Active power distribution network simulation calculation method based on one graph of power grid

Abstract

The invention discloses an active power distribution network simulation calculation method based on a power grid graph, and belongs to the technical field of power system simulation. The invention relates to an active power distribution network simulation calculation method based on a power grid graph. The method comprises the following steps: establishing a power network topology model, constructing a dynamic simulation model, monitoring and controlling in real time, and analyzing and optimizing a simulation result. The problems that an existing simulation calculation method often cannot meet the multi-scale and multi-time-scale dynamic simulation requirement, and real-time monitoring and control are lacked are solved. According to the method, multi-scale and multi-time-scale dynamic simulation of the active power distribution network is realized through the calculation steps of establishing a power network topology model, constructing a dynamic simulation model, carrying out real-time monitoring and control, and analyzing and optimizing a simulation result, and the operation state and performance of a power system are predicted. And the safety stability and the power supply reliability of the power grid are improved through real-time monitoring and control.

Description

Technical Field

[0001] The present invention relates to the technical field of power system simulation, and in particular to an active distribution network simulation calculation method based on a power grid graph. Background Art

[0002] With the access of distributed power sources (such as solar energy and wind energy) and the increase of new loads such as electric vehicle charging stations and energy storage systems, traditional passive distribution networks are transforming into active distribution networks. This transformation requires us to model and simulate the distribution network more accurately to ensure its safe, economical and efficient operation. However, existing simulation calculation methods often cannot meet the dynamic simulation requirements of multiple scales and time scales, especially in real-time monitoring and control. Summary of the invention

[0003] The purpose of the present invention is to provide an active distribution network simulation calculation method based on a power grid map, by establishing a power network topology model, constructing a dynamic simulation model, carrying out real-time monitoring and control, and analyzing and optimizing the calculation steps of the simulation results to realize multi-scale and multi-time scale dynamic simulation of the active distribution network, predict the operating status and performance of the power system, and improve the safety stability and power supply reliability of the power grid through real-time monitoring and control, thereby solving the problems raised in the above-mentioned background technology.

[0004] To achieve the above object, the present invention provides the following technical solutions:

[0005] The active distribution network simulation calculation method based on a power grid map includes the following steps:

[0006] S01. Establish a power network topology model: collect geographic information, equipment information and line parameters of the distribution network, use the node-branch model or node-injection model to describe the structure of the power network, and form a basic framework of a power grid map;

[0007] S02. Build dynamic simulation model: Design simulation algorithms that can handle multi-time scale processes simultaneously, adopt hierarchical approach, combine long-term trend analysis with short-term dynamic simulation, and build mathematical models of renewable energy and flexible loads;

[0008] S03, Real-time monitoring and control: Deploy smart sensors at key nodes in combination with IoT technology, develop fault detection algorithms, adjust generator output, adjust transformer tap positions and control switch status based on real-time monitoring results;

[0009] Get the normal operating data X0 in the historical database, and use its mean (mean(X0) and standard deviation std(X0) to analyze the historical data X old Standardize to obtain the standardized training data X;

[0010] For historical data X old The expression for standardization using the mean(X0) and standard deviation std(X0) of normal data X0 is:

[0011] X=(X old -mean(X0)) / std(X0) (1)

[0012] Where X=[X1,X2,...,X n ] T ∈R n×m , n is the number of samples, m is the number of variables;

[0013] Extract slow features S = [S1, S2, ..., S M ];

[0014] Extracting slow features from training data X involves the following steps:

[0015] (7) Obtain normal operating condition data X0 from the historical database;

[0016] (8) Calculate the whitening matrix: Perform singular value decomposition on the covariance matrix of the normal data X0: Get the whitening matrix Q = Λ -1 / 2 U T , the whitening transformation of X0 is:

[0017] (9) Find the transformation matrix: The first-order derivative of the whitened data B The covariance matrix of is subjected to singular value decomposition: The transformation matrix W = PΛ -1 / 2 U T ;

[0018] (10) All slow features of the training data X are calculated according to the following formula: S = WX, where S = [S1, S2, ..., S m ]∈R n×m ;

[0019] (11) Calculate the slowness Δ(X 0j ), according to the formula Determine the number of slow features M, where q = 0.1 represents the set {Δ(X j )} 0.1 upper quantile;

[0020] (12) Get the slow features S of the training data X = [S1, S2, ..., S M ]∈R n×M , where M < m, represents the number of main slow features, and n is the number of samples;

[0021] S04. Analysis and optimization of simulation results: Predict the operating status and performance of the power system based on simulation calculations, evaluate the safety stability and power supply reliability of the power grid, and optimize the operation strategy of the power grid based on simulation results.

