Source network load storage collaborative optimization planning system and method based on local power distribution system

Abstract

The invention relates to the technical field of power distribution systems, and particularly discloses a source-grid-load-storage collaborative optimization planning system and method based on a local power distribution system, and the system comprises an uncertainty modeling module which is used for building a source-load output and fault uncertainty model in the local power distribution system based on collected data by employing the theories of probability statistics, random process and the like; the analogue simulation module is used for building a simulation model by utilizing PSCAD (Power System Computer Aided Design) and MATLAB / Simulink power system simulation software, simulating different uncertain scenes such as peak load, valley load and extreme weather, and observing and analyzing the operation response of the system; the operation characteristic analysis module is used for deeply analyzing the operation characteristics of the system under the influence of multiple uncertainties; according to the invention, through the multi-region local power distribution system optimization planning sub-module, the energy complementarity and the collaborative support capability among multiple regions can be considered, and the cross-region level source network load storage collaborative optimization planning is realized, which is helpful for promoting the efficient energy configuration and utilization among the regions and promoting the further development of the smart power grid.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of power distribution systems, and particularly relates to a source-grid-load-storage collaborative optimization planning system and method based on a local power distribution system. BACKGROUND

[0002] With large-scale access of renewable energy and gradual opening of the electricity market, the local power distribution system is facing unprecedented challenges and opportunities. The traditional power distribution system planning method is often based on deterministic conditions, and it is difficult to effectively cope with the complex influence of uncertain factors such as source-load output and faults. In particular, in peak load, off-peak load and extreme weather scenarios, the operation characteristics and stability of the system are easily impacted, resulting in a decrease in power supply reliability and energy utilization efficiency.

[0003] In addition, with the development of smart grid technology, source-grid-load-storage collaborative optimization has become an important means to improve the performance of the local power distribution system. However, how to realize the collaborative optimization planning of source-grid-load-storage under multiple uncertain conditions is still a problem to be solved. The existing planning methods mostly lack comprehensive consideration of uncertain factors, resulting in poor planning scheme effect in actual application and difficulty in meeting the long-term operation demand of the system. Therefore, a source-grid-load-storage collaborative optimization planning system and method based on a local power distribution system is proposed. SUMMARY

[0004] The purpose of the present application is to provide a source-grid-load-storage collaborative optimization planning system and method based on a local power distribution system to solve the problem that the existing planning methods mostly lack comprehensive consideration of uncertain factors, resulting in poor planning scheme effect in actual application and difficulty in meeting the long-term operation demand of the system.

[0005] To achieve the above-mentioned purpose, the present application provides the following technical solutions:

[0006] The source-grid-load-storage collaborative optimization planning system based on a local power distribution system comprises:

[0007] An uncertainty modeling module is used to establish an uncertainty model of source-load output and faults in the local power distribution system based on collected data and theories such as probability statistics and random processes;

[0008] A simulation module uses PSCAD, MATLAB / Simulink power system simulation software to build a simulation model to simulate different uncertain scenarios such as peak load, off-peak load and extreme weather, and to observe and analyze the operation response of the system;

[0009] An operation characteristic analysis module is used to in-depth analyze the operation characteristics of the system under the influence of multiple uncertainties and to reveal the collaborative support mechanism of internal source-grid-load-storage under multiple types of operation scenarios;

[0010] A synergistic optimization planning module, in combination with the disclosed synergistic support mechanism, establishes a multi-objective optimization planning model for an independent local power distribution system and a multi-region strong local power distribution system considering multi-source synergistic support;

[0011] An intelligent solution module, which adopts an intelligent optimization algorithm and improves and adaptively adjusts the algorithm in combination with actual constraint conditions and objective functions to solve an optimal source-grid-load-storage configuration scheme;

[0012] A power supply guarantee analysis system module, which is used to develop a local power distribution system power supply guarantee analysis system and perform typical demonstration applications;

[0013] A data-driven and knowledge fusion module, which is used to integrate external multi-source heterogeneous data, including meteorological data, energy policy data and market price data, and use deep learning technology to mine potential laws and knowledge from the data to provide more comprehensive data support and knowledge supplement for uncertainty modeling, simulation, operation characteristic analysis, synergistic optimization planning, etc., and improve the accuracy and adaptability of system planning.

[0014] Preferably, the uncertainty modeling module further comprises a source-load-output uncertainty modeling submodule and a fault uncertainty modeling submodule, the source-load-output uncertainty modeling submodule is used to establish an uncertainty model of source-load-output, and the fault uncertainty modeling submodule is used to establish an uncertainty model of fault;

[0015] The uncertainty modeling module further comprises an uncertainty dynamic adjustment submodule, which is used to monitor data and system operation state in real time, and use an online learning algorithm to dynamically adjust uncertainty model parameters of source-load-output and fault, so that the model can timely reflect the latest changes of the system and improve the real-time performance and accuracy of the model.

[0016] Preferably, the synergistic optimization planning module further comprises an independent local power distribution system optimization planning submodule and a multi-region local power distribution system optimization planning submodule, the independent local power distribution system optimization planning submodule is used for source-grid-load-storage synergistic optimization planning of an independent local power distribution system, and the multi-region local power distribution system optimization planning submodule is used for source-grid-storage synergistic optimization planning of a multi-region local power distribution system considering multi-source synergistic support;

[0017] The synergistic optimization planning module further comprises a distributed synergistic optimization submodule, which adopts a distributed computing architecture, decomposes optimization tasks of a multi-region local power distribution system into multiple subtasks, allocates the subtasks to different computing nodes for parallel optimization solution, improves the efficiency and scalability of optimization solution, and realizes a globally optimal source-grid-load-storage configuration scheme through information interaction and synergy between nodes.

[0018] Preferably, the intelligent optimization algorithm adopted by the intelligent solving module, the intelligent solving module further comprises a genetic algorithm, a particle swarm optimization algorithm and an ant colony algorithm, and can improve and adaptively adjust the algorithm according to specific constraint conditions and objective functions.

[0019] The genetic algorithm comprises an adaptive crossover probability algorithm formula and a mutation probability algorithm formula, the particle swarm optimization algorithm comprises a dynamic inertia weight formula and a nonlinear velocity update formula, and the ant colony algorithm comprises a dynamic pheromone evaporation coefficient and a heuristic information improvement formula.

[0020] Preferably, the power supply guarantee analysis system module can analyze the power supply guarantee capability of a local power distribution system in real time, provide early warning and decision support, and realize source-grid-load-storage collaborative optimization planning of the local power distribution system in actual application, thereby improving the power supply guarantee capability of the system, efficient consumption of energy resources and cost efficiency.

