Gridding new energy distribution network optimization scheduling method, system, device and medium

Abstract

The invention discloses a grid new energy distribution network optimization scheduling method, system, device and medium, and the method comprises the steps: collecting electrical parameters and new energy data, and generating an initial grid topological graph; marking nodes of the initial grid topological graph and carrying out merging operation or isolation operation, generating a final network topological graph and calculating partition boundary parameters; constructing a five-dimensional objective function derivative set, and obtaining a Pareto optimal solution set; generating a distribution network optimization scheduling strategy; and calculating a four-dimensional deviation, generating a distribution network optimization scheduling execution report, updating the five-dimensional objective function derivative set, and generating a closed-loop optimization instruction. According to the method, through topology reconstruction, new energy output fluctuation is adapted, and the defect of static partition is overcome; a five-dimensional objective function derivative set is combined with a Pareto optimal solution, multi-objective collaborative optimization is achieved, and the problem of single economical efficiency emphasis is solved; and a closed-loop iteration mechanism continuously optimizes scheduling, so that global optimum operation of the distribution network is guaranteed, and adaptability and reliability are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent optimization scheduling of power systems, and in particular to a grid-based new energy distribution network optimization scheduling method, system, device and medium. BACKGROUND

[0002] Under the trend of global energy structure transformation towards clean and low-carbon, the penetration rate of new energy such as wind power and photovoltaic in distribution network continues to increase. Modern distribution network has changed from traditional passive radial network to active network with interaction of source, network, load and storage, and needs to consider multiple scheduling objectives such as economy, stability, environmental protection and efficiency. Grid-based distribution network optimization scheduling, as a core technology of intelligent transformation of power systems, can realize regulation and control of new energy output fluctuation through partition management, and provide basic support for safe and efficient operation of distribution network. At present, it has formed a standard technology chain of data acquisition, power flow calculation and optimization decision, and has become a key technology direction for coping with high penetration of new energy.

[0003] However, the existing grid-based new energy distribution network optimization scheduling technology has significant limitations. On the one hand, the topology partition method is static, and conventional technology relies on traditional clustering methods such as K-means to construct a fixed grid topology, which cannot dynamically adapt to the temporal and spatial fluctuation characteristics of new energy output, resulting in problems such as voltage out-of-limit and rapid increase of network loss in high penetration areas. On the other hand, the multi-objective coordination capability is insufficient. Existing multi-objective optimization algorithms mostly use linear programming to handle a single economic objective, and combine heuristic algorithms to meet safety constraints. The federal learning framework only aggregates node parameters through simple weighted averaging, which is difficult to accurately quantify the conflict degree between economic, stability and other objectives, and cannot achieve global optimal scheduling. SUMMARY

[0004] In view of the above existing problems, the present application is proposed.

[0005] Therefore, the present application provides a grid-based new energy distribution network optimization scheduling method, system, device and medium to solve the two core shortcomings of the existing grid-based new energy distribution network optimization scheduling technology: one is that the topology partition is static, relying on traditional clustering such as K-means to build a fixed grid, which is difficult to adapt to the temporal and spatial fluctuation of new energy output, and is prone to voltage out-of-limit and rapid increase of network loss in high penetration areas; the other is that the multi-objective coordination is weak, the algorithm focuses on a single economy, and the federal learning simply aggregates parameters by weighted averaging, which is difficult to quantify the conflict between objectives and difficult to achieve global optimization.

[0006] To solve the above technical problems, the present application provides the following technical solutions: In a first aspect, the present application provides a grid-based new energy distribution network optimization scheduling method, comprising: Collecting electrical parameters and new energy data, and standardizing the electrical parameters and new energy data to generate an initial grid topology map; Mark the nodes of the initial grid topology map, and perform a merging operation or an isolation operation on the marked nodes to generate a final network topology map, and calculate partition boundary parameters of the final network topology map; According to the final grid topology map and the partition boundary parameters, a five-dimensional objective function derivative set is constructed, and a Pareto optimal solution set is obtained through the five-dimensional objective function derivative set; The Pareto optimal solution is analyzed to generate a distribution network optimization scheduling strategy; The distribution network optimization scheduling strategy is executed, voltage deviation data and network loss change data are collected to calculate four-dimensional deviation, and a distribution network optimization scheduling execution report is generated; According to the distribution network optimization scheduling execution report, the five-dimensional objective function derivative set is updated to generate a closed-loop optimization instruction.

[0007] As a preferred scheme of the grid new energy distribution network optimization scheduling method, the step of generating an initial grid topology map comprises: The collected electrical parameters and new energy data are time-synchronized and abnormally cleaned to obtain a standardized data set; According to the standardized data set, an impedance matrix and a new energy density vector are analyzed and obtained; According to the impedance matrix and the new energy density vector, an initial grid topology map is generated through clustering and merging or splitting.

[0008] The beneficial effects of the preferred technical scheme are that: through time synchronization and abnormal cleaning, the timeliness and reliability of electrical parameters and new energy data are ensured, and a standardized data set is provided for subsequent topology generation; according to the clustering operation of the impedance matrix and the new energy density vector, the generated initial grid topology map can match the actual situation of the electrical connection and new energy distribution of the distribution network, and lay a foundation for subsequent dynamic topology reconstruction.

[0009] As a preferred scheme of the grid new energy distribution network optimization scheduling method, the step of generating a final grid topology map and calculating partition boundary parameters comprises: According to the initial grid topology map, the electrical distance and new energy density characteristics of the nodes are calculated; According to the electrical distance and new energy density characteristics, feature extraction and fusion are performed, and the nodes of the initial grid topology map are marked; According to the marked initial grid topology map, a merging and isolation operation is performed on the nodes to generate a final grid topology map; According to the final grid topology map, the partition boundary parameters are obtained through a boundary calculation algorithm.

[0010] The beneficial effects of the preferred technical scheme are that: the electrical distance and the new energy density characteristics are calculated according to the initial grid topology map, the nodes of the initial grid topology map are marked after extraction and fusion, the node merging and isolation operation is more suitable for the actual operation characteristics of the distribution network, and the generated final grid topology map adapts to the new energy fluctuation; the boundary calculation algorithm is combined to obtain the partition boundary parameters, the topology support is provided for the construction and global optimization of the derivative set of the five-dimensional target function, and the adaptability and reliability of the distribution network scheduling are improved.

[0011] As a preferred scheme of the grid-based new energy distribution network optimization scheduling method, the step of obtaining the Pareto optimal solution set comprises: According to the final grid topology map and the partition boundary parameters, the operation optimization parameters of each node in the final grid topology map are collected; The operation optimization parameters are fused and processed by the distributed parameter aggregation algorithm to generate a five-dimensional target function derivative set; The five-dimensional target function derivative set is solved by the multi-objective optimization algorithm to obtain the Pareto optimal solution set.

