Intelligent control method and system for series reactor for improving power quality

Abstract

The present invention discloses an intelligent control method and system for series reactors for improving power quality, which relates to the field of power optimization technology. The method includes: identifying system parameters and topological structure of the target distribution system; determining multiple distribution monitoring nodes, and building a distribution twin topological network; calling historical distribution records, and training the distribution twin topological network; collecting real-time monitoring data sets, and inputting them into the converged distribution twin network for data prediction of subsidiary branches; optimizing inductance parameters of multiple series reactors based on the monitoring data distribution map; and mapping and controlling the reactance distribution topology according to the inductance distribution control strategy. The method solves the technical problem that the existing series reactor control cannot monitor and accurately adjust the parameters of the reactor in real time, which makes it difficult to achieve rapid response and effective compensation for power quality problems, resulting in poor stability and reliability of power quality, and achieves the technical effect of improving the stability and reliability of power quality.

Description

Technical Field

[0001] The present application relates to the field related to power optimization technology, and specifically to an intelligent control method and system for series inductors for improving power quality. Background Art

[0002] In the current power system, power quality, as one of the core elements of power supply, directly affects the stable operation of the power grid and the user's power experience. With the rapid development of industrialization and informatization, the requirements for power quality are getting higher and higher. The influence of load changes and power supply fluctuations in the power system will lead to the emergence of power quality problems, such as voltage fluctuations and harmonic pollution. As an important power equipment, the series reactor is used to limit fault current or for load distribution in parallel circuits. It has many functions such as improving the reactive power operation of the power system, preventing the self-excited resonance of the generator, and reducing the transient overvoltage of the power frequency. The series reactor can adjust the current and smoothly supply power to the load, thereby reducing the amplitude and frequency of voltage fluctuations. However, traditional reactor control often relies on manual operation or simple automatic control systems, and usually can only achieve basic switch control, but cannot accurately adjust and monitor the parameters of the reactor in real time. With the continuous development of the power system, the requirements for power quality are also increasing. Problems such as voltage fluctuations and harmonic pollution are gradually highlighted, which brings challenges to the stable operation of power equipment and power grids. Traditional reactor control methods can no longer meet these new requirements.

[0003] Therefore, in the current series reactor control related technologies, there is a technical problem that the parameters of the reactor cannot be monitored and adjusted accurately in real time, which makes it difficult to achieve rapid response and effective compensation for power quality problems, resulting in poor stability and reliability of power quality. Summary of the invention

[0004] The present application provides an intelligent control method and system for series inductors for improving power quality, and adopts technical means such as building a twin network and optimizing inductance parameters to solve the technical problem that the existing series inductor control cannot monitor and accurately adjust the parameters of the reactor in real time, resulting in difficulty in achieving rapid response and effective compensation for power quality problems, resulting in poor stability and reliability of power quality, thereby achieving the technical effect of improving the stability and reliability of power quality.

[0005] The present application provides an intelligent control method for series inductors for improving power quality, the method comprising: identifying system parameters and topological structure of a target distribution system; determining multiple distribution monitoring nodes, and building a distribution twin topological network in combination with the system parameters and topological structure; calling historical distribution records, training the distribution twin topological network based on the historical distribution records, and generating a convergent distribution twin network; collecting real-time monitoring data sets of the multiple distribution monitoring nodes, and inputting the real-time monitoring data sets into the convergent distribution twin network for data prediction of subsidiary branches, and obtaining a monitoring data distribution map; based on the monitoring data distribution map, combined with the grid load state, the capacitor state and the distribution twin topological network, optimizing the inductance parameters of the multiple series inductors of the target distribution system, and generating an inductance distribution control strategy; building a reactance distribution topology based on the multiple series inductors, and mapping and controlling the reactance distribution topology according to the inductance distribution control strategy.

[0006] In a possible implementation method, multiple distribution monitoring nodes are determined, and a distribution twin topology network is built in combination with the system parameters and the topological structure, and the following processing is performed: based on the digital twin technology, the target distribution system is simulated and modeled in combination with the system parameters and the topological structure to construct an initial distribution twin topology network; multiple distribution monitoring nodes and multiple subsidiary branches are determined, wherein the subsidiary branches are branches that do not have distribution monitoring nodes; the multiple distribution monitoring nodes and the multiple subsidiary branches are embedded in the initial distribution twin topology network, and the positioning and identification are performed to generate the distribution twin topology network.