[0021] Preferably, the source-load output uncertainty modeling submodule and the fault uncertainty modeling submodule of the uncertainty modeling module are established based on processed data and by using probability statistics and random process theory.

[0022] Preferably, the uncertainty modeling module further has the function of establishing uncertainty models of source-load output and faults, thereby providing a theoretical basis for simulation and operation characteristic analysis.

[0023] Preferably, the source-load output uncertainty modeling submodule comprises data preprocessing, probability distribution fitting and random process simulation.

[0024] The data preprocessing is used for cleaning, denoising and normalizing historical data of collected sources and loads, so as to improve data quality, the data preprocessing collects wind power generation and photovoltaic power generation as the source, and collects power load as the load; the probability distribution fitting is used for fitting the probability distribution of source-load output data by using probability statistics theory, so as to establish an uncertainty model of source-load output, the probability distribution fitting is normal distribution and Weibull distribution; the random process simulation adopts random process theory and simulates the dynamic change process of source-load output, thereby further refining the uncertainty model, and the random process theory is any one of Markov chain or time series analysis.

[0025] Preferably, the fault uncertainty modeling submodule includes fault type classification, fault probability evaluation, and fault impact analysis. The fault type classification is used to classify possible faults in the local power distribution system, and the fault type classification is equipment failure, line failure, and weather factor caused failure. The fault probability evaluation is used to evaluate the occurrence probability of each type of fault based on historical fault data and expert experience, using methods such as Bayesian network and fault tree analysis. The fault impact analysis is used to simulate the impact on the system after the fault occurs to establish an uncertainty model of the fault, and the simulation of the impact on the system after the fault occurs includes fault duration, impact range, and recovery time.

[0026] Preferably, the independent local power distribution system optimization planning submodule includes objective function setting, constraint condition construction, and optimization algorithm application.

[0027] The objective function setting is used to set the optimization objective function according to the system operation characteristic analysis result, and the objective function is to minimize the cost and maximize the energy utilization efficiency. The constraint condition construction is used to construct the constraint condition of system operation, and the constraint condition is voltage stability, power balance, and device capacity limit. The optimization algorithm application adopts intelligent optimization algorithm to solve the optimal source network load storage configuration scheme, and the intelligent optimization algorithm is genetic algorithm and particle swarm optimization algorithm.

[0028] The multi-region local power distribution system optimization planning submodule includes multi-source collaborative support mechanism, cross-region optimization planning, and risk assessment and response.

[0029] The multi-source collaborative support mechanism is used to consider the energy complementarity and collaborative support capability among multiple regions to establish a multi-source collaborative support mechanism. The cross-region optimization planning is used to perform collaborative optimization planning of source network load storage at the cross-region level to realize efficient configuration and utilization of energy resources. The risk assessment and response is used to perform risk assessment on the optimization planning scheme, develop response measures, and ensure safe and reliable operation of the system.

[0030] Based on the local power distribution system source network load storage collaborative optimization planning method, the following steps are included:

[0031] Step 1: Collect and process data, collect historical data of sources and loads in each region, as well as meteorological data, energy policy data, and market price data and other multi-source heterogeneous data; and perform preprocessing operations such as cleaning, denoising, and normalization on the collected data to ensure data quality and consistency;

[0032] Step two: modeling the uncertainty module, using probability and statistics theory to fit the probability distribution of source and load data, and using normal distribution and Weibull distribution to establish the uncertainty model of source and load; at the same time, using random process theory to simulate the dynamic change process of source and load, further refining the uncertainty model;

[0033] Modeling the uncertainty module, classifying the possible faults in the local power distribution system, including equipment failure, line failure and weather-induced failure; at the same time, based on historical fault data, using Bayesian network, fault tree analysis and other methods to evaluate the occurrence probability of each type of fault, and simulating the impact on the system after the fault occurs, including fault duration, impact range and recovery time, etc., to establish the uncertainty model of the fault;

[0034] Adjusting the uncertainty dynamically, real-time monitoring of data and system operation state, using online learning algorithm to dynamically adjust the uncertainty model parameters of source and load and fault, so that the model can timely reflect the latest changes of the system;

[0035] Step three: establishing regional interconnection and collaborative support mechanism, analyzing the energy complementarity between regions, considering the power supply characteristics and load characteristics of different regions; at the same time, establishing a multi-source collaborative support mechanism, clarifying the roles and responsibilities of each region in the source-grid-load-storage collaborative support, and determining the mode and strategy of energy complementarity and collaborative support;

[0036] Step four: constructing a cross-regional optimization planning model, setting the objective function, setting the optimization objective function according to the system operation characteristic analysis results, considering factors such as minimizing cost and maximizing energy utilization efficiency;

[0037] Constructing the constraint condition, constructing the constraint condition of system operation, including voltage stability, power balance and equipment capacity limit, at the same time considering the factors of line current carrying capacity, transformer capacity limit and distributed power access capacity in low-voltage level of distribution network; and constraining multi-regional collaborative planning, considering the collaborative constraints brought by regional interconnection, including the capacity limit of cross-regional transmission line, the power exchange constraint between regions, to ensure the reasonable and safe energy flow between regions;

[0038] Step five: for distributed collaborative optimization solution, a distributed computing architecture is adopted, and the optimization task of the multi-region local power distribution system is decomposed into multiple sub-tasks and distributed to different computing nodes for parallel optimization solution; at the same time, through information interaction and cooperation between the computing nodes, the intermediate results and constraint conditions in the optimization process are shared, the coordination and consistency of the whole optimization process are ensured, and intelligent optimization algorithm is used, and the algorithm is improved and adaptively adjusted in combination with actual constraint conditions and objective function to solve the optimal source network load storage configuration scheme;

[0039] Step six: risk assessment is carried out on the optimization planning scheme, risk factors are identified, and corresponding countermeasures are formulated for different risk factors, the countermeasures include establishing a risk early warning mechanism, reserving emergency energy and signing long-term energy supply contracts to ensure the safe and reliable operation of the system;

[0040] Step seven: according to the optimization planning scheme, the construction and reconstruction of source network load storage are implemented, the reconstruction work includes the access of distributed power supply, the configuration of energy storage equipment and the upgrading and reconstruction of power grid, and in the process of system operation, the running state and performance index of the system are monitored in real time, the performance index includes energy utilization rate, cost index and power supply reliability, and according to the monitoring results, the optimization planning scheme is dynamically adjusted and optimized to adapt to the changes of system operation environment and the development of user demand.

