 = second penalty function iteration multiplier × second penalty function factor;

[0110] The update process of these two factors ensures that the system controls the error more and more strictly in each iteration, helping the optimization process converge faster;

[0111] Step S304: Update the iteration counter k=k+1, and return to step S302 until the convergence criterion is met;

[0112] The convergence criterion is when the power exchange error between the virtual energy distribution sub-models meets the convergence conditions, and the convergence conditions include:

[0113] |First power interaction information-first power scheduling|<first convergence criterion;

[0114] |Second power interaction information-second power scheduling|<second convergence criterion;

[0115] Step S305: When the convergence criteria are met at the same time, the iterative process ends and the optimization results are output. The optimization results include the power scheduling plan and power exchange strategy of each virtual energy distribution sub-model.

[0116] By introducing robust optimization methods, the uncertainty sources in the system are handled and each sub-model is ensured to maintain efficient operation under the most unfavorable conditions.

[0117] By generating columns and constraints, the constraints in the scheduling problem are gradually optimized to ensure the coordination of the system between the global and local aspects.

[0118] The optimization scheduling process gradually optimizes the power distribution of the energy system through an iterative process, which ensures the effective distribution of power, optimizes energy utilization efficiency and meets the system constraints.

[0119] The second-order scheduling algorithm includes:

[0120] The column constraint generation algorithm CCG is adopted to correct the decision deviation in the first-order scheduling algorithm through interactive iteration with the first-order scheduling algorithm.

[0121] For the worst scenarios monitored in real time, such as extreme weather and sudden load changes, the energy allocation strategy is dynamically adjusted and the real-time scheduling plan under multi-energy flow coupling is optimized to ensure economy and safety.

[0122] The second-order scheduling algorithm specifically includes:

[0123] Step S311: Initialize the lower bound of the subproblem to negative infinity, the upper bound of the subproblem to positive infinity, set the iteration counter r of the subproblem to 1, and set the third convergence criterion to determine whether the algorithm converges;

[0124] Step S312: The convergence condition is set as: subproblem upper bound - subproblem lower bound ≥ the third convergence criterion;

[0125] Step S313: iteratively optimize the optimization results of the first-order scheduling algorithm, update the dual variables and objective constraints in the calculation process of the first-order scheduling algorithm, and update the upper bound and lower bound of the subproblem;

[0126] Step S314: Update r=r+1, and the new solution will be used as the input value for the next round;

[0127] Step S315: Optimize the currently updated subproblem as the optimization object, recursively optimize the upper bound and lower bound of the subproblem of the subproblem, and obtain the upper bound and lower bound that meet the third convergence criterion;

[0128] This process ensures the dynamic tracking and adjustment of all constraints, ensuring that the quality of the solution is improved as much as possible in each iteration. The update of the constraints ensures that the subproblem can be improved within the feasible solution space after each optimization.

[0129] Step S316: Stop iteration when convergence condition is met;

[0130] The algorithm alternates between solving a main problem and a subproblem, continuously updating the subproblem's upper and lower bounds and constraints. With each iteration, the algorithm updates the current solution and introduces new dual variables and constraints, ultimately finding the global optimal solution. The core idea of ​​the algorithm is to optimize each subproblem through iterative solving and updating, gradually converging to the optimal solution.

[0131] All data volumes of the regional energy distributed control method based on edge computing are within unit time.

[0132] This method decouples the edge node energy distribution model into multiple small integrated energy sub-models for analysis and optimization, and coordinates with other sub-models through power interaction to ensure global optimization. This regional energy distributed control method based on edge computing can effectively address the challenges of numerous devices and users and the high computational complexity in regional energy internets. By decoupling the edge node energy distribution model from the virtual energy distribution sub-model and utilizing discrete multi-objective cascade optimization scheduling, it not only simplifies scheduling calculations but also maintains the local autonomy of each sub-model while ensuring global optimization. A dual-order optimization scheduling strategy is adopted, achieving a balance between global optimization and local optimization through the collaborative work of first-order and second-order scheduling algorithms. This approach can address system uncertainty and further improve stability and optimization results.