[0022] Preferably, the method for establishing the power network topology model is specifically as follows:

[0023] Using a geographic information system (GIS) platform to collect the geographical location, equipment type, equipment number, line length, and line type of the distribution network;

[0024] Using the node-branch model in graph theory, the power network is abstracted into a collection of nodes and branches, where nodes represent substations, distribution transformers and user terminals, and branches represent transmission lines and distribution lines;

[0025] Determine the electrical parameters of each node and branch, including but not limited to resistance, reactance, admittance, rated capacity and maximum load, and consider the changing characteristics under different operating conditions;

[0026] The graphical conversion function module can quickly generate a graphical power topology model from the tabular distribution network parameters.

[0027] Preferably, the method for constructing the dynamic simulation model is specifically as follows:

[0028] Establish a multi-time scale simulation algorithm, including fast response at the second level and load change at the hour level, where fast response objects at the second level include but are not limited to short circuit faults and instantaneous fluctuations, and load change objects at the hour level include but are not limited to daily load curves and seasonal load changes;

[0029] Adopting a hierarchical approach, long-term trend analysis is combined with short-term dynamic simulation to form a multi-level simulation system.

[0030] Establish mathematical models for various types of renewable energy; Establish mathematical models for flexible loads including electric vehicles;

[0031] The control strategy and protection strategy of the power electronic inverter model of the distributed power generation are developed by rapid prototype controller, and then the control strategy is downloaded to the distributed power generation simulator.

[0032] Preferably, the real-time monitoring and control method is specifically:

[0033] Deploy smart sensors at key nodes, including but not limited to current transformers, voltage transformers, temperature sensors, and humidity sensors, for real-time data collection;

[0034] Develop fault detection algorithms, including but not limited to short circuit detection, ground fault detection and overload detection, to quickly locate the fault point and take measures to restore power supply when an abnormal situation occurs;

[0035] Adjust the output of the generator set according to the real-time monitoring results;

[0036] Adjust the transformer tap position to maintain the system voltage within the normal range;

[0037] Control switch status, perform load transfer and fault isolation;

[0038] The distributed power generation simulator is connected to the real-time digital simulation computing core via optical fiber communication to perform joint simulation of the active distribution network.

[0039] Preferably, the simulation result analysis and optimization method further includes:

[0040] Provide visualization tools to display simulation results in charts;

[0041] Use power quality index information to calculate user economic losses and the cost of power grid investment and construction;

[0042] Use non-dominated sorting genetic algorithm for multi-objective optimization and obtain Pareto optimal solution;

[0043] The non-dominated sorting genetic algorithm is optimized for multiple objectives using local search of external populations and internal and external populations, including the following four aspects:

[0044] (1) Setting up an external population, retaining the parents and offspring of the internal population, the individuals generated by the local search of the internal population, and the optimal individuals generated by the local search of the external population into the external population;

[0045] (2) Internal population local search: perform internal population local search on half of the individuals distributed in the internal population, and retain the best individuals generated by the internal population local search in the external population;

[0046] The new individuals generated by local search of the internal population are:

[0047]

[0048] In formula (3), r and is a random number uniformly distributed between 0 and 1. is the j-th element of the i-th individual distributed in the internal population; is the j-th element of the new individual generated by the i-th individual distributed in the internal population through the local search of the internal population; j = 1, 2, ..., D; D is the dimension of the decision vector; i = 1, 2, ..., P / 2; P is the population size of the internal population;

[0049] (3) External population local search: In the second half of the optimization process, based on the internal population local search, the external population local search operation is performed on half of the individuals distributed in the external population, and the optimal individuals generated by the external population local search are retained in the external population;

[0050] (4) Natural selection: The parent population of the internal population, the child population of the internal population, the population generated by the local search of the internal population, and the population generated by the local search of the external population are sorted according to their fitness, and the next generation population of the internal population is obtained through the survival of the fittest.