[0041] Compared with the prior art, the beneficial effects of the present application are:

[0042] Through the uncertainty modeling module, the present application can establish the uncertainty model of source load output and fault based on the collected data by using probability statistics, random process and other theories, which provides a theoretical basis for simulation and operation characteristic analysis, which helps to more accurately reflect the actual operation of the system and improve the accuracy and reliability of the planning scheme;

[0043] The simulation module can simulate different uncertain scenarios such as peak load, valley load and extreme weather, observe and analyze the running response of the system, which helps to evaluate the adaptability of the planning scheme under different scenarios and ensure the stable operation of the system under various conditions;

[0044] The operation characteristic analysis module can deeply analyze the operation characteristics of the system under the influence of multiple uncertainties, and reveal the collaborative support mechanism of internal source network load storage under multiple types of operation scenarios, which helps to develop more reasonable resource allocation strategies and maximize the efficient consumption of energy resources and cost-effectiveness;

[0045] The intelligent solving module adopts intelligent optimization algorithms such as genetic algorithm, particle swarm optimization algorithm and ant colony algorithm, and improves and adaptively adjusts the algorithm in combination with actual constraint conditions and objective functions, which is helpful for quickly solving an optimal source-grid-load-storage configuration scheme and improving the efficiency and accuracy of planning work;

[0046] The power supply guarantee analysis system module can analyze the power supply guarantee capability of the local power distribution system in real time, provide early warning and decision support, which is helpful for timely discovering and processing potential power supply risks and ensuring the safe and reliable operation of the system;

[0047] Through the multi-region local power distribution system optimization planning sub-module, the energy complementarity and collaborative support capability among multiple regions can be considered, and the source-grid-load-storage collaborative optimization planning across regional levels can be realized, which is helpful for promoting efficient energy configuration and utilization among regions and further developing the smart grid. BRIEF DESCRIPTION OF DRAWINGS

[0048] Figure 1 The module operation diagram of the application;

[0049] Figure 2 The module operation diagram of the application; DETAILED DESCRIPTION

[0050] The technical solutions in the embodiments of the application will be clearly and completely described below with reference to the drawings in the embodiments of the application. Obviously, the described embodiments are only part of the embodiments of the application, rather than all the embodiments of the application. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the application.

[0051] As shown in Figure 1 The local power distribution system source-grid-load-storage collaborative optimization planning system based on the local power distribution system source-grid-load-storage collaborative optimization planning system comprises:

[0052] An uncertainty modeling module is used to establish an uncertainty model of source-load output and faults in the local power distribution system based on collected data by using probability statistics, random process and other theories;

[0053] Specifically, the uncertainty modeling module further comprises a source-load output uncertainty modeling sub-module and a fault uncertainty modeling sub-module, the source-load output uncertainty modeling sub-module is used to establish an uncertainty model of source-load output, and the fault uncertainty modeling sub-module is used to establish an uncertainty model of faults;

[0054] Specifically, the source-load-power uncertainty modeling submodule and the fault uncertainty modeling submodule of the uncertainty modeling module are established based on processed data and using probability statistics and random process theory; the uncertainty modeling module also has the function of establishing uncertainty models of source-load-power and faults, providing a theoretical basis for simulation and operating characteristic analysis; the source-load-power uncertainty modeling submodule includes data preprocessing, probability distribution fitting, and random process simulation;

[0055] Specifically, the fault uncertainty modeling submodule includes fault type classification, fault probability evaluation, and fault impact analysis; the fault type classification is used to classify possible faults in the local power distribution system, and the fault type classification is equipment failure, line failure, and failure caused by weather factors; the fault probability evaluation is used to evaluate the occurrence probability of each type of fault based on historical fault data and expert experience, using methods such as Bayesian network and fault tree analysis; the fault impact analysis is used to simulate the impact on the system after the fault occurs to establish an uncertainty model of the fault; the simulation of the impact on the system after the fault occurs includes fault duration, impact range, and recovery time;

[0056] The uncertainty modeling module also includes an uncertainty dynamic adjustment submodule for real-time monitoring of data and system operating state, using online learning algorithms to dynamically adjust the uncertainty model parameters of source-load-power and faults, so that the model can timely reflect the latest changes in the system and improve the real-time performance and accuracy of the model;

[0057] As can be seen from the above, the uncertainty of distributed new energy is constructed:

[0058] The source-load-power uncertainty modeling submodule is responsible for constructing the uncertainty model of distributed new energy, and the specific steps are as follows:

[0059] Data preprocessing: collect historical data of wind power generation and photovoltaic power generation, as well as power load data; clean the data to remove outliers and noise; normalize the data for subsequent probability distribution fitting and random process simulation;

[0060] Probability distribution fitting: use probability statistics theory to fit the probability distribution of source-load-power data; for example, the output of wind power generation and photovoltaic power generation may conform to normal distribution or Weibull distribution; the probability distribution obtained by fitting can quantify the uncertainty of source-load-power;

[0061] Random process simulation: use random process theory such as Markov chain or time series analysis to simulate the dynamic change process of source-load-power; through simulation, the uncertainty model can be further refined to consider the time correlation and dynamic change characteristics of source-load-power;

[0062] Uncertainty modeling of power grid failures:

[0063] The fault uncertainty modeling submodule is responsible for building the uncertainty model of power grid failures. The specific steps are as follows:

[0064] Fault type classification: classify the possible faults in the local power distribution system, such as equipment failure, line failure and weather-induced failure;

[0065] Fault probability assessment: based on historical fault data, use Bayesian network, fault tree analysis and other methods to assess the probability of each type of fault; through assessment, the uncertainty of power grid failure can be quantified;

[0066] Fault impact analysis: simulate the impact of the fault on the system after the fault occurs, including fault duration, impact range and recovery time; through simulation, the uncertainty model of the fault can be established, providing a basis for subsequent collaborative optimization planning;

[0067] The principle of the application of the uncertainty dynamic adjustment submodule in the planning system is as follows:

[0068] Sensor deployment: deploy various sensors in the local power distribution system to monitor source-load output data and system operation state data in real time.