[0012] The beneficial effects of the preferred technical scheme are that: the operation optimization parameters of the nodes are collected in combination with the final grid topology map and the partition boundary parameters, the relevance of the operation optimization parameters and the topology is guaranteed; the data is fused by the distributed parameter aggregation algorithm to generate a five-dimensional target function derivative set, and the data security and aggregation effectiveness are considered; and then the Pareto optimal solution set is obtained by the multi-objective optimization algorithm, and global optimal solution support is provided for subsequent scheduling strategy generation.

[0013] As a preferred scheme of the grid-based new energy distribution network optimization scheduling method, the step of generating the distribution network optimization scheduling strategy comprises: The Pareto optimal solution set is encoded into an executable instruction set through a communication protocol, and the executable instruction set is distributed to a terminal for execution; Execution state data is collected, and the distribution network optimization scheduling strategy is generated according to the execution state data.

[0014] The beneficial effects of the preferred technical scheme are that: the Pareto optimal solution set is encoded into an executable instruction set through a communication protocol, and the executable instruction set is distributed to a terminal for execution; the execution state data of the terminal is collected synchronously, so that the generated distribution network optimization scheduling strategy can be adapted to the actual operation of the equipment, the strategy deviation from the execution scene is avoided, and the practicality and landing of the scheduling strategy are improved.

[0015] As a preferred scheme of the grid-based new energy distribution network optimization scheduling method, the step of generating the distribution network optimization scheduling execution report comprises: Real-time monitoring of voltage deviation data and network loss change data and four-dimensional deviation analysis are performed to obtain four-dimensional deviation analysis results. According to the four-dimensional deviation analysis results, a distribution network optimization scheduling execution report is generated.

[0016] The beneficial effects of the preferred technical solution are: real-time monitoring of voltage deviation and network loss change data to capture distribution network operation state fluctuations; four-dimensional deviation analysis of actual deviations to make the four-dimensional deviation analysis results more comprehensive; and the distribution network optimization scheduling execution report generated according to the four-dimensional deviation analysis results can provide a basis for subsequent generation of closed-loop optimization instructions to ensure the closed-loop effectiveness of scheduling optimization.

[0017] As a preferred scheme of the grid-based new energy distribution network optimization scheduling method, the step of generating a closed-loop optimization instruction and triggering a new distribution network optimization scheduling includes: A five-dimensional target function partial derivative set is generated by updating the five-dimensional target function derivative set through the distribution network optimization scheduling execution report; According to the five-dimensional target function partial derivative set, a closed-loop optimization instruction set is calculated and generated through a multi-objective optimization algorithm; According to the closed-loop optimization instruction set, a new distribution network optimization scheduling is triggered, and the closed-loop optimization instruction set is distributed to each regional execution terminal.

[0018] The beneficial effects of the preferred technical solution are: a five-dimensional target function partial derivative set is generated according to the distribution network optimization scheduling execution report to ensure that the target function fits the actual operation deviation of the distribution network; a closed-loop optimization instruction set is generated through a multi-objective optimization algorithm to ensure the scientificity and pertinence of the instructions; a new distribution network optimization scheduling is triggered, and the closed-loop optimization instruction set is distributed to the terminal to form a closed-loop iteration of execution, feedback, and optimization, adapt to the changes of the distribution network, and continuously improve the global optimality and stability of the scheduling.

[0019] In a second aspect, the present application provides a grid-based new energy distribution network optimization scheduling system, which includes: A data acquisition module is configured to acquire electrical parameters and new energy data and generate an initial grid topology map; A topology reconstruction module is configured to generate a final grid topology map and partition boundary parameters according to the initial grid topology map; An optimization calculation module is configured to aggregate node optimization parameters and generate a five-dimensional target function derivative set according to the final grid topology map and partition boundary parameters, and obtain a Pareto optimal solution set by using a multi-objective optimization algorithm; An instruction scheduling module is configured to generate a distribution network optimization scheduling strategy according to the Pareto optimal solution; A monitoring module is configured to generate a distribution network optimization scheduling execution report according to the distribution network optimization scheduling strategy; An iterative optimization module generates a five-dimensional target function partial derivative set according to the distribution network optimization scheduling execution report, and generates a closed-loop optimization instruction.

[0020] In a third aspect, the present application provides an electronic device, comprising: a memory and a processor; The memory is used to store computer executable instructions, and the processor is used to execute the computer executable instructions, which realize the steps of the grid new energy distribution network optimization scheduling method when executed by the processor.

[0021] In a fourth aspect, the present application provides a computer readable storage medium, which stores computer executable instructions, which realize the steps of the grid new energy distribution network optimization scheduling method when executed by the processor.

[0022] Compared with the prior art, the present application has the following beneficial effects: by topology reconstruction, i.e. generation of an initial grid topology map to a final grid topology map, adapting to new energy output fluctuation, solving the defect of static partition; with a five-dimensional target function derivative set combined with a Pareto optimal solution, realizing multi-objective collaborative optimization, making up for the single economic emphasis problem; closed-loop iterative mechanism continuously optimizes scheduling, guarantees the global optimal of distribution network operation, and improves the adaptability and reliability. BRIEF DESCRIPTION OF DRAWINGS

[0023] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiment description. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can be obtained without creative labor on the basis of these drawings.

[0024] Figure 1 The overall flowchart of the grid new energy distribution network optimization scheduling method of an embodiment of the present application is shown. DETAILED DESCRIPTION

[0025] In order to make the above-mentioned purposes, features and advantages of the present application more apparent and easy to understand, the specific embodiments of the present application will be described in detail below with reference to the drawings of the specification. Obviously, the described embodiments are only a part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor should be within the scope of protection of the present application.

[0026] Embodiment 1, refer to Figure 1 For an embodiment of the present application, a grid new energy distribution network optimization scheduling method is provided, comprising: S100, collect electrical parameters and new energy data, and standardize the electrical parameters and new energy data to generate an initial grid topology map.

[0027] S200, mark the nodes of the initial grid topology map, and perform a merging operation or an isolation operation on the marked nodes to generate a final network topology map, and calculate a partition boundary parameter of the final network topology map.

[0028] S300, construct a five-dimensional target function derivative set according to the final grid topology map and the partition boundary parameter, and obtain a Pareto optimal solution set through the five-dimensional target function derivative set.

[0029] S400, analyze the Pareto optimal solution to generate a distribution network optimization scheduling strategy.