[0051] Preferably, the power grid map specifically includes:

[0052] Data import and export unit, used to import and export data in various formats and interact with external systems and external tools;

[0053] The query and statistics unit is used to provide a variety of query and statistics functions, including but not limited to equipment query, load query, fault query and historical data analysis, to quickly search and analyze power grid data and assist users in making decisions;

[0054] Multi-user access and permission management unit, used for multi-user access and permission management, so that users can only access data and functions authorized by the user;

[0055] System integration unit, used for data intercommunication and business collaboration between different systems;

[0056] Data verification unit, used to establish a data verification mechanism to prevent erroneous data from entering the system;

[0057] Operation interface unit, used to provide a user-friendly operation interface, supporting graphical modeling and parameter setting;

[0058] Simulation mode unit, used to support multiple simulation requirements and multi-scenario simulation analysis;

[0059] The simulation report unit is used to provide detailed simulation reports to help users fully understand the simulation results and support decision making.

[0060] Preferably, the calculation method further comprises the following steps:

[0061] Large-scale parallel computing and distributed processing through cloud platform deployment;

[0062] Provide API interface to support the integration and expansion of third-party applications;

[0063] Provide ongoing technical support and maintenance services;

[0064] By using big data analysis technology, the operating parameters of the active distribution network are imported into the artificial intelligence big data platform for analysis, the operating parameters are optimized, and the intelligence level of the system is improved.

[0065] Compared with the prior art, the present invention has the following beneficial effects:

[0066] 1. The present invention designs a multi-time scale simulation algorithm and combines it with a hierarchical method to simultaneously handle fast responses at the second level and load changes at the hour level. This method ensures the comprehensiveness and accuracy of the simulation model and can more realistically reflect the operating status of the active distribution network.

[0067] 2. The present invention realizes real-time data collection by deploying intelligent sensors at key nodes, develops fault detection algorithms that can quickly locate fault points and take effective measures to restore power supply, connects the distributed power supply simulator with the real-time digital simulation computing core through optical fiber communication, realizes joint simulation of active distribution networks, improves the system's fault handling capability and reliability, and ensures the safe and stable operation of the power grid.

[0068] 3. The present invention predicts the operating status and performance of the power system through simulation calculation, evaluates the safety stability and power supply reliability of the power grid, and uses a non-dominated sorting genetic algorithm for multi-objective optimization, thereby ensuring the optimality and feasibility of the planning and solving the problem that traditional simulation methods lack comprehensive analysis and optimization methods for simulation results. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] Figure 1 It is a flow chart of the simulation calculation method of the present invention;

[0070] Figure 2 It is a relationship diagram of simulation calculation data transmission of the present invention. DETAILED DESCRIPTION

[0071] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0072] In order to solve the problem that existing simulation calculation methods often cannot meet the needs of multi-scale and multi-time scale dynamic simulation and lack real-time monitoring and control, please refer to Figure 1-2 , this embodiment provides the following technical solutions:

[0073] The active distribution network simulation calculation method based on a power grid map includes the following steps:

[0074] S01. Establish a power network topology model: collect geographic information, equipment information and line parameters of the distribution network, use the node-branch model or node-injection model to describe the structure of the power network, and form a basic framework of a power grid map;

[0075] S02. Build dynamic simulation model: Design simulation algorithms that can handle multi-time scale processes simultaneously, adopt hierarchical approach, combine long-term trend analysis with short-term dynamic simulation, and build mathematical models of renewable energy and flexible loads;

[0076] S03, Real-time monitoring and control: Deploy smart sensors at key nodes in combination with IoT technology, develop fault detection algorithms, adjust generator output, adjust transformer tap positions and control switch status based on real-time monitoring results;