[0069] Data collection and transmission: through the data collection system, the data monitored by the sensors are transmitted to the uncertainty dynamic adjustment submodule in real time;

[0070] Algorithm selection: select appropriate online learning algorithms, such as online gradient descent algorithm, online support vector machine algorithm, etc., for dynamic adjustment of uncertainty model parameters;

[0071] Model parameter update: based on the real-time monitored data, use online learning algorithms to continuously update the uncertainty model parameters of source-load output and fault, so that the model can timely reflect the latest changes of the system;

[0072] Real-time feedback: the adjusted model parameters are fed back to the uncertainty modeling module in real time to update the uncertainty model established by it;

[0073] Model evaluation and optimization: periodically evaluate the adjusted uncertainty model, and further optimize the online learning algorithm and model parameter update strategy according to the evaluation results, to improve the real-time performance and accuracy of the model;

[0074] Through the above technical solution, real-time monitoring of data and system operation state can be realized, and online learning algorithm is used to dynamically adjust the uncertainty model parameters of source load output and fault, so that the model can timely reflect the latest changes of the system. For example, when the system operation state changes suddenly or new situations occur in external environmental factors, the model can be quickly adjusted to avoid planning deviation caused by model lag;

[0075] By dynamically adjusting the parameters, the model can more accurately describe the uncertainty of the system, improving the accuracy of uncertainty modeling. This is crucial for subsequent simulation, operation characteristic analysis and collaborative optimization planning, because a more accurate model can provide more reliable reference, thus obtaining a better planning scheme;

[0076] The ability of the system to respond to various uncertainties and changes is enhanced, so that the planning system can better adapt to the complex and changing actual operation environment. Whether it is sudden fluctuations in source load output or changes in the probability of fault occurrence, the model parameters can be dynamically adjusted to respond in time, ensuring the stable operation of the system and the effectiveness of the planning.

[0077] Through the uncertainty modeling module, the present application can establish the uncertainty model of source load output and fault based on the collected data, using probability statistics, random process and other theories, to provide a theoretical basis for simulation and operation characteristic analysis, which helps to more accurately reflect the actual operation of the system and improve the accuracy and reliability of the planning scheme;

[0078] The constructed uncertainty model can be applied to the source network load storage collaborative planning scheme to improve the power supply capacity, and the specific steps are as follows:

[0079] Simulation: Use PSCAD, MATLAB / Simulink and other power system simulation software to build a simulation model; simulate different uncertainty scenarios such as peak load, off-peak load, extreme weather; observe and analyze the system's response under these scenarios to provide data support for subsequent collaborative optimization planning;

[0080] Operation characteristic analysis: in-depth analysis of the operation characteristics of the system under the influence of multiple uncertainties; reveal the collaborative support mechanism of internal source network load storage under multiple types of operation scenarios; through analysis, the weak links and potential risks of the system under different uncertainty scenarios can be determined;

[0081] Coordinated optimization planning: Based on the revealed coordination support mechanism, a multi-objective optimization planning model for independent local power distribution systems and multi-region strong local power distribution systems considering multi-source coordination support is established. The uncertainty of distributed new energy and the uncertainty of power grid failure are considered in the model, and the optimization objective function and constraint conditions are set. The optimal source-grid-load-storage configuration scheme is solved by intelligent optimization algorithm to improve the power supply capacity and reliability of the system.

[0082] Simulation module: PSCAD, MATLAB / Simulink power system simulation software is used to build a simulation model to simulate different uncertainty scenarios such as peak load, off-peak load, extreme weather, etc., and observe and analyze the system's operation response.

[0083] As can be seen from the above, the simulation module can simulate different uncertainty scenarios such as peak load, off-peak load, extreme weather, etc., and observe and analyze the system's operation response, which helps to evaluate the adaptability of the planning scheme under different scenarios and ensure the stable operation of the system under various conditions.

[0084] Operation characteristic analysis module: in-depth analysis of the operation characteristics of the system under the influence of multiple uncertainties, revealing the coordination support mechanism of internal source-grid-load-storage under multiple types of operation scenarios.

[0085] Coordinated optimization planning module: based on the revealed coordination support mechanism, a multi-objective optimization planning model for independent local power distribution systems and multi-region strong local power distribution systems considering multi-source coordination support is established.

[0086] Specifically, the coordinated optimization planning module further includes an independent local power distribution system optimization planning submodule and a multi-region local power distribution system optimization planning submodule. The independent local power distribution system optimization planning submodule is used for source-grid-load-storage coordinated optimization planning of independent local power distribution systems, and the multi-region local power distribution system optimization planning submodule is used for source-grid-storage coordinated optimization planning of multi-region local power distribution systems considering multi-source coordination support.

[0087] The coordinated optimization planning module further includes a distributed coordinated optimization submodule. The distributed coordinated optimization submodule adopts a distributed computing architecture, decomposes the optimization task of the multi-region local power distribution system into multiple subtasks, assigns them to different computing nodes for parallel optimization solution, improves the efficiency and scalability of optimization solution, and realizes the globally optimal source-grid-load-storage configuration scheme through information interaction and coordination between nodes.

[0088] As can be seen from the above, the principle of the application of the distributed coordinated optimization submodule in the planning system is as follows:

[0089] Computing node deployment: Deploy multiple computing nodes in the multi-region local power distribution system, each with certain computing and storage capabilities;

[0090] Network communication establishment: Establish network communication mechanisms between computing nodes to ensure efficient information exchange and collaborative computation among nodes;

[0091] Task decomposition: Decompose the optimization task of the multi-region local power distribution system into multiple subtasks, each corresponding to the optimization problem of a region or part of the system;

[0092] Task allocation: According to the computing capacity and load of the computing nodes, allocate the decomposed subtasks to different computing nodes for parallel optimization solution;

[0093] Local optimization solution: Each computing node uses intelligent optimization algorithms to solve the assigned subtasks for local optimization, obtaining a locally optimal source-grid-load-storage configuration scheme;

[0094] Information exchange and collaboration: Computing nodes regularly exchange information, share local optimization results and intermediate calculation data, adjust optimization direction and parameters through collaborative mechanisms, and gradually approach the global optimal solution;

[0095] Result aggregation and analysis: Aggregate and analyze the local optimization results of each computing node, considering the energy complementarity and collaborative support capacity between regions;

[0096] Global optimal determination: Through further optimization and adjustment, determine the globally optimal source-grid-load-storage configuration scheme and feed it back to other submodules of the collaborative optimization planning module to guide the actual operation and planning of the system;

[0097] Through the above technical solutions, a distributed computing architecture is adopted to decompose the optimization task of the multi-region local power distribution system into multiple subtasks and allocate them to different computing nodes for parallel optimization solution. This approach fully utilizes the computing resources of multiple computing nodes, greatly shortens the optimization solution time, and improves the system response speed;

[0098] The distributed architecture allows the system to easily expand computing nodes to handle larger-scale multi-region local power distribution system optimization problems. When the system size increases or the optimization task increases, only the corresponding computing nodes need to be added, without the need for large-scale modification of the entire system, reducing the system expansion cost and difficulty;

[0099] Through information interaction and cooperation between nodes, on the basis of parallel solving of local sub-tasks, the energy complementarity and collaborative support capability between regions can be comprehensively considered to realize a globally optimal source-grid-load-storage configuration scheme, avoid the problem of poor overall performance caused by local optimal solution, and improve the overall operation efficiency and energy utilization efficiency of the system.