[0030] S500, execute the distribution network optimization scheduling strategy, collect voltage deviation data and network loss change data to calculate four-dimensional deviation, and generate a distribution network optimization scheduling execution report.

[0031] S600, update the five-dimensional target function derivative set according to the distribution network optimization scheduling execution report to generate a closed-loop optimization instruction.

[0032] It should be noted that in the current new energy distribution network, the new energy output has temporal and spatial fluctuation characteristics, and the distribution network needs to consider economic efficiency, stability and other multi-objective scheduling requirements. The traditional method relies on static grid topology and is difficult to adapt to new energy fluctuations, which can easily cause voltage out-of-limit in high-penetration areas and rapid increase of network loss. At the same time, single-objective optimization algorithms are mostly used, which are difficult to quantify objective conflicts and cannot achieve global optimal scheduling, affecting the efficiency and stability of the distribution network.

[0033] Therefore, in view of the poor adaptability of the above-mentioned static topology and the weak multi-objective coordination, through the steps of the multi-objective optimization algorithm S100-S600, the standardized data is collected to generate an initial grid topology map as a foundation; then the nodes of the initial grid topology map are marked, and the final grid topology map is generated through the merging operation and the isolation operation of the multi-objective optimization algorithm, and the partition boundary parameter is calculated to improve the fitting degree of the topology and the actual operation; then a five-dimensional target function derivative set is constructed and a Pareto optimal solution set is solved to realize multi-objective coordination; then a distribution network optimization scheduling strategy is generated according to the Pareto optimal solution set, and a distribution network optimization scheduling strategy is executed to obtain a distribution network optimization scheduling execution report; finally, the five-dimensional target function derivative set is updated through the distribution network optimization scheduling execution report, a closed-loop instruction is generated, and technical support is provided for distribution network scheduling and stable operation.

[0034] Embodiment 2, refer to Figure 1 For an embodiment of the present application, based on the above-mentioned embodiment, a grid-based new energy distribution network optimization scheduling method is provided.

[0035] In the embodiment of the application, in the power distribution network of a suburban industrial park with a new energy penetration rate of 40%, 3 photovoltaic power stations, 2 wind power plants, and 12 user load nodes are taken as an application scenario, and the steps of generating an initial grid topology graph in A1-A3 in S100 are implemented as follows: A1, time synchronization and abnormal cleaning are performed on the collected electrical parameters and new energy data to obtain a standardized data set.

[0036] Specifically, the electrical parameter collection content includes the voltage, current, and power of the 12 load nodes, and the new energy data includes the new energy output and installed capacity of various new energy power stations such as photovoltaic and wind power. The electrical parameters and new energy data are unified to a second-level time stamp by using a time synchronization protocol, the voltage abnormal value is detected by using an abnormal detection algorithm, and the data greater than 10% of the nominal voltage is removed, and finally 1440 groups / day of standardized data sets are obtained.

[0037] A2, impedance matrix and new energy density vector are obtained by analyzing the standardized data set.

[0038] Specifically, according to the voltage parameters and current parameters in the standardized data set, the mutual impedance between the 12 load nodes and the new energy power stations is calculated by inverting the node admittance matrix, and a 15x15 impedance matrix is formed, The electrical coupling strength between nodes is quantified by using a normalized impedance distance formula, and the formula is as follows: ; Wherein, represents the electrical coupling strength between nodes and ; represents the mutual impedance between node and node ; represents the modulus operation of mutual impedance; According to the ratio of new energy installed capacity to regional total load, the spatial distribution difference is corrected by using a spatial distribution correction algorithm, and a new energy density vector of 15 nodes is generated, wherein the new energy density calculation formula is as follows: Wherein, represents the new energy density; represents the installed capacity of the region ; represents the total load of the region; represents the spatial distribution correction factor according to the Gaussian kernel density estimation, is a regional coordinate vector, which is used to quantify the geographical aggregation degree of new energy resources in the region; Gaussian kernel density estimation method; the new energy density vector element ranges from 0.2 to 0.8, and the higher the value, the greater the new energy penetration intensity of the node periphery.

[0039] A3, generating an initial grid topology map by clustering and merging or splitting the impedance matrix and the new energy density vector.

[0040] Specifically, the electrical similarity calculated by the clustering algorithm through the impedance matrix and the new energy distribution similarity calculated by the new energy density vector are used as double basis; The threshold of the electrical similarity after normalization is 0.82, and the threshold of the new energy distribution similarity is 0.75; the nodes with similar electrical similarity and new energy distribution similarity are merged, such as photovoltaic power station 1 and load nodes 1-3, which are merged into one grid unit because the mutual impedance The calculation result is The nodes with different electrical similarity and new energy distribution similarity are split, such as wind power station 2, which is split into an independent grid unit because the mutual impedance is greater than 2Ω, The value is small, and the electrical coupling is weak; finally, an initial grid topology map of 8 grid units is generated, and each unit contains 1-3 original nodes.

[0041] In an optional embodiment, data enhancement and robustness verification can also be added in step S100, and the steps are as follows: after step A1, supplement the abnormal new energy output under extreme weather, expand the standardized data set to cover normal and extreme scenarios; check the data robustness by using the 3σ criterion, and remove 32 abnormal samples with new energy output fluctuation greater than 30% / h, and the newly generated standardized data set can improve the accuracy of grid unit merging and splitting operations in subsequent clustering, and avoid topology deviation caused by extreme weather.

[0042] In another optional embodiment, the clustering threshold adjustment mechanism can also be optimized in step S100, and the steps are as follows: in step A3, instead of using a fixed threshold, the threshold of electrical similarity is adjusted according to the new energy penetration rate: when the new energy penetration rate is greater than 45%, the threshold of electrical similarity is lowered to 0.78 to enhance the adaptation to new energy fluctuations; when the new energy penetration rate is less than 35%, the threshold is raised to 0.85 to ensure the stability of the load side; taking the new energy penetration rate fluctuation of 38%-46% in a certain week as an example, the initial grid topology map generated after adjusting the threshold, compared with the fixed threshold scheme, reduces the number of voltage out-of-limit times, and lays a better foundation for subsequent topology reconstruction.

[0043] In the embodiments of the present application, the step S200 of generating the final grid topology map and calculating the partition boundary parameters includes the following steps B1-B4: B1、According to the initial grid topology, the electrical distance and new energy density characteristics of the nodes are calculated.