[0133] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media containing computer-usable program code. The storage medium may be implemented by any type of volatile or non-volatile storage device, or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0134] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A regional energy distributed control method based on edge computing, characterized in that: The following steps are involved: Step S1: Collect regional energy data of edge nodes and establish an energy distribution model of edge nodes; Step S2: Perform discrete multi-objective cascade optimization scheduling on the edge node energy distribution model and decouple hierarchical optimization; Step S3: Generate a two-stage optimization schedule based on the columns and constraints, the two-stage optimization schedule includes iterative operation of the first-order scheduling algorithm and the second-order scheduling algorithm; The step S2 comprises: Setting optimization goals of the edge node energy distribution model includes reducing energy loss of the edge node energy distribution model, improving energy utilization of the edge node energy distribution model, and reducing the cost of the edge node energy distribution model; Discretization for scheduling decisions involves discretizing all energy data for each time period into categories. For each energy data category, the continuous data is divided into a number of discrete intervals at predetermined intervals. The categories include operating status, energy consumption data, and environmental parameters. Decoupling hierarchical optimization of the edge node energy distribution model includes: The regional edge node energy distribution model is decoupled into several virtual energy distribution sub-models, and each virtual energy distribution sub-model is centrally managed. This unified management includes managing the power interaction between virtual energy distribution sub-models through a dispatchable virtual power plant. The internal scheduling of each virtual energy distribution sub-model is autonomously optimized by the local controller. The target cascade method is used to coordinate the edge node energy distribution model and the virtual energy distribution sub-model through penalty functions and consistency parameters. The coordination process specifically includes: The edge node energy distribution model is used as the top-level management node to coordinate the global energy supply and demand balance and interactively optimize the multi-dimensional energy data flow; The virtual energy distribution sub-model is used as the lower-level execution node, and the scheduling strategy is adjusted according to the local device constraints, and the interactive power is fed back to the upper-level management node. The local device constraints include preset energy storage device charging and discharging restrictions and the dynamic characteristics of the regional energy Internet. The dynamic characteristics include dynamic feature quantities obtained by neural network feature extraction of preprocessed multi-dimensional energy data streams.

2. The regional energy distributed control method based on edge computing according to claim 1, characterized in that: Collecting edge node regional energy data includes: The energy data of distributed energy devices is collected in real time through edge computing nodes deployed in the region. The energy data includes operating status, energy consumption data and environmental parameters, forming a multi-dimensional energy data stream. The edge computing nodes include intelligent gateways and sensor networks, the distributed energy equipment includes photovoltaics, energy storage and gas turbines, the operating states include startup, shutdown and stagnation, the energy consumption data include temperature, humidity, pressure, current, voltage, power, energy output and energy consumption, and the environmental parameters include temperature, humidity, air pressure and light intensity; The multi-dimensional energy data stream is converged to the edge computing node through the intelligent gateway, and is marked, synchronized and preprocessed in time series in the edge computing node. The preprocessing includes denoising and normalization.

3. The regional energy distributed control method based on edge computing according to claim 2, characterized in that: Establishing an edge node energy distribution model based on the energy data includes: The geographical location partitions and functional levels of edge nodes are integrated through the Geographic Information System (GIS), and the multi-energy flow coupling characteristics are hierarchically described in combination with the geographical location partitions and functional levels of edge nodes. The geographical location partitions include city centers, suburbs, and edge areas, the functional levels include core nodes and peripheral nodes, and the multi-energy flow coupling characteristics include first temperature, second temperature, electricity, and gas. The hierarchical description includes: Divide the current edge node energy distribution model into several levels; Utilize virtual power plant technology to aggregate distributed energy resources into a virtual entity, forming a dispatchable virtual power plant; Based on the integrated functions of the virtual power plant, the energy distribution model of the edge node is intelligently scheduled through an optimization algorithm. The goals of intelligent scheduling include maximizing energy utilization efficiency, minimizing scheduling costs, meeting the needs of each node, and optimizing energy flow. The optimized edge node energy distribution model is deployed in the actual regional energy Internet, which includes all edge nodes, core nodes, energy stations, and energy storage equipment.