[0100] Specifically, the independent local power distribution system optimization planning submodule includes target function setting, constraint condition construction, and optimization algorithm application.

[0101] The target function setting is used to set the optimization target function according to the system operation characteristic analysis result, the target function being to minimize cost and maximize energy utilization efficiency; the constraint condition construction is used to construct the constraint conditions for system operation, the constraint conditions being voltage stability, power balance, and device capacity limit; and the optimization algorithm application adopts an intelligent optimization algorithm to solve the optimal source-grid-load-storage configuration scheme, the intelligent optimization algorithm being a genetic algorithm and a particle swarm optimization algorithm.

[0102] The multi-region local power distribution system optimization planning submodule includes a multi-source collaborative support mechanism, cross-region optimization planning, and risk assessment and response.

[0103] The multi-source collaborative support mechanism is used to consider the energy complementarity and collaborative support capability between multi-regions to establish a multi-source collaborative support mechanism; the cross-region optimization planning is used to perform collaborative optimization planning of sources, grids, loads, and storages at the cross-region level to realize efficient configuration and utilization of energy resources; and the risk assessment and response is used to perform risk assessment on the optimization planning scheme, develop response measures, and ensure safe and reliable operation of the system.

[0104] From the above, the independent local power distribution system optimization planning submodule is responsible for planning of the local power distribution system, and the specific implementation steps are as follows:

[0105] Target function setting: set the optimization target function according to the system operation characteristic analysis result, such as minimizing cost and maximizing energy utilization efficiency;

[0106] Constraint condition construction: construct the constraint conditions for system operation, such as voltage stability, power balance, and device capacity limit; and optimization algorithm application: adopt an intelligent optimization algorithm to solve the optimal source-grid-load-storage configuration scheme.

[0107] The multi-region local power distribution system optimization planning submodule is responsible for planning of multi-region sources, grids, loads, and storages, and the specific steps are as follows:

[0108] Multi-source collaborative support mechanism: consider the energy complementarity and collaborative support capability between multi-regions to establish a multi-source collaborative support mechanism; through collaborative support, realize optimization configuration and sharing of energy resources between multi-regions;

[0109] Cross-regional optimization planning: collaborative optimization planning of source, network, load and storage at the cross-regional level; considering power transmission and distribution between different regions, achieving efficient allocation and utilization of energy resources;

[0110] Risk assessment and response: risk assessment of optimization planning scheme, identification of potential risks and uncertainty factors; develop response measures such as establishing backup power, optimizing dispatching strategy, etc. to ensure safe and reliable operation of the system.

[0111] The operation characteristic analysis module can deeply analyze the operation characteristics of the system under the influence of multiple uncertainties, and reveal the collaborative support mechanism of internal source, network, load and storage under multiple types of operation scenarios, which helps to develop more reasonable resource allocation strategies and maximize the efficient utilization and cost-effectiveness of energy resources.

[0112] Intelligent solving module, using intelligent optimization algorithm, and combining actual constraint conditions and objective function to improve and adaptively adjust the algorithm to solve the optimal source, network, load and storage configuration scheme;

[0113] Specifically, the intelligent optimization algorithm used by the intelligent solving module includes genetic algorithm, particle swarm optimization algorithm and ant colony algorithm, and can improve and adaptively adjust the algorithm according to specific constraint conditions and objective functions;

[0114] The genetic algorithm includes adaptive crossover probability algorithm formula and mutation probability algorithm formula, the particle swarm optimization algorithm includes dynamic inertia weight formula and nonlinear velocity update formula, and the ant colony algorithm includes dynamic pheromone evaporation coefficient and heuristic information improvement formula;

[0115] Specifically, the power supply guarantee analysis system module can analyze the power supply guarantee capability of the local power distribution system in real time, provide early warning and decision support, and the system can realize the collaborative optimization planning of source, network, load and storage of the local power distribution system in practical application, improve the power supply guarantee capability of the system, the efficient utilization of energy resources and cost-effectiveness;

[0116] The data preprocessing is used for cleaning, denoising and normalizing the collected historical data of source and load to improve data quality, the data preprocessing collects wind power and photovoltaic power as source, and collects power load as load; the probability distribution fitting is used to use probability statistics theory to fit the probability distribution of source and load output data to establish an uncertainty model of source and load output, the probability distribution fitting is normal distribution and Weibull distribution; the random process simulation adopts random process theory and simulates the dynamic change process of source and load output to further refine the uncertainty model, the random process theory is any one of Markov chain or time series analysis;

[0117] As can be seen from the above, the intelligent solving module adopts intelligent optimization algorithms such as genetic algorithm, particle swarm optimization algorithm and ant colony algorithm, and combines actual constraint conditions and objective functions to improve and adapt the algorithm, which helps to quickly solve the optimal source network load storage configuration scheme and improve the efficiency and accuracy of planning work.

[0118] The power supply guarantee analysis system module is used to develop a local power distribution system power supply guarantee analysis system for typical demonstration application;

[0119] As can be seen from the above, the power supply guarantee analysis system module can analyze the power supply guarantee capability of the local power distribution system in real time, provide early warning and decision support, which helps to discover and handle potential power supply risks in time and ensure the safe and reliable operation of the system.

[0120] The data-driven and knowledge fusion module is used to integrate external multi-source heterogeneous data, including meteorological data, energy policy data and market price data, and use deep learning technology to mine potential rules and knowledge from the data, providing more comprehensive data support and knowledge supplement for uncertainty modeling, simulation, operation characteristic analysis, collaborative optimization planning, etc., improving the accuracy and adaptability of system planning;

[0121] As can be seen from the above, the principle of the application of the data-driven and knowledge fusion module in the planning system is as follows:

[0122] Data source access: Establish data interfaces with meteorological departments, energy policy publishing agencies, energy market trading platforms, etc., to obtain meteorological data, energy policy data and market price data in real time or regularly;

[0123] Data cleaning and preprocessing: Clean the collected multi-source heterogeneous data, remove noise, missing values and outliers, and perform format unification and standardization processing for subsequent analysis;

[0124] Model construction: According to the data characteristics and analysis goals, select appropriate deep learning models, such as convolutional neural network (CNN) for processing meteorological image data, and recurrent neural network (RNN) and its variants for processing time series data;

[0125] Model training: Use preprocessed data to train the deep learning model, adjust the model parameters, and make it able to accurately mine the potential rules and knowledge in the data;

[0126] Knowledge extraction: Extract useful knowledge from the trained model, such as the association rules between meteorological conditions and source load output, the influence mode of energy policy on system operation, etc.;