[0044] Specifically, according to the 8 grid units of the initial grid topology, the weighted electrical distance between the nodes in the unit is calculated, the weight is taken as the reciprocal of the mutual impedance value, and the coupling correlation of the low impedance node is highlighted: the average mutual impedance between the nodes in the unit of photovoltaic power station 1 is 0.45Ω, and the corresponding weighted electrical distance is 0.32Ω; the independent unit of wind power station 2 has a mutual impedance of 2.1Ω, and the weighted electrical distance is 2.1Ω; meanwhile, the new energy density characteristics of each unit are extracted, and according to the new energy density vector range of 0.2-0.8 in step A2, the average density of the unit is calculated: the photovoltaic unit is 0.68, the wind power unit is 0.52, and the pure load unit is 0.25; the density fluctuation rate is ±8% / h for the photovoltaic unit and ±15% / h for the wind power unit, which matches the real-time output fluctuation characteristics in the new energy data, forming an 8-row 3-column feature matrix, where the row corresponds to the unit, the column corresponds to the weighted electrical distance, the density average and the density fluctuation rate.

[0045] B2、According to the electrical distance and new energy density characteristics, the features are extracted and fused, and the nodes of the initial grid topology are marked.

[0046] Specifically, principal component analysis is used to fuse the electrical distance and new energy density characteristics to obtain comprehensive characteristic values in the range of 0-1.0; the marking rules are set: the units with comprehensive characteristic values ≥0.75 are marked as high-coupling high-permeability areas, such as the unit of photovoltaic power station 1, the comprehensive characteristic value is 0.82, the density average is 0.68, and the electrical distance is 0.32Ω; the comprehensive characteristic value ranges from 0.5 to 0.75, which is marked as a medium-coupling medium-permeability area, such as the units of load nodes 4-6, the comprehensive characteristic value is 0.63, the density average is 0.45, and the electrical distance is 0.85Ω; the comprehensive characteristic value ≤0.5 is marked as a low-coupling low-permeability area, such as the unit of wind power station 2, the comprehensive characteristic value is 0.41, the density average is 0.52, and the electrical distance is 2.1Ω, and the 8 units are respectively assigned with unique identification codes G1-G8.

[0047] B3、According to the marked initial grid topology, the nodes are merged and isolated to generate the final grid topology.

[0048] Specifically, the merging operation is performed on the units marked as high coupling and high permeability area: such as photovoltaic power station 1 unit, identification code G1, comprehensive characteristic value 0.82 and adjacent photovoltaic power station 2 unit, identification code G2, comprehensive characteristic value 0.78, because the difference between the comprehensive characteristic values of the two units is less than 0.05, which meets the node distribution density of the industrial park distribution network, and is merged into a new energy aggregation area A; the isolation operation is performed on the units marked as low coupling and low permeability area and mutual impedance > 2Ω: such as wind power station 2 unit, identification code G5, mutual impedance 2.1Ω and adjacent load node 12 unit, identification code G6, density average 0.25, because the electrical connection is weak and the permeability difference is large, the density difference is 0.27, the electrical isolation boundary is set to avoid the influence of wind power fluctuation on load stability. Finally, the final grid topology diagram of 6 grid units is generated, which is reduced by 2 compared with the initial 8 units.

[0049] B4、According to the final grid topology map, the boundary calculation algorithm is used to obtain the partition boundary parameters.

[0050] Specifically, according to the topology structure of the final 6 grid units, the network flow optimization theory is used to analyze the node connection relationship between the units, and the key transmission path and potential blocking point are identified: among the 3 connection lines between the new energy aggregation area A and the adjacent load area, 2 connection lines with transmission capacity ≥ 5MW are selected as the key path. The boundary calculation is carried out combined with multiple constraint parameters: the electrical connection strength, the reciprocal quantization of mutual impedance value, such as 0.32Ω corresponding to strength value 3.125; power transmission limit, calculated according to the line carrying capacity, such as 10kV line power transmission limit 8MW; operation safety margin, i.e. 15% redundancy is reserved, and the actual allowed power is calculated as .

[0051] The partition boundary is determined through collaborative calculation and iterative optimization, such as the electrical boundary, taking the S502, S503 circuit breakers on the key path as the isolation point, and the electrical separation position of the new energy aggregation area A and the external area is determined; the power boundary, the maximum allowed power exchange value of the new energy aggregation area A to the outside is set as 7MW, considering the limit value after safety margin; the topology boundary, the connection relationship matrix of each unit is output, 6×6 matrix, the matrix element value represents the connection strength between units, ranging from 0 to 1.0, among which the connection strength between the new energy aggregation area A and the load area is 0.85, and the connection strength between the new energy aggregation area A and the wind power station 2 unit is 0.12, and finally the partition boundary parameters are obtained.

[0052] In an optional embodiment, a feature time effectiveness correction mechanism can also be added in step S200, and the steps are: in B2 step, a time decay factor is introduced to the new energy density feature, the decay coefficient is 0.1 / 24h, and the weight of long-term static data is corrected; for example, the density feature value of a certain unit 3 days ago is 0.72, which is corrected to 0.72x(1-0.1x3)=0.504, avoiding the interference of historical data on the current label, so that the matching degree of the label result with the real-time new energy output fluctuation is improved.

[0053] In another optional embodiment, the boundary adjustment strategy can also be optimized in step S200, and the steps are: after B4 step, the update threshold of the boundary parameter is set, when the power exchange at a certain boundary exceeds the upper limit by 10% for 3 consecutive periods, the boundary recalculation is automatically triggered; for example, the boundary power of new energy aggregation area A continuously exceeds 9MW, the new boundary exchange power is recalculated to be 10MW, and the electrical distance weight of the adjacent unit is adjusted to 0.65, improving the adaptability of the topology to power fluctuation.

[0054] In the embodiment of the application, the step S300 of obtaining the Pareto optimal solution set includes the following steps C1-C3: C1, according to the final grid topology map and the partition boundary parameter, the running optimization parameters of each node in the final grid topology map are collected.

[0055] Specifically, according to the final grid topology map and the partition boundary parameter, such as the external power exchange of new energy aggregation area A ≤ 7MW, the electrical isolation point S502, etc., the running optimization parameters of each node are collected: the electrical state parameters, i.e. the voltage, current and power of 15 original nodes in 6 units, such as the voltage of photovoltaic power station 1 node in new energy aggregation area A 0.415kV, the output current 420A, the new energy output 2.8MW, the voltage of load node 1 0.408kV, the input current 350A, and the consumption 0.8MW; the new energy operation parameters, i.e. the real-time new energy output (2.8MW, 2.5MW, 2.6MW) and the new energy output fluctuation rate (±7% / h) of photovoltaic power station, and the real-time new energy output (3.2MW, 3.0MW) and the new energy output fluctuation rate (±14% / h) of wind power plant; the boundary constraint condition, i.e. the external power exchange of new energy aggregation area A 6.8MW (close to the upper limit of 7MW), the S502 breaker tripping current threshold 800A (the current is 450A at present), and the voltage deviation allowed range ±5% (the voltage deviation of new energy aggregation area A +3%), the running optimization parameters are obtained, and a running optimization parameter set containing 15 nodes and 6 types of parameters is formed.