[0007] In a possible implementation, the distribution twin topology network is trained based on the historical distribution records to generate a converged distribution twin network, and the following processing is performed: information is extracted from the historical distribution records based on the multiple distribution monitoring nodes and the multiple affiliated branches to obtain multiple sample monitoring data sets and multiple sample data sets; based on the principles of machine learning, the multiple sample monitoring data sets are used as training data, and the multiple sample data sets are used as supervision data to perform supervised training and verification on the distribution twin topology network to obtain a converged distribution twin network that meets the expected training constraints.

[0008] In a possible implementation, before the real-time monitoring data set is input into the converged distribution twin network for data prediction of the subsidiary branches, the following processing is also performed: configuring a data preprocessing strategy, the data preprocessing strategy includes a data cleaning strategy and a Fourier transform filtering and denoising strategy; configuring a data preprocessing scheme based on the data preprocessing strategy, and integrating and building a data preprocessing model; preprocessing the real-time monitoring data set through the data preprocessing model to obtain a standard monitoring data set that is input into the converged distribution twin network for data prediction of the subsidiary branches.

[0009] In a possible implementation, based on the monitoring data distribution diagram, combined with the grid load state, the capacitor state and the distribution twin topology network, the inductance parameters of multiple series reactors of the target distribution system are optimized to generate an inductance distribution control strategy, and the following processing is also performed: the capacitor state is a capacitance parameter of the capacitor, wherein the capacitor and the series reactor correspond one to one, and there is a first mapping relationship between the capacitance parameter and the inductance parameter; the first mapping relationship expression is:

[0010]

[0011] Among them, L is the inductance parameter, f1 is the harmonic frequency, and C is the capacitance parameter; based on the monitoring data distribution diagram, the distribution twin topology network is assigned values ​​in combination with the grid load state and multiple capacitance parameters to generate a real-time distribution topology network; for the purpose of improving the power quality, the inductance parameters of multiple series reactors are optimized based on the power quality evaluation function and the real-time distribution topology network to generate the inductance distribution control strategy.

[0012] In a possible implementation, based on the power quality evaluation function and the real-time distribution topology network, the inductance parameters of multiple series reactors are optimized to generate the inductance distribution control strategy, and the following processing is also performed: multiple rated inductance parameter thresholds of the multiple series reactors are obtained; multiple inductance parameter combinations are randomly generated based on the multiple rated inductance parameter thresholds, and are set as multiple initial solutions; the real-time distribution topology network is simulated for harmonic control using the multiple initial solutions to generate multiple simulated harmonic control results, wherein the simulated harmonic control results include voltage stability coefficient, frequency stability coefficient, and harmonic content; the multiple simulated harmonic control results are evaluated using the power quality evaluation function to determine multiple quality Fitness; arranging the multiple initial solutions from large to small according to the quality fitness, determining N head solutions and M tail solutions based on the initial solution sequence, wherein N is less than M, and the sum of N and M is the number of initial solutions; randomly clustering the M tail solutions based on the N head solutions to generate N domains, and within the N domains, optimizing and iteratively updating the N domains according to a preset optimization strategy, wherein the preset optimization strategy includes a predetermined optimization step size and a predetermined update strategy, and the predetermined update strategy includes a taboo update strategy and an allowed update strategy; until the preset optimization number threshold is met, outputting the current N domains, and selecting the head solution of the domain with the largest comprehensive fitness among the current N domains to be set as the inductance distribution control strategy.

[0013] In a possible implementation, the power quality evaluation function is used to evaluate the multiple simulated harmonic control results, and the following processing is also performed: the expression of the power quality evaluation function is:

[0014]

[0015] Among them, F represents the adaptability of power quality, Q is the number of series reactors, and v i is the weight value of the ith series reactor, where the weight value is positively correlated with the circuit importance of the ith series reactor, w1 is the weight of the voltage stability coefficient, V ei is the voltage fluctuation variance of the circuit associated with the i-th series reactor, w2 is the weight of the frequency stability coefficient, Z ei is the frequency fluctuation variance of the circuit associated with the i-th series reactor, w3 is the weight of the harmonic content, X i is the harmonic content of the circuit associated with the i-th series reactor.