[0127] Uncertainty modeling: Incorporate the association knowledge between meteorological data mined by deep learning and source-load output into the uncertainty modeling module to improve the accuracy of source-load output uncertainty model;

[0128] Simulation: Adjust the parameters in the simulation module using market price data and energy policy data to simulate the system operation under different policy environments and market conditions;

[0129] Operation characteristic analysis: In-depth analysis of the operation characteristics of the system under different scenarios based on the mined knowledge, providing a more comprehensive basis for collaborative optimization planning;

[0130] Collaborative optimization planning: Supplement the optimization objective function and constraint conditions with knowledge to make the optimization planning scheme more in line with the actual situation;

[0131] Through the above technical solutions, the addition of data-driven and knowledge fusion modules in the planning system can integrate meteorological, energy policy, market price and other multi-source heterogeneous data, breaking the limitation of single data source in traditional planning systems and providing a more comprehensive data basis for system planning. For example, meteorological data can more accurately predict source-load output, energy policy data can reflect the impact of policy guidance on the system in a timely manner, and market price data can help optimize the economy of the system;

[0132] Using deep learning technology to mine potential laws and knowledge from massive data, these knowledge can deeply reflect the operation mechanism of the system and the internal relationship between external environmental factors and the system. For example, the complex association between meteorological conditions and source-load output is mined to provide more accurate basis for uncertainty modeling, making the model more reflect the actual situation;

[0133] Providing more comprehensive data support and knowledge supplement for uncertainty modeling, simulation, operation characteristic analysis, collaborative optimization planning and other modules helps to improve the accuracy of system planning. At the same time, due to the consideration of multiple factors and potential laws, the planning scheme can better adapt to different scenarios and changes, enhancing the adaptability and robustness of the system.

[0134] In summary, through the multi-regional local power distribution system optimization planning sub-module, the energy complementarity and collaborative support capacity between multiple regions can be considered, realizing cross-regional source-grid-load-storage collaborative optimization planning, which helps to promote efficient energy allocation and utilization between regions and further development of smart grid;

[0135] The uncertainty modeling, simulation, operational characteristics analysis, collaborative optimization planning, intelligent solution, power supply guarantee analysis and data-driven and knowledge integration modules are systematically integrated to form a closed-loop optimization system. Through data interaction and function complementation among the modules, the whole process automation from uncertainty modeling to actual planning scheme is realized, and the planning efficiency and reliability are significantly improved.

[0136] An online learning algorithm is introduced in the source load output and fault modeling to dynamically adjust the model parameters through real-time monitoring data, so that the system can quickly adapt to environmental changes. The traditional static model only relies on historical data, while the dynamic adjustment sub-module realizes real-time updating of model parameters through adaptive mechanism, improving the model prediction accuracy.

[0137] A distributed computing architecture is proposed to decompose multi-region optimization tasks into sub-tasks for parallel solution, and to realize global optimization through information interaction among nodes. The traditional centralized optimization has a calculation bottleneck, and the distributed architecture significantly reduces the computational complexity, which is suitable for large-scale multi-region system collaborative planning. Multi-source heterogeneous data such as weather, energy policy and market price are integrated, and deep learning technology is used to mine potential laws to provide data-driven decision support for planning.

[0138] For the first time, external data is combined with internal data of the power system to enhance the economic efficiency and policy adaptability of the planning. The power supply guarantee analysis system module supports real-time analysis of the power supply capacity of local distribution systems, dynamically evaluates the risks and proposes countermeasures. The traditional power supply analysis is mostly offline, and the real-time analysis function can quickly respond to sudden failures or load changes to improve system resilience.

[0139] Normal distribution and Weibull distribution are used to fit the probability distribution of source load output, and Markov chain or time series analysis is used to simulate dynamic changes. Based on Bayesian network and fault tree analysis, the fault probability is evaluated, and a multi-dimensional model is established by simulating the impact of faults. Through the combination of probability statistics and random process, high-precision uncertainty quantification of source load output and faults is realized.

[0140] Independent local distribution system optimization aims to minimize cost and maximize energy utilization efficiency, while multi-region system introduces multi-source collaborative support mechanism to realize cross-regional resource complementation. Traditional planning mainly focuses on a single target, while multi-objective model considers economic efficiency, efficiency and reliability, which meets the needs of new power systems.

[0141] Genetic algorithm, particle swarm optimization algorithm and ant colony algorithm are improved, such as adaptive crossover probability, dynamic inertia weight and nonlinear velocity update. Traditional algorithms are prone to local optimum, and the improved algorithms improve global search ability and accelerate convergence through dynamic parameter adjustment.

[0142] Introducing risk assessment and response modules in multi-region optimization planning to quantify the risks of planning schemes and develop emergency plans; traditional planning lacks dynamic risk assessment, and this module improves the system's risk resistance through probability analysis and scenario simulation;

[0143] Developing a power supply guarantee analysis system to support demonstration applications in typical scenarios and promote the transformation of technological achievements into practical engineering; a complete closed-loop design from theoretical models to engineering implementation enhances the practicality and promotional value of the system;

[0144] This system breaks through the bottlenecks of traditional distribution system planning, such as insufficient quantification of uncertainty, low computational efficiency, and lack of coordination mechanisms, by using key technologies such as modular design, dynamic modeling, distributed computing, multi-source data fusion, and intelligent algorithm optimization, providing an innovative solution for building a highly resilient and reliable local power distribution system.

[0145] Based on the local power distribution system source-network-load-storage collaborative optimization planning method, the following steps are included:

[0146] Step one: Collect and process data, collect historical data of sources and loads in each region, as well as meteorological data, energy policy data, and market price data and other multi-source heterogeneous data; and perform preprocessing operations such as cleaning, denoising, and normalization on the collected data to ensure data quality and consistency;

[0147] Step two: Model the uncertainty module, use probability and statistics theory to fit the probability distribution of source and load output data, and use normal distribution and Weibull distribution to establish the uncertainty model of source and load output; at the same time, use stochastic process theory to simulate the dynamic change process of source and load output, further refining the uncertainty model;

[0148] Model the uncertainty module, classify the possible faults in the local power distribution system, including device faults, line faults, and weather-induced faults; at the same time, based on historical fault data, use Bayesian networks and fault tree analysis to evaluate the occurrence probability of each type of fault, and simulate the impact on the system after the fault occurs, including fault duration, impact range, and recovery time, to establish the uncertainty model of the fault;

[0149] Adjust the uncertainty dynamically, monitor data and system operation status in real time, and use online learning algorithms to dynamically adjust the uncertainty model parameters of source and load output and faults, so that the model can timely reflect the latest changes in the system;