[0056] C2, the running optimization parameters are fused and processed by a distributed parameter aggregation algorithm to generate a five-dimensional target function derivative set.

[0057] Specifically, the homomorphic encryption method is used to encrypt the operation optimization parameters of each grid unit, such as the encrypted parameter of new energy aggregation area A as ciphertext E(2.8, 420, 6.8), which is uploaded to the central server through a high-speed communication network; the central server decrypts and verifies the encrypted data, such as verifying that the power data of new energy aggregation area A and the matching degree of boundary parameters meet the requirements, and then dynamically allocating aggregation weights according to the data volume ratio, new energy aggregation area A data volume ratio 35%, wind power plant 2 unit data volume ratio 15%; The economic efficiency is quantified by the loss cost, the stability is quantified by the voltage deviation, the environmental protection is quantified by the carbon emission, the efficiency is quantified by the device utilization rate, and the standby is quantified by the callable standby capacity. Five dimensions, such as new energy aggregation area A loss 0.05 yuan / kWh, deviation +3%, carbon emission 0.3 kg per MW·h, photovoltaic utilization rate 93%, standby 0.2 MW. The change rate of each dimension is the derivative after calculating the weighted average, and a five-dimensional target function derivative set is generated. For example, the derivative of new energy aggregation area A is [0.002 yuan / (kWh·unit), 0.001% / unit, 0.005 kg / (MW·h·unit), 0.003% / unit, 0.004 MW / unit].

[0058] C3, solve the five-dimensional target function derivative set by a multi-objective optimization algorithm to obtain a Pareto optimal solution set.

[0059] Specifically, 100 groups of initial populations are generated by uniform sampling, each group corresponding to a set of scheduling parameters, such as new energy aggregation area A power distribution, wind power plant output adjustment, etc., to ensure that the solution space covers different optimization tendencies such as economic efficiency priority and stability priority; through non-dominated sorting and adaptive mechanism, the initial population is iteratively selected and screened, and 40 groups of non-dominated solutions are reserved, such as the solution of "loss cost 0.048 yuan / kWh, voltage deviation +2.5%"; 20 groups of elite solutions are selected, and the population diversity is enhanced by simulated binary crossover and polynomial mutation to generate 80 new solutions; combined with the constraint domination principle of voltage deviation ≤±5% and power exchange ≤ boundary value, 12 groups of invalid solutions with power exchange exceeding 7 MW are excluded, and finally a Pareto optimal solution set containing 25 groups of solutions is obtained from 120 groups of iterative solutions, each solution corresponding to the five-dimensional balance state of economic efficiency, stability, environmental protection, efficiency and standby, such as a group of solutions: loss 0.049 yuan / kWh, voltage deviation +2.8%, carbon emission 0.29 kg / (MW·h), device utilization rate 94%, standby 0.21 MW.

[0060] In an optional embodiment, a parameter updating mechanism can also be added in step S300, and the steps are as follows: after step C1, an updating period of the running optimization parameter is set, for example, updating once every 15 minutes, and when the new energy output fluctuation rate exceeds a threshold, for example, greater than ±10% / h, instant updating is triggered; for example, when the new energy output of the photovoltaic power station in the new energy aggregation area A suddenly changes from 2.8 MW to 2.2 MW, the new parameters such as voltage 0.410 kV and current 380 A are collected in real time, and steps C2-C3 are re-executed, so that the Pareto optimal solution set can be matched with the new energy fluctuation in time.

[0061] In another optional embodiment, the multi-objective weight adaptive adjustment can also be optimized in step S300, and the steps are as follows: when the five-dimensional objective function derivative set is generated in step C2, a scene recognition model is introduced to distinguish scenes such as peak load and large new energy generation, and the aggregation weight of the five-dimensional objective is adjusted; for example, in the large new energy generation scene, the environmental protection weight is increased and the economic weight is reduced, so that the generated Pareto optimal solution set is more suitable for the scene demand.

[0062] In the embodiments of the present application, the generation of the distribution network optimization scheduling strategy in step S400 includes the following steps D1-D2: D1, the Pareto optimal solution set is encoded into an executable instruction set through a communication protocol, and the executable instruction set is distributed to a terminal for execution.

[0063] Specifically, a structured coding protocol is used, and economic targets such as network loss 0.049 yuan / kWh are mapped to the operation code area as operation code 0x10; stability (voltage deviation +2.8%) and environmental protection (carbon emission 0.29 kg / (MW·h)) targets are packaged into the parameter area, and the parameter segment is [0x02, 0x1C], and a standard data frame is generated; Through a time division multiplexing channel of an optical fiber communication protocol, an executable instruction set composed of a standard data frame sequence containing power generation instructions, topology switching commands and energy storage scheduling parameters, such as output adjustment of the photovoltaic power station 1 to 2.7 MW, the S502 circuit breaker remains closed, and the standby energy storage of the new energy aggregation area A calls 0.15 MW, is distributed to each grid terminal device, such as the intelligent inverter of the photovoltaic power station 1, the converter of the wind farm 2, and the intelligent electric meter of the load node, etc. Each terminal device is equipped with a special analysis component, which decodes and executes the received standard data frame sequence in real time, such as the intelligent inverter of the photovoltaic power station 1 adjusting the output from 2.8 MW to 2.7 MW according to the instruction, synchronously collecting voltage, current and other data in the execution process, and feeding back the execution state in the same format of the standard data frame, such as 0xAA, 0x20, 0x03, 0x1E, 0x7A, 0xBB, wherein 0x20 is the feedback operation code.

[0064] D2, collect execution state data, and generate a distribution network optimization scheduling strategy according to the execution state data.

[0065] Specifically, each grid terminal device collects execution state data in real time during execution of the scheduling instruction, including voltage state, power response, and device operating parameters, such as photovoltaic power station 1 node voltage 0.412 kV, actual new energy output 2.7 MW, load node 1 voltage 0.406 kV, wind farm 2 actual new energy output 3.1 MW, smart inverter temperature 45℃, circuit breaker tripping coil state normal, etc., to form a complete execution state data set. The collected voltage, power and other data are subjected to feature extraction and standardization processing, and then a multi-objective strategy fusion algorithm is used to synchronously calculate economic targets, safety constraint targets and new energy consumption targets, such as actual network loss 0.051 yuan / kWh, voltage deviation +3.0%, photovoltaic consumption rate 92%, and wind power consumption rate 93%. A weight distribution mechanism is used to balance the conflicts between the targets, a fuzzy membership degree optimization ranking is performed according to the Pareto optimal solution set, and finally a distribution network optimization scheduling strategy is output, such as maintaining the output of photovoltaic power station 1 at 2.7 MW, the output of wind farm 2 at 3.1 MW, calling the standby energy storage 0.15 MW in new energy aggregation area A to suppress power fluctuation, and keeping the S502 circuit breaker closed to ensure power transmission.