[0077] Get the normal operating data X0 in the historical database, and use its mean (mean(X0) and standard deviation std(X0) to analyze the historical data X old Standardize to obtain the standardized training data X;

[0078] For historical data X old The expression for standardization using the mean(X0) and standard deviation std(X0) of normal data X0 is:

[0079] X=(X old -mean(X0)) / std(X0) (1)

[0080] Where X=[X1,X2,...,X n ] T ∈R n×m , n is the number of samples, m is the number of variables;

[0081] Extract slow features S = [S1, S2, ..., S M ];

[0082] Extracting slow features from training data X involves the following steps:

[0083] (13) Obtain normal operating condition data X0 from the historical database;

[0084] (14) Calculate the whitening matrix: Perform singular value decomposition on the covariance matrix of the normal data X0: Get the whitening matrix Q = Λ -1 / 2 U T , the whitening transformation of X0 is:

[0085] (15) Find the transformation matrix: The first-order derivative of the whitened data B The covariance matrix of is subjected to singular value decomposition: The transformation matrix W = PΛ -1 / 2 U T ;

[0086] (16) All slow features of the training data X are calculated according to the following formula: S = WX, where S = [S1, S2, ..., S m ]∫R n×m ;

[0087] (17) Calculate the slowness Δ(X 0j ), according to the formula Determine the number of slow features M, where q = 0.1 represents the set {Δ(X j )} 0.1 upper quantile;

[0088] (18) Get the slow features S of the training data X = [S1, S2, ..., S M ]∈R n×M , where M < m, represents the number of main slow features, and n is the number of samples;

[0089] S04. Analysis and optimization of simulation results: Predict the operating status and performance of the power system based on simulation calculations, evaluate the safety stability and power supply reliability of the power grid, and optimize the operation strategy of the power grid based on simulation results.

[0090] The specific method for establishing the power network topology model is as follows:

[0091] Using a geographic information system (GIS) platform to collect the geographical location, equipment type, equipment number, line length, and line type of the distribution network;

[0092] Using the node-branch model in graph theory, the power network is abstracted into a collection of nodes and branches, where nodes represent substations, distribution transformers and user terminals, and branches represent transmission lines and distribution lines;

[0093] Determine the electrical parameters of each node and branch, including but not limited to resistance, reactance, admittance, rated capacity and maximum load, and consider the changing characteristics under different operating conditions;

[0094] The graphical conversion function module can quickly generate a graphical power topology model from the tabular distribution network parameters, reducing modeling time and labor costs, and establishing a comprehensive and accurate power network topology model to provide basic data for subsequent simulation calculations.

[0095] The construction method of the dynamic simulation model is as follows:

[0096] Establish a multi-time scale simulation algorithm, including fast response at the second level and load change at the hour level, where fast response objects at the second level include but are not limited to short circuit faults and instantaneous fluctuations, and load change objects at the hour level include but are not limited to daily load curves and seasonal load changes;

[0097] A hierarchical approach is adopted to combine long-term trend analysis with short-term dynamic simulation to form a multi-level simulation system. Long-term trend analysis includes but is not limited to annual load growth forecast and equipment aging analysis, and short-term dynamic simulation includes but is not limited to real-time power balance and frequency regulation.

[0098] Establish mathematical models for various types of renewable energy, including output power prediction models, and consider the impact of factors such as weather forecasts, light intensity, and wind speed; establish mathematical models for flexible loads including electric vehicles, including charging behavior prediction models, and consider the impact of factors such as user behavior habits and charging station distribution;

[0099] The control strategy and protection strategy of the power electronic inverter model for distributed power generation modeling are developed through rapid prototype controller development, and then the control strategy is downloaded to the distributed power generation simulator to ensure the accuracy and real-time performance of the simulation model.