[0016] The present application also provides an intelligent control system for a series reactor for improving power quality, including:

[0017] A parameter and topology identification module, wherein the parameter and topology identification module is used to identify system parameters and topology of a target power distribution system;

[0018] A distribution twin topology network building module, which is used to determine multiple distribution monitoring nodes and build a distribution twin topology network in combination with the system parameters and topology structure;

[0019] A convergent distribution twin network generation module, which is used to call historical distribution records, train the distribution twin topology network based on the historical distribution records, and generate a convergent distribution twin network;

[0020] A monitoring data distribution map acquisition module, wherein the monitoring data distribution map acquisition module is used to collect real-time monitoring data sets of the multiple distribution monitoring nodes, and input the real-time monitoring data sets into the converged distribution twin network to perform data prediction of the subsidiary branches, and obtain a monitoring data distribution map;

[0021] An inductance parameter optimization module, which is used to optimize the inductance parameters of multiple series reactors of the target distribution system based on the monitoring data distribution diagram, combined with the grid load state, the capacitor state and the distribution twin topology network, and generate an inductance distribution control strategy;

[0022] A reactance distribution topology control module is used to build a reactance distribution topology based on the multiple series reactors, and to map and control the reactance distribution topology according to the inductance distribution control strategy.

[0023] The intelligent control method and system of series reactors for improving power quality proposed in this application is intended to identify the system parameters and topological structure of the target distribution system; determine multiple distribution monitoring nodes and build a distribution twin topological network; call historical distribution records to train the distribution twin topological network; collect real-time monitoring data sets and input them into the convergent distribution twin network for data prediction of subsidiary branches; optimize the inductance parameters of multiple series reactors based on the monitoring data distribution map and generate an inductance distribution control strategy; build a reactance distribution topology based on multiple series reactors, and map and control the reactance distribution topology according to the inductance distribution control strategy. It solves the technical problem that the existing series reactor control cannot monitor and accurately adjust the parameters of the reactor in real time, which makes it difficult to achieve rapid response and effective compensation for power quality problems, resulting in poor stability and reliability of power quality, realizes intelligent control of series reactors, and achieves the technical effect of improving the stability and reliability of power quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the technical solution of the embodiment of the present disclosure, the accompanying drawings of the embodiment of the present disclosure will be briefly introduced below. A flow chart is used in the present application to illustrate the operations performed by the system according to the embodiment of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps can be processed in reverse order or simultaneously as needed. At the same time, other operations can also be added to these processes, or one or more operations can be removed from these processes.

[0025] Figure 1 A flow chart of an intelligent control method for a series reactor for improving power quality provided in an embodiment of the present application;

[0026] Figure 2 A schematic diagram of the structure of an intelligent control system for series inductors for improving power quality provided in an embodiment of the present application.

[0027] Explanation of the accompanying drawings: parameter and topology structure identification module 10, distribution twin topology network construction module 20, convergent distribution twin network generation module 30, monitoring data distribution map acquisition module 40, inductance parameter optimization module 50, reactance distribution topology control module 60. DETAILED DESCRIPTION

[0028] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.

[0029] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0030] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict, and the terms "first\second" involved are merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by technicians in the technical field of this application. The terms used herein are for the purpose of describing the embodiments of the present application only.

[0031] The present application provides an intelligent control method for a series reactor for improving power quality, such as Figure 1 As shown, the method includes:

[0032] Step S100, identify the system parameters and topology of the target distribution system. The system parameters of the target distribution system refer to some key parameter characteristics of the distribution system, mainly including impedance, capacitance, inductance, etc.; the topology refers to the connection method and layout between various devices and components in the distribution system. The topology of the distribution network includes overhead lines, poles, cables, distribution transformers, switchgear, reactive compensation capacitors and other distribution equipment and ancillary facilities. The equipment and components are combined together through different connection methods to form different power network structures. Common power topologies include tree topology, ring topology and mesh topology, etc.; specifically, impedance is the complex representation of the ratio of voltage to current in a circuit, which reflects The capacitance is a physical quantity that measures the ability of a capacitor to store electric charge. It is composed of two conductive plates and an insulating medium between them. It can store electric charge and generate current in the circuit. The value of the