[0066] In an optional embodiment, an instruction verification mechanism can also be added in step S400, and the steps are as follows: before the executable instruction set is distributed in step D1, the executable instruction set is subjected to multiple verifications, including instruction format verification such as verifying the legality of data frame length and operation code, and boundary constraint verification such as whether the power generation instruction is within the device limit, and the new energy output of photovoltaic power station 1 is 2.7 MW, which is within the 0-3 MW installed capacity range; if the verification fails, return to S300 to regenerate the Pareto optimal solution set and encode, to ensure the safety of instruction execution.

[0067] In another optional embodiment, the adaptive processing of execution state data can also be optimized in step S400, and the steps are as follows: when collecting execution state data in step D2, an adaptive algorithm is used to select data compression and encoding schemes to match the computing power, storage space and communication bandwidth characteristics of the terminal device; for example, for a load node smart meter with weak computing power, a lightweight encoding scheme is used to transmit key voltage and power data, so that different terminal devices can efficiently transmit and process execution state data, providing comprehensive data support for scheduling strategy generation.

[0068] In the embodiments of the present application, generating a distribution network optimization scheduling execution report in step S500 includes the following steps E1-E2: E1, real-time monitoring of voltage deviation and network loss change data and four-dimensional deviation analysis are performed to obtain four-dimensional deviation analysis results.

[0069] Specifically, the system monitors in real time the voltage deviation data of 15 original nodes within 6 grid units, such as the voltage deviation of node 1 of the photovoltaic power station being +3.2% and the voltage deviation of node 4 of the load being 2.5%, as well as the network loss change data, such as the network loss of 0.053 yuan / kWh in the new energy aggregation zone A and the network loss of 0.061 yuan / kWh in the load zone. The deviation values ​​for each of the four dimensions—economic efficiency, stability, environmental friendliness, and efficiency—are calculated. The formula for calculating economic deviation is: ; in, Indicates an economic deviation; This represents the weighting coefficient for power generation costs, typically ranging from 0.5 to 1.0. This represents the weighting coefficient for network loss rate, typically ranging from 0.2 to 0.5. Indicates the unit The coefficient of the quadratic term of fuel cost; Indicates the unit The coefficient for the primary term of fuel cost; Indicates the unit The fixed operating cost constant term; This represents the total active power loss across the entire network; Indicates the first The active power output of the Taiwanese generator set; This represents the quadratic term of the unit's output. This represents the total active power load demand of the entire network; This indicates the total number of generator sets participating in the dispatch.

[0070] For example, the grid loss in the new energy aggregation zone A is 0.053 yuan / kWh, and the total active power load demand of the entire grid is... Take photovoltaic power station 1 (unit) ), fuel cost secondary term One term Fixed costs Contribute your efforts Power generation cost weighting Network loss rate weight Substitute into the formula: .

[0071] The formula for calculating stability deviation is: ; in, Indicates stability deviation; This represents the number of sampling points within the statistical time window; express Measured voltage value at time; express Time frequency; This represents the voltage stability weighting coefficient, which typically ranges from 0.1 to 1.0. This represents the frequency stability weighting coefficient, which typically ranges from 0.1 to 1.0. Indicates the voltage reference value; Indicates the maximum permissible voltage deviation; Indicates the rated operating frequency.

[0072] For example, taking the sampling points of the statistical window. Voltage stability weight Frequency stability weights Voltage reference value Maximum allowable voltage deviation Rated frequency The sum of squares of the sampled voltage deviations was obtained. Total frequency deviation Substitute into the formula: .

[0073] The formula for calculating environmental deviation is: ; in, This indicates a deviation in environmental protection. This represents the weighting coefficient of the carbon market quota allocation model, typically ranging from 0.6 to 0.8. Indicates the first Carbon dioxide emission monitoring data of the unit; This indicates the mapping between administrative regions and power grid zones; This indicates the average carbon dioxide emission level across the entire network.

[0074] For example, let's set up an administrative and power grid zoning mapping. Carbon market weight Carbon emissions from Unit A in the new energy aggregation zone Units in the load area Average emissions across the entire network Total carbon emissions of the entire network Substitute into the formula: , .

[0075] The formula for calculating efficiency deviation is: ; in, Indicates efficiency bias; Indicates efficiency weight; Indicates equipment Load rate; Representation device actual response time of the device rated response time of the device total number of devices participating in the dispatch

[0076] devices participating in the dispatch efficiency weight smart inverter of photovoltaic power station 1, load rate actual response time rated response time remaining devices calculate Substitute the formula: Calculate to form a four-dimensional deviation analysis result set.

[0077] E2, according to the four-dimensional deviation analysis result, generate a distribution network optimization dispatch execution report.

[0078] Specifically, in the report, the deviation of each dimension is summarized, such as the overall network loss of the economic dimension is slightly higher than expected, the voltage deviation of the stability dimension is within the allowed range, the carbon emissions of the environmental protection dimension have a small increase, and the utilization rate of part of the new energy equipment of the efficiency dimension does not reach the optimal expectation; evaluate the execution effect of the distribution network optimization dispatch strategy, and point out that the distribution network optimization dispatch strategy performs well in ensuring voltage stability, but there is room for optimization in network loss control and equipment efficiency improvement; combined with the four-dimensional deviation analysis result, the improvement direction is proposed, such as in the future, for the new energy aggregation area A with slightly high network loss, the line flow distribution can be optimized; for the insufficient utilization rate of photovoltaic equipment, adjust the output dispatch instruction, etc., to generate a distribution network optimization dispatch execution report.

[0079] In an optional implementation, a deviation warning mechanism can also be added in step S500, and the steps are: in the four-dimensional deviation analysis process of E1 step, set the deviation warning threshold of each dimension, such as economic dimension deviation exceeds +0.005 yuan / kWh, stability dimension deviation exceeds-3%, etc.; when the deviation of a certain dimension reaches or exceeds the warning threshold, such as the network loss deviation of new energy aggregation area A reaches +0.006 yuan / kWh, a warning signal is triggered, reminding the operation and maintenance personnel to pay attention and investigate the cause in time.