4. The regional energy distributed control method based on edge computing according to claim 3 is characterized in that: The step S3 comprises: The first-order scheduling algorithm includes initializing parameters of the optimized edge node energy distribution model, the parameters including a first penalty function factor, a second penalty function factor, a first convergence criterion, a second convergence criterion, an iteration counter k=1, a first penalty function iteration multiplier, and a second penalty function iteration multiplier; The first penalty function factor is used to control the penalty function factor of the electric energy interaction information error; The second penalty function factor is used to control the penalty function factor of the scheduling error of the virtual energy distribution sub-model; The first convergence criterion is the convergence error threshold of the power interaction between the virtual energy distribution sub-models; The second convergence criterion is the convergence error threshold of the power interaction between the virtual energy distribution sub-model and the edge node energy distribution model; The first penalty function iteration multiplier is used to update the multiplication factor of the first penalty function factor; The second penalty function iteration multiplier is used to update the multiplication factor of the second penalty function factor; The first interaction power variable and the second interaction power variable are updated iteratively until a convergence condition is satisfied, wherein the convergence condition is that the iteration reaches a preset iteration threshold.

5. The regional energy distributed control method based on edge computing according to claim 4, characterized in that: The first-order scheduling algorithm includes: Step S301: respectively obtaining first power interaction information and second power interaction information between each virtual energy distribution sub-model; The first electric energy interaction information is the electric energy interaction unit quantity from the edge node energy distribution model to the virtual energy distribution sub-model; The second electric energy interaction information is the electric energy interaction unit quantity from the virtual energy distribution sub-model to the edge node energy distribution model; Step S302: Solve the optimization problem of all virtual energy distribution sub-models based on the optimization algorithm to obtain the first power scheduling and the second power scheduling between the virtual energy distribution sub-models; The first power scheduling is the power output by the virtual energy distribution sub-model to the edge node energy distribution model; The second electric energy dispatch is the electric energy output by the virtual energy distribution sub-model to other virtual energy distribution sub-models; Step S303: Update the penalty function factor according to the optimization result. The updating process includes: First penalty function factor + 1 = first penalty function iteration multiplier × first penalty function factor; Second penalty function factor + 1 = second penalty function iteration multiplier × second penalty function factor; Step S304: Update the iteration counter k=k+1, and return to step S302 until the convergence criterion is met; The convergence criterion is when the power exchange error between the virtual energy distribution sub-models meets the convergence conditions, and the convergence conditions include: |First power interaction information-first power scheduling|<first convergence criterion; |Second power interaction information-second power scheduling|<second convergence criterion; Step S305: When the convergence criteria are met at the same time, the iterative process ends and the optimization results are output. The optimization results include the power scheduling plan and power exchange strategy of each virtual energy distribution sub-model.

6. The regional energy distributed control method based on edge computing according to claim 5, characterized in that: The second-order scheduling algorithm includes: The column constraint generation algorithm CCG is adopted to correct the decision deviation in the first-order scheduling algorithm through interactive iteration with the first-order scheduling algorithm.

7. The regional energy distributed control method based on edge computing according to claim 6, characterized in that: The second-order scheduling algorithm specifically includes: Step S311: Initialize the lower bound of the subproblem to negative infinity, the upper bound of the subproblem to positive infinity, set the iteration counter of the subproblem r=1, and set the third convergence criterion; Step S312: The convergence condition is set as: subproblem upper bound - subproblem lower bound ≥ the third convergence criterion; Step S313: iteratively optimize the optimization results of the first-order scheduling algorithm, update the dual variables and objective constraints in the calculation process of the first-order scheduling algorithm, and update the upper bound and lower bound of the subproblem; Step S314: Update r=r+1; Step S315: Optimize the currently updated subproblem as the optimization object, recursively optimize the upper bound and lower bound of the subproblem of the subproblem, and obtain the upper bound and lower bound that meet the third convergence criterion; Step S316: Stop iteration when convergence condition is met; All data volumes of the regional energy distributed control method based on edge computing are within unit time.

Images