[0100] The specific methods of real-time monitoring and control are:

[0101] Deploy smart sensors at key nodes, including but not limited to current transformers, voltage transformers, temperature sensors, and humidity sensors, for real-time data collection;

[0102] Develop fault detection algorithms, including but not limited to short circuit detection, ground fault detection and overload detection, to quickly locate the fault point and take measures to restore power supply when an abnormal situation occurs;

[0103] Adjust the output of the generator set according to the real-time monitoring results, including but not limited to adjusting the active power and reactive power of the generator;

[0104] Adjust transformer tap positions, including but not limited to manual and automatic adjustments, to maintain system voltage within normal range;

[0105] Control switch status, including but not limited to circuit breakers, disconnectors, load switches, etc., to transfer loads and isolate faults;

[0106] The distributed power simulation machine is connected to the real-time digital simulation computing core through optical fiber communication to conduct joint simulation of the active distribution network, thereby improving the real-time and authenticity of the simulation.

[0107] Ensure safe and stable operation of the power grid through real-time monitoring and control.

[0108] The simulation results analysis and optimization methods are as follows:

[0109] Predict the operating status and performance of the power system through simulation calculations, including but not limited to node voltage, branch current, system frequency, active power, reactive power, etc.;

[0110] Evaluate the safety stability and power supply reliability of the power grid, including but not limited to static safety analysis, transient stability analysis, voltage stability analysis, frequency stability analysis, etc.;

[0111] According to the simulation results, optimize the operation strategy of the power grid, including but not limited to load balancing, reactive power compensation, voltage regulation, frequency regulation, etc., to improve the economy and reliability of the system;

[0112] Provide visualization tools to display simulation results in charts and graphs, which is convenient for managers to make decisions;

[0113] Use power quality index information to calculate user economic losses and the cost of grid investment and construction to ensure the economic feasibility of planning;

[0114] The non-dominated sorting genetic algorithm is used for multi-objective optimization to obtain the Pareto optimal solution, ensuring the optimality and feasibility of the planning.

[0115] The non-dominated sorting genetic algorithm is optimized for multiple objectives using local search of external populations and internal and external populations, including the following four aspects:

[0116] (1) Setting up an external population, retaining the parents and offspring of the internal population, the individuals generated by the local search of the internal population, and the optimal individuals generated by the local search of the external population into the external population;

[0117] (2) Internal population local search: perform internal population local search on half of the individuals distributed in the internal population, and retain the best individuals generated by the internal population local search in the external population;

[0118] The new individuals generated by local search of the internal population are:

[0119]

[0120] In formula (3), r and is a random number uniformly distributed between 0 and 1. is the j-th element of the i-th individual distributed in the internal population; is the j-th element of the new individual generated by the i-th individual distributed in the internal population through the local search of the internal population; j = 1, 2, ..., D; D is the dimension of the decision vector; i = 1, 2, ..., P / 2; P is the population size of the internal population;

[0121] (3) External population local search: In the second half of the optimization process, based on the internal population local search, the external population local search operation is performed on half of the individuals distributed in the external population, and the optimal individuals generated by the external population local search are retained in the external population;

[0122] (4) Natural selection: The parent population of the internal population, the child population of the internal population, the population generated by the local search of the internal population, and the population generated by the local search of the external population are sorted according to their fitness, and the next generation population of the internal population is obtained through the survival of the fittest.

[0123] The detailed model construction process and calculation method are as follows:

[0124] S11. Data preprocessing

[0125] The data of several attributes are obtained from the active distribution network simulation database and the data is cleaned. The data anomalies and bad points are removed, and the cleaned data is standardized to ensure that the data is analyzed under the same dimension, thereby improving the efficiency and accuracy of subsequent algorithms.

[0126] S12. Model training and verification

[0127] Divide the historical distribution network parameters into training set and validation set;

[0128] Train the preset initial active distribution network control model according to the training set, obtain multiple active distribution network training models, define the training function and start training, including model training, verification and printing loss values;

[0129] Multiple active distribution network training models are verified according to the verification set, and an active distribution network control model is selected from the multiple active distribution network training models according to the verification results of each active distribution network training model. The best performing model is selected through the loss function and accuracy of the verification set.

[0130] S13, Real-time simulation calculation

[0131] The distributed power simulator, the grid-connected inverter controlled by the distributed power simulator and the power topology model are connected in sequence to obtain a digital simulation calculation model, and the distributed power simulator and the grid-connected inverter controlled by the RCP controller are connected to the power topology model through optical fibers, auxiliary boards, etc., to perform active distribution network joint simulation;

[0132] The digital simulation calculation model is used for simulation to obtain the operating parameters of the active distribution network. The electrical parameters of the distributed photovoltaic and distribution network connection points can be observed through the oscilloscope control in the configuration software.