[0150] Step three: Establish regional interconnection and collaborative support mechanism, analyze the energy complementarity between regions, consider the power supply characteristics and load characteristics of different regions; At the same time, establish a multi-source collaborative support mechanism, clarify the roles and responsibilities of each region in the source network load storage collaborative support, and determine the mode and strategy of energy complementarity and collaborative support;

[0151] Step four: Construct a cross-regional optimization planning model, set the objective function, and set the optimization objective function according to the system operation characteristic analysis results, considering factors such as minimizing cost and maximizing energy utilization efficiency;

[0152] The constraint conditions are constructed, and the constraint conditions of system operation are constructed, including voltage stability, power balance and device capacity limit, while considering the line current carrying capacity, transformer capacity limit and distributed power access capacity in low-voltage level in distribution network; And the multi-regional collaborative planning is constrained, considering the collaborative constraints brought by regional interconnection, including the capacity limit of cross-regional transmission lines, the power exchange constraints between regions, to ensure the reasonable and safe energy flow between regions;

[0153] Step five: Distributed collaborative optimization solution, using distributed computing architecture, decomposing the optimization task of multi-regional local power distribution system into multiple subtasks and distributing them to different computing nodes for parallel optimization solution; At the same time, through information interaction and collaboration between computing nodes, share the intermediate results and constraint conditions in the optimization process, ensure the coordination and consistency of the whole optimization process, and use intelligent optimization algorithm, and combine with actual constraint conditions and objective function to improve and adapt the algorithm, and solve the optimal source network load storage configuration scheme;

[0154] Step six: Risk assessment of optimization planning scheme, identify risk factors, and develop corresponding measures for different risk factors, including establishing a risk early warning mechanism, reserving emergency energy and signing long-term energy supply contracts to ensure safe and reliable operation of the system;

[0155] Step seven: According to the optimization planning scheme, implement the construction and transformation of source network load storage, including the access of distributed power supply, the configuration of energy storage equipment and the upgrading and reconstruction of power grid, and in the process of system operation, real-time monitoring of system operation state and performance index, including energy utilization rate, cost index and power supply reliability, and according to the monitoring results, dynamically adjust and optimize the optimization planning scheme to adapt to the changes of system operation environment and the development of user demand.

[0156] Further, the design application is applied to the collaborative optimization planning of power distribution system source network load storage, through the multi-region local power distribution system optimization planning submodule, the energy complementarity and collaborative support ability between multiple regions can be considered, the cross-region level source network load storage collaborative optimization planning is realized, which is helpful to promote the efficient energy allocation and utilization between regions and promote the further development of smart grid

[0157] While embodiments of the application have been shown and described, it is to be understood that the embodiments described are merely exemplary of the principles and application of the present application. Numerous modifications, changes, omissions, substitutions and adaptations can be made by those skilled in the art without departing from the application, which is defined by the following claims and their equivalents.

Claims

1. A local power distribution system source-grid-load-storage collaborative optimization planning system, characterized in that, The application relates to a local power supply system uncertainty modeling and optimization method and system. The application comprises: An uncertainty modeling module, which is used for establishing an uncertainty model of source-load output and faults in a local power supply system based on collected data by using probability statistics, random process and the like theory; A simulation modeling module, which is used for simulating different uncertainty scenes such as peak load, valley load and extreme weather by using PSCAD, MATLAB / Simulink power system simulation software to build a simulation model, observing and analyzing the operation response of the system; An operation characteristic analysis module, which is used for deeply analyzing the operation characteristics of the system under the influence of multiple uncertainties, and revealing the collaborative support mechanism of internal source-grid-load-storage under multiple operation scenes; A collaborative optimization planning module, which is used for establishing a multi-objective optimization planning model of an independent local power supply system and a multi-region strong local power supply system considering multi-source collaborative support by combining the revealed collaborative support mechanism; An intelligent solving module, which is used for solving an optimal source-grid-load-storage configuration scheme by using an intelligent optimization algorithm and improving and adaptively adjusting the algorithm in combination with actual constraint conditions and objective functions; A power supply guarantee analysis system module, which is used for developing a local power supply system power supply guarantee analysis system and performing typical demonstration application; 2.The local power distribution system source-network-load-storage collaborative optimization planning system according to claim 1, characterized in that: A data driving and knowledge fusion module, which is used for integrating external multi-source heterogeneous data including meteorological data, energy policy data and market price data, and using deep learning technology to mine potential rules and knowledge from the data. The uncertainty modeling module further comprises a source-load output uncertainty modeling submodule and a fault uncertainty modeling submodule, the source-load output uncertainty modeling submodule is used for establishing an uncertainty model of source-load output, and the fault uncertainty modeling submodule is used for establishing an uncertainty model of faults; The uncertainty modeling module further comprises an uncertainty dynamic adjustment submodule, which is used for monitoring data and system operation states in real time, dynamically adjusting uncertainty model parameters of source-load output and faults by using an online learning algorithm, and enabling the model to timely reflect the latest changes of the system; The collaborative optimization planning module further comprises an independent local power supply system optimization planning submodule and a multi-region local power supply system optimization planning submodule, the independent local power supply system optimization planning submodule is used for source-grid-load-storage collaborative optimization planning of an independent local power supply system, and the multi-region local power supply system optimization planning submodule is used for source-grid-storage collaborative optimization planning of a multi-region local power supply system considering multi-source collaborative support; 3.The local power distribution system source-network-load-storage collaborative optimization planning system according to claim 1, characterized in that: The collaborative optimization planning module further comprises a distributed collaborative optimization submodule, which adopts a distributed computing architecture, decomposes the optimization task of the multi-region local power supply system into multiple subtasks, allocates the subtasks to different computing nodes for parallel optimization solving, and simultaneously realizes information interaction and cooperation between the nodes. The intelligent solving module adopts an intelligent optimization algorithm, and the intelligent solving module further comprises a genetic algorithm, a particle swarm optimization algorithm and an ant colony algorithm; The genetic algorithm comprises an adaptive crossover probability algorithm formula and a mutation probability algorithm formula, the particle swarm optimization algorithm comprises a dynamic inertia weight formula and a nonlinear velocity updating formula, and the ant colony algorithm comprises a dynamic pheromone evaporation coefficient and a heuristic information improvement formula. 4.The local power distribution system source-network-load-storage collaborative optimization planning system according to claim 1, characterized in that: The power supply guarantee analysis system module can analyze the power supply guarantee capability of the local power distribution system in real time.