[0080] In another optional implementation, the visualization presentation of the report can also be optimized in step S500, and the steps are: in the step of generating a distribution network optimization dispatch execution report in E2, the four-dimensional deviation data is intuitively displayed in the form of line chart, column chart, etc. by using a combination of charts, such as using a line chart to display the network loss deviation trend in different time periods, and using a column chart to compare the voltage deviation of each grid unit, so that the report is easier to read and more convenient for analysis and decision-making.

[0081] ​In the embodiments of the present application, generating the closed-loop optimization instruction and triggering the new distribution network optimization scheduling in step S600 includes the following steps F1-F3: F1, update the five-dimensional target function derivative set by the distribution network optimization scheduling execution report, and generate the five-dimensional target function partial derivative set.

[0082] Specifically, combined with the four-dimensional deviation analysis results in the execution report, such as economic network loss deviation +0.003 yuan / kWh, stability voltage deviation-1.8%, environmental protection carbon emission deviation +0.01 kg / (MW·h), efficiency utilization rate deviation-2%, update the five-dimensional target function derivative set, and map the deviation values of each dimension to the correction coefficients of the five-dimensional target function, such as economic deviation +0.003 yuan / kWh corresponding to correction coefficient 1.1, stability deviation-1.8% corresponding to correction coefficient 0.95; adjust the five-dimensional target function derivative set [0.002 yuan / (kWh·unit), 0.001% / unit, 0.005 kg / (MW·h·unit), 0.003% / unit, 0.004 MW / unit] according to the correction coefficients, such as economic derivative update to 0.002*1.1=0.0022 yuan / (kWh·unit), stability derivative update to 0.001*0.95=0.00095% / unit; integrate to generate the five-dimensional target function partial derivative set [0.0022 yuan / (kWh·unit), 0.00095% / unit, 0.0052 kg / (MW·h·unit), 0.0028% / unit, 0.0041 MW / unit].

[0083] F2, calculate and generate a closed-loop optimization instruction set according to the five-dimensional target function partial derivative set by a multi-objective optimization algorithm.

[0084] Specifically, with the five-dimensional target function partial derivative set as a constraint, 80 groups of initial scheduling solutions are generated, such as photovoltaic power station 1 output adjustment to 2.6 MW, wind farm 2 output maintenance to 3.1 MW, and new energy aggregation area A standby energy storage calling 0.18 MW; through non-dominated sorting, 3 groups of solutions that violate the boundary constraint are removed, such as 3 groups of power exchange exceeding 7 MW, combined with simulated binary crossover and polynomial mutation to enhance the diversity of solutions, and 25 groups of high-quality solutions are selected; the optimal solution considering each target is selected from the high-quality solutions, encoded into a closed-loop optimization instruction set according to a communication protocol, including power generation instruction, energy storage scheduling instruction, and topology control instruction, such as photovoltaic power station 1 output 2.6 MW, new energy aggregation area A energy storage discharge 0.18 MW, S502 circuit breaker continuously closed, to ensure that the closed-loop optimization instruction set can be directly parsed and executed by the terminal.

[0085] F3, trigger a new distribution network optimization scheduling according to the closed-loop optimization instruction set, and distribute the closed-loop optimization instruction set to each regional execution terminal.

[0086] Specifically, through the network central control system, the closed-loop optimization instruction set is taken as the input to trigger a new network optimization scheduling process, and the scheduling task account is updated synchronously, such as recording the task number, the triggering time, the involved grid unit and the like; the special transmission channel of the optical fiber communication network is used to distribute the closed-loop optimization instruction set to each execution terminal according to the grid unit partition, such as distributing the power generation instruction of the photovoltaic power station 1 to its intelligent inverter and distributing the energy storage scheduling instruction to the energy storage converter of the new energy aggregation area A; after each terminal receives the instruction, it feeds back the receiving state in real time, such as instruction receiving success, starting execution and the like, and adjusts the operating parameters according to the instruction requirements, such as the intelligent inverter of the photovoltaic power station 1 reduces the output from 2.7MW to 2.6MW, and the energy storage converter starts the discharge mode to output 0.18MW, forming a closed-loop scheduling cycle of execution, feedback and optimization.

[0087] In an optional embodiment, an instruction priority mechanism can also be added in step S600, and the steps are as follows: before the F3 step of distributing the instructions, the priority of the closed-loop optimization instruction is set according to the emergency degree of the network operation, such as setting the adjustment instruction of the voltage deviation exceeding the early warning as the highest priority and setting the regular output fine-tuning instruction as the ordinary priority; the instructions are transmitted in the order of priority to ensure that the emergency instructions are executed first and the safety of the network operation is ensured.

[0088] In another optional embodiment, the closed-loop cycle can also be adaptively adjusted in step S600, and the steps are as follows: the closed-loop optimization cycle is adjusted according to the change amplitude of the five-dimensional target function partial derivative in the F1 step, such as shortening the cycle from the regular 30min to 15min if the overall change exceeds 10%; if the overall change is less than 10%, the regular 30min cycle is maintained, so that the closed-loop scheduling is more in line with the actual operation state of the network.

[0089] In summary, the collected electrical parameters and new energy data are first synchronously cleaned, the impedance matrix and density vector are analyzed, and the initial grid topology graph is generated through clustering operation; then the final grid topology graph and partition boundary parameters are obtained through feature calculation, node marking and merging operation and isolation operation; then the operation optimization parameters are collected, the five-dimensional target function derivative set is fused to generate, the Pareto optimal solution is solved; then the encoding instruction is distributed for execution to generate the scheduling strategy and the execution report, and finally the closed-loop optimization instruction is generated. The optional optimization process is also provided to support the scheduling and stable operation of the network and to help the new energy consumption.

[0090] Embodiment 3, the above is a schematic scheme of the grid new energy distribution network optimization scheduling method. It should be noted that the technical scheme of the grid new energy distribution network optimization scheduling system belongs to the same concept as the technical scheme of the grid new energy distribution network optimization scheduling method described above. The technical scheme of the grid new energy distribution network optimization scheduling system in this embodiment is not described in detail. The details can be referred to the description of the technical scheme of the grid new energy distribution network optimization scheduling method described above.

[0091] The embodiment also provides a grid new energy distribution network optimization scheduling system, comprising: A data acquisition module is configured to acquire electrical parameters and new energy data, and generate an initial grid topology map. A topology reconstruction module is configured to generate a final grid topology map and partition boundary parameters according to the initial grid topology map. An optimization calculation module is configured to aggregate node optimization parameters and generate a five-dimensional target function derivative set according to the final grid topology map and partition boundary parameters, and obtain a Pareto optimal solution set by using a multi-objective optimization algorithm. An instruction scheduling module is configured to generate a distribution network optimization scheduling strategy according to the Pareto optimal solution. A monitoring module is configured to generate a distribution network optimization scheduling execution report according to the distribution network optimization scheduling strategy. An iterative optimization module is configured to update the five-dimensional target function derivative set according to the distribution network optimization scheduling execution report, and generate a closed-loop optimization instruction.