[0133] S14. Control parameter optimization

[0134] The active distribution network operation parameters are input into the active distribution network control model created based on the historical distribution network parameters to obtain the current active distribution network control parameters;

[0135] The threshold range is determined according to the average value and standard deviation of the historical active distribution network control parameters, and the simulation parameters in the power topology model are corrected according to the comparison result between the current active distribution network control parameters and the threshold range.

[0136] S15. Simulation results analysis and optimization

[0137] Predict the operating status and performance of the power system through simulation calculations, and evaluate the safety stability and power supply reliability of the power grid;

[0138] The non-dominated sorting genetic algorithm is used for multi-objective optimization to obtain the Pareto optimal solution set. The optimization objectives include minimizing the economic loss of users and minimizing the investment cost of power grid construction.

[0139] Generate detailed simulation reports, including simulation process, simulation results, optimization suggestions, etc., to help users fully understand the simulation results and support decision making.

[0140] A power grid diagram specifically includes:

[0141] Data import and export unit, used to import and export data in various formats and interact with external systems and external tools;

[0142] The query and statistics unit is used to provide a variety of query and statistics functions, including but not limited to equipment query, load query, fault query and historical data analysis, to quickly search and analyze power grid data and assist users in making decisions;

[0143] Multi-user access and permission management unit, used for multi-user access and permission management. Users can only access data and functions authorized by the user, ensuring data security and privacy protection;

[0144] System integration unit, used for data intercommunication and business collaboration between different systems, to achieve data sharing and business collaboration;

[0145] Data verification unit, used to establish a data verification mechanism to prevent erroneous data from entering the system and ensure the accuracy and completeness of input data;

[0146] Operation interface unit, used to provide a user-friendly operation interface, supporting graphical modeling and parameter setting;

[0147] Simulation mode unit, used to support multiple simulation requirements and multi-scenario simulation analysis, including but not limited to steady-state simulation, transient simulation and dynamic simulation;

[0148] The simulation report unit is used to provide detailed simulation reports to help users fully understand the simulation results and support decision making.

[0149] The calculation method also includes the following steps:

[0150] Support cloud platform deployment for large-scale parallel computing and distributed processing;

[0151] Provide API interface to support the integration and expansion of third-party applications;

[0152] Provide continuous technical support and maintenance services to ensure stable operation and continuous optimization of the system;

[0153] Supports multi-language versions to meet the usage requirements of different regions and countries;

[0154] By using big data analysis technology, the operating parameters of the active distribution network are imported into the artificial intelligence big data platform for big data analysis, the operating parameters are optimized, and the intelligence level of the system is improved.

[0155] Working principle: By collecting geographic information, equipment information and line parameters of the distribution network, the node-branch model in graph theory is used to describe the structure of the power network, forming the basic framework of a power grid map; designing multi-time scale simulation algorithms, combining hierarchical methods, combining long-term trend analysis with short-term dynamic simulation, and establishing mathematical models for renewable energy such as photovoltaics and wind power and flexible loads such as electric vehicles; deploying intelligent sensors at key nodes to achieve real-time data collection. Developing fault detection algorithms, adjusting the output of generator sets, adjusting the position of transformer taps, and controlling the switch status according to real-time monitoring results to ensure the safe and stable operation of the power grid; predicting the operating status and performance of the power system through simulation calculations, and evaluating the safety, stability and power supply reliability of the power grid. Optimizing the operation strategy of the power grid based on the simulation results, providing detailed simulation reports, and supporting decision-making.

[0156] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0157] While the embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that many changes, modifications, substitutions and variations can be made to the embodiments without departing from the principles and spirit of the invention.