5. The local power distribution system source-network-load-storage collaborative optimization planning system according to claim 1, wherein: The source-load-output uncertainty modeling submodule and the fault uncertainty modeling submodule of the uncertainty modeling module are established based on processed data and by using probability statistics and random process theory. 6.The local power distribution system source-network-load-storage collaborative optimization planning system according to claim 1, characterized in that: The uncertainty modeling module also has the function of establishing uncertainty models of source-load-output and faults, providing a theoretical basis for simulation and operation characteristic analysis.

7. The local power distribution system source-network-load-storage collaborative optimization planning system according to claim 2, wherein: The source-load-output uncertainty modeling submodule includes data preprocessing, probability distribution fitting and random process simulation. The data preprocessing is used for cleaning, denoising and normalizing the collected historical data of sources and loads. The data preprocessing collects wind power generation and photovoltaic power generation as sources, and collects power load as loads. 8.The local power distribution system source-network-load-storage collaborative optimization planning system according to claim 2, characterized in that: The probability distribution fitting is used for probability distribution fitting of source-load-output data by using probability statistics theory, to establish an uncertainty model of source-load-output. The random process simulation simulates the dynamic change process of source-load-output by using random process theory, to further refine the uncertainty model. 9.The local power distribution system source-network-load-storage collaborative optimization planning system according to claim 2, characterized in that: The random process theory is any one of Markov chain or time series analysis. The fault uncertainty modeling submodule includes fault type classification, fault probability evaluation and fault impact analysis. The fault type classification is used for classifying possible faults in the local power distribution system. The fault type classification includes equipment faults, line faults and faults caused by weather factors. The fault probability evaluation is used for evaluating the occurrence probability of each type of fault by using Bayesian network, fault tree analysis and other methods based on historical fault data and expert experience. The fault impact analysis is used for simulating the impact of faults on the system after the faults occur, to establish an uncertainty model of faults. The simulation of the impact of faults on the system after the faults occur includes fault duration, impact range and recovery time. The independent local power distribution system optimization planning submodule includes objective function setting, constraint condition construction and optimization algorithm application. The objective function setting is used for setting the optimization objective function according to the system operation characteristic analysis result. The objective function is to minimize cost and maximize energy utilization efficiency. The constraint condition construction is used for constructing the constraint conditions of system operation. The constraint conditions are voltage stability, power balance and equipment capacity limit. The optimization algorithm application adopts intelligent optimization algorithms to solve the optimal source-grid-load-storage configuration scheme. The intelligent optimization algorithms are genetic algorithm and particle swarm optimization algorithm. The multi-region local power distribution system optimization planning submodule includes multi-source collaborative support mechanism, cross-region optimization planning and risk assessment and response. The multi-source collaborative support mechanism is used to consider the energy complementarity and collaborative support capability among multiple regions, so as to establish a multi-source collaborative support mechanism; the cross-regional optimization planning is used to perform collaborative optimization planning of sources, networks, loads and storages at a cross-regional level, so as to realize efficient configuration and utilization of energy resources; and the risk assessment and response are used to perform risk assessment on the optimization planning scheme, and develop response measures to ensure safe and reliable operation of the system.

10. A method for source-grid-load-storage collaborative optimization planning based on a local power distribution system, characterized in that, The method comprises the following steps: Step one: collecting and processing data, collecting historical data of sources and loads in each region, as well as meteorological data, energy policy data and market price data and other multi-source heterogeneous data; and performing preprocessing operations such as cleaning, denoising and normalization on the collected data to ensure the quality and consistency of the data; Step two: modeling the uncertainty module, using probability and statistics theory to fit the probability distribution of source and load output data, and using normal distribution and Weibull distribution to establish the uncertainty model of source and load output; at the same time, using stochastic process theory to simulate the dynamic change process of source and load output, and further refining the uncertainty model; Modeling the fault uncertainty module, classifying the possible faults in the local power distribution system, including device faults, line faults and weather factor caused faults; at the same time, based on historical fault data, using Bayesian network, fault tree analysis and other methods to evaluate the occurrence probability of each type of fault, and simulating the influence on the system after the fault occurs, including fault duration, influence range and recovery time, etc., to establish the uncertainty model of the fault; Adjusting the uncertainty dynamics in real time, monitoring the data and system operation state in real time, and using online learning algorithm to dynamically adjust the uncertainty model parameters of source and load output and fault, so that the model can timely reflect the latest changes of the system; Step three: establishing the regional interconnection and collaborative support mechanism, analyzing the energy complementarity among regions, considering the power source characteristics and load characteristics of different regions; at the same time, establishing a multi-source collaborative support mechanism, clarifying the roles and responsibilities of each region in the source-network-load-storage collaborative support, and determining the mode and strategy of energy complementarity and collaborative support; Step four: constructing a cross-regional optimization planning model, setting the objective function, setting the optimization objective function according to the system operation characteristic analysis result, and comprehensively considering factors such as minimizing cost and maximizing energy utilization efficiency; Constructing the constraint condition, constructing the constraint condition of system operation, including voltage stability, power balance and device capacity limitation, at the same time, considering the factors such as current carrying capacity of line, capacity limitation of transformer and access capacity of distributed power in low-voltage level of distribution network; and constraining the multi-regional collaborative planning, considering the collaborative constraints brought by regional interconnection, including capacity limitation of cross-regional transmission line, power exchange constraint between regions, to ensure that the energy flow among regions is reasonable and safe; Step five: for distributed collaborative optimization solution, a distributed computing architecture is adopted to decompose the optimization task of multi-region local power distribution system into multiple sub-tasks, which are distributed to different computing nodes for parallel optimization solution; at the same time, through information interaction and cooperation among the computing nodes, the intermediate results and constraint conditions in the optimization process are shared to ensure the coordination and consistency of the whole optimization process, and an intelligent optimization algorithm is used, which is improved and adaptively adjusted in combination with actual constraint conditions and objective function to solve the optimal source-net-load-storage configuration scheme; Step six: risk assessment is performed on the optimized planning scheme to identify the risk factors faced, and corresponding countermeasures are developed for different risk factors, including establishing a risk early warning mechanism, reserving emergency energy and signing long-term energy supply contracts to ensure safe and reliable operation of the system; Step seven: according to the optimized planning scheme, the construction and reconstruction of source-net-load-storage are implemented, the reconstruction work including the access of distributed power sources, the configuration of energy storage devices and the upgrading and reconstruction of power grids, and in the process of system operation, the operating state and performance indicators of the system are monitored in real time, the performance indicators including energy utilization rate, cost indicators and power supply reliability, and according to the monitoring results, the optimized planning scheme is dynamically adjusted and optimized to adapt to the changes of system operation environment and the development of user demand.

Images