[0092] The embodiment also provides an electronic device suitable for grid new energy distribution network optimization scheduling, comprising a memory and a processor. The memory is configured to store computer executable instructions, and the processor is configured to execute the computer executable instructions to implement the grid new energy distribution network optimization scheduling method proposed in the above embodiment.

[0093] The embodiment also provides a storage medium having a computer program stored thereon. The program is executed by a processor to implement the grid new energy distribution network optimization scheduling method proposed in the above embodiment.

[0094] The storage medium proposed in the embodiment belongs to the same inventive concept as the grid new energy distribution network optimization scheduling method proposed in the above embodiment. The technical details not described in detail in the embodiment can be referred to the above embodiment, and the embodiment has the same beneficial effects as the above embodiment.

[0095] Those skilled in the art can clearly understand the present application by the above description of the embodiments, and the present application can be realized by software and necessary general hardware, and of course, can also be realized by hardware. Based on such understanding, the technical solutions of the present application or the part that contributes to the prior art can be embodied in the form of a software product, which can be stored in a computer readable storage medium, such as a floppy disk, a read-only memory (ROM), a random access memory (RAM), a FLASH, a hard disk, or an optical disc, and includes a number of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute the methods of various embodiments of the present application.

[0096] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present application but not limit the present application, and although the present application is described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application can be modified or replaced equivalently without departing from the spirit and scope of the present application, and all should be covered in the scope of the claims of the present application.

Claims

1. A method for grid new energy distribution network optimization scheduling, characterized in that, The method comprises the following steps: Collecting electrical parameters and new energy data, and standardizing the electrical parameters and new energy data to generate an initial grid topology map; Marking the nodes of the initial grid topology map, and performing a merging operation or an isolation operation on the marked nodes to generate a final network topology map, and calculating partition boundary parameters of the final network topology map; According to the final grid topology map and the partition boundary parameters, a five-dimensional target function derivative set is constructed, and a Pareto optimal solution set is obtained through the five-dimensional target function derivative set; Analyzing the Pareto optimal solution to generate a distribution network optimization scheduling strategy; Executing the distribution network optimization scheduling strategy, collecting voltage deviation data and network loss change data to calculate four-dimensional deviation, and generating a distribution network optimization scheduling execution report; According to the distribution network optimization scheduling execution report, updating the five-dimensional target function derivative set to generate a closed-loop optimization instruction. 2.The meshed new energy distribution network optimization scheduling method of claim 1, wherein, The step of generating an initial grid topology map comprises: Time synchronization and abnormal cleaning are performed on the collected electrical parameters and new energy data to obtain a standardized data set; According to the standardized data set, an impedance matrix and a new energy density vector are obtained; According to the impedance matrix and the new energy density vector, an initial grid topology map is generated through clustering and merging or splitting. 3.The meshed new energy distribution network optimization scheduling method of claim 2, wherein, The step of generating a final grid topology map and calculating partition boundary parameters comprises: According to the initial grid topology map, the electrical distance and new energy density characteristics of the nodes are calculated; According to the electrical distance and new energy density characteristics, feature extraction and fusion are performed, and the nodes of the initial grid topology map are marked; According to the marked initial grid topology map, a merging and isolation operation is performed on the nodes to generate a final grid topology map; According to the final grid topology map, the partition boundary parameters are obtained through a boundary calculation algorithm. 4.The meshed new energy distribution network optimization scheduling method of claim 3, wherein, The step of obtaining a Pareto optimal solution set comprises: According to the final grid topology map and the partition boundary parameters, the running optimization parameters of each node in the final grid topology map are collected; The running optimization parameters are fused and processed through a distributed parameter aggregation algorithm to generate a five-dimensional target function derivative set; The five-dimensional target function derivative set is solved through a multi-objective optimization algorithm to obtain a Pareto optimal solution set.

5. The method of claim 4, wherein, The step of generating a distribution network optimization scheduling strategy comprises: The Pareto optimal solution set is encoded into an executable instruction set through a communication protocol, and the executable instruction set is distributed to a terminal for execution; Execution state data is collected, and a distribution network optimization scheduling strategy is generated according to the execution state data. 6.The meshed new energy distribution network optimization scheduling method of claim 5, wherein, The step of generating a distribution network optimization scheduling execution report comprises: Real-time monitoring of voltage deviation data and network loss change data and four-dimensional deviation analysis are performed to obtain four-dimensional deviation analysis results; According to the four-dimensional deviation analysis results, a distribution network optimization scheduling execution report is generated.

7. The method of Claim 6, wherein, The step of generating a closed-loop optimization instruction and triggering a new distribution network optimization scheduling comprises: The five-dimensional target function derivative set is updated through the distribution network optimization scheduling execution report to generate a five-dimensional target function partial derivative set; According to the five-dimensional target function partial derivative set, a closed-loop optimization instruction set is generated through a multi-objective optimization algorithm; According to the closed-loop optimization instruction set, a new distribution network optimization scheduling is triggered, and the closed-loop optimization instruction set is distributed to each regional execution terminal.

8. A grid new energy distribution network optimization scheduling system, applying the method of any one of claims 1-7, characterized in that, The method comprises the following steps: A data acquisition module is configured to acquire electrical parameters and new energy data and generate an initial grid topology map; A topology reconstruction module is configured to generate a final grid topology map and partition boundary parameters according to the initial grid topology map; An optimization calculation module is configured to aggregate node optimization parameters and generate a five-dimensional target function derivative set according to the final grid topology map and partition boundary parameters, and obtain a Pareto optimal solution set by using a multi-objective optimization algorithm; An instruction scheduling module is configured to generate a distribution network optimization scheduling strategy according to the Pareto optimal solution set; A monitoring module is configured to generate a distribution network optimization scheduling execution report according to the distribution network optimization scheduling strategy; An iterative optimization module is configured to update the five-dimensional target function derivative set according to the distribution network optimization scheduling execution report and generate a closed-loop optimization instruction. 9.An electronic device, comprising: a memory and a processor; the memory is configured to store computer executable instructions, and the processor is configured to execute the computer executable instructions, and the computer executable instructions, when executed by the processor, implement the steps of the grid new energy distribution network optimization scheduling method of any one of claims 1 to 7. 10.A computer readable storage medium storing computer executable instructions, and the computer executable instructions, when executed by a processor, implement the steps of the grid new energy distribution network optimization scheduling method of any one of claims 1 to 7.

Images