Claims

1. An active distribution network simulation calculation method based on a power grid map, characterized in that: The following steps are involved: S01. Establish a power network topology model: collect geographic information, equipment information and line parameters of the distribution network, use the node-branch model or node-injection model to describe the structure of the power network, and form a basic framework of a power grid map; S02. Build dynamic simulation model: Design simulation algorithms that can handle multiple time scales simultaneously, adopt hierarchical approach, combine long-term trend analysis with short-term dynamic simulation, and build mathematical models of renewable energy and flexible loads; S03, Real-time monitoring and control: Deploy smart sensors at key nodes in combination with IoT technology, develop fault detection algorithms, adjust generator output, adjust transformer tap positions and control switch status based on real-time monitoring results; Get the normal operating data X0 in the historical database, and use its mean (mean(X0) and standard deviation std(X0) to analyze the historical data X old Standardize to obtain the standardized training data X; For historical data X old The expression for standardization using the mean(X0) and standard deviation std(X0) of normal data X0 is: X=(X old -mean(X0)) / std(X0) (1) Where X=[X1,X2,...,X n ] T ∈R n×m , n is the number of samples, m is the number of variables; Extract slow features S = [S1, S2, ..., S M ]; Extracting slow features from training data X involves the following steps: (1) Obtain normal operating condition data X0 from the historical database; (2) Calculate the whitening matrix: Perform singular value decomposition on the covariance matrix of the normal data X0: Get the whitening matrix Q = Λ -1 / 2 U T , the whitening transformation of X0 is: (3) Find the transformation matrix: The first-order derivative of the whitened data B The covariance matrix of is subjected to singular value decomposition: The transformation matrix W = PΛ -1 / 2 U T ; (4) All slow features of training data X are calculated according to the following formula: S = WX, where S = [S1, S2, ..., S m ]∈R n×m ; (5) Calculate the slowness Δ(X 0j 0, according to the formula Determine the number of slow features M, where q = 0.1 represents the set {Δ(X j )} 0.1 upper quantile; (6) Obtain the slow features S of the training data X = [S1, S2, ..., S M ]∈R n×M , where M < m, represents the number of main slow features, and n is the number of samples; S04. Analysis and optimization of simulation results: Predict the operating status and performance of the power system based on simulation calculations, evaluate the safety stability and power supply reliability of the power grid, and optimize the operation strategy of the power grid based on simulation results.

2. The active distribution network simulation calculation method based on a power grid one map according to claim 1 is characterized in that: The method for establishing the power network topology model is specifically as follows: Use the GIS platform to collect the geographical location, equipment type, equipment number, line length and line type of the distribution network; Using the node-branch model in graph theory, the power network is abstracted into a collection of nodes and branches, where nodes represent substations, distribution transformers and user terminals, and branches represent transmission lines and distribution lines; Determine the electrical parameters of each node and branch, including but not limited to resistance, reactance, admittance, rated capacity and maximum load, and consider the changing characteristics under different operating conditions; The graphical conversion function module can quickly generate a graphical power topology model from the tabular distribution network parameters.

3. The active distribution network simulation calculation method based on a power grid one map according to claim 2 is characterized in that: The construction method of the dynamic simulation model is specifically as follows: Establish a multi-time scale simulation algorithm, including fast response at the second level and load change at the hour level, where fast response objects at the second level include but are not limited to short circuit faults and instantaneous fluctuations, and load change objects at the hour level include but are not limited to daily load curves and seasonal load changes; The state variables of the multi-time scale simulation algorithm are divided into short time scale variables x s and the long time scale variable x L ; According to the different time scales of variables, the multi-time scale simulation algorithm is divided into two parts, as shown in the following formula (2): Where: G LL (x L ) and G ss (x s ) are the influence of long- and short-time scale variables on themselves; G s (x L ) and G LS (x s ) are the couplings between long and short time scale variables, respectively. A hierarchical approach is adopted to combine long-term trend analysis with short-term dynamic simulation to form a multi-level simulation system.

4. The active distribution network simulation calculation method based on a power grid one map according to claim 3 is characterized in that: The method for constructing the dynamic simulation model specifically also includes: Establish mathematical models for various types of renewable energy; Establish mathematical models for flexible loads including electric vehicles; The control strategy and protection strategy of the power electronic inverter model of the distributed power source modeling are developed through rapid prototype controller development, and then the control strategy is downloaded to the distributed power source simulator.

5. The active distribution network simulation calculation method based on a power grid one map according to claim 4 is characterized in that: The real-time monitoring and control method is specifically as follows: Deploy smart sensors at key nodes, including but not limited to current transformers, voltage transformers, temperature sensors, and humidity sensors, for real-time data collection; Develop fault detection algorithms, including but not limited to short circuit detection, ground fault detection and overload detection. When an abnormal situation occurs, quickly locate the fault point and take measures to restore power supply.

6. The active distribution network simulation calculation method based on a power grid one map according to claim 5 is characterized in that: The real-time monitoring and control method specifically includes: Adjust the output of the generator set according to the real-time monitoring results; Adjust the transformer tap position to maintain the system voltage within the normal range; Control switch status, perform load transfer and fault isolation; The distributed power generation simulator is connected to the real-time digital simulation computing core via optical fiber communication to perform joint simulation of the active distribution network.

7. The active distribution network simulation calculation method based on a power grid one map according to claim 6 is characterized in that: The simulation result analysis and optimization method also includes: Provide visualization tools to display simulation results in charts; Use power quality index information to calculate user economic losses and the cost of power grid investment and construction; Use non-dominated sorting genetic algorithm for multi-objective optimization and obtain Pareto optimal solution; The non-dominated sorting genetic algorithm is optimized for multiple objectives using local search of external populations and internal and external populations, including the following four aspects: (1) Setting up an external population, retaining the parents and offspring of the internal population, the individuals generated by the local search of the internal population, and the optimal individuals generated by the local search of the external population into the external population; (2) Internal population local search: perform internal population local search on half of the individuals distributed in the internal population, and retain the best individuals generated by the internal population local search in the external population; The new individuals generated by local search of the internal population are: In formula (3), r and is a random number uniformly distributed between 0 and 1. is the j-th element of the i-th individual distributed in the internal population; is the j-th element of the new individual generated by the i-th individual distributed in the internal population through the local search of the internal population; j = 1, 2, ..., D; D is the dimension of the decision vector; i = 1, 2, ..., P / 2; P is the population size of the internal population; (3) External population local search: In the second half of the optimization process, based on the internal population local search, the external population local search operation is performed on half of the individuals distributed in the external population, and the optimal individuals generated by the external population local search are retained in the external population; (4) Natural selection: The parent population of the internal population, the child population of the internal population, the population generated by the local search of the internal population, and the population generated by the local search of the external population are sorted according to their fitness, and the next generation population of the internal population is obtained through the survival of the fittest.

8. The active distribution network simulation calculation method based on a power grid one map according to claim 7 is characterized in that: The power grid map specifically includes: Data import and export unit, used to import and export data in various formats and interact with external systems and external tools; The query and statistics unit is used to provide a variety of query and statistics functions, including but not limited to equipment query, load query, fault query and historical data analysis, to quickly search and analyze power grid data and assist users in making decisions; The multi-user access and permission management unit is used for multi-user access and permission management. Users can only access the data and functions authorized for the user.

9. The active distribution network simulation calculation method based on a power grid one map according to claim 8 is characterized in that: The power grid map specifically includes: System integration unit, used for data intercommunication and business collaboration between different systems; Data verification unit, used to establish a data verification mechanism to prevent erroneous data from entering the system; Operation interface unit, used to provide a user-friendly operation interface, supporting graphical modeling and parameter setting; Simulation mode unit, used to support multiple simulation requirements and multi-scenario simulation analysis; The simulation report unit is used to provide detailed simulation reports to help users fully understand the simulation results and support decision making.

10. The active distribution network simulation calculation method based on a power grid one map according to claim 9 is characterized in that: The calculation method further comprises the following steps: Large-scale parallel computing and distributed processing through cloud platform deployment; Provide API interface to support the integration and expansion of third-party applications; Provide ongoing technical support and maintenance services; Using big data analysis technology, the operating parameters of the active distribution network are imported into the artificial intelligence big data platform for analysis and optimization of the operating parameters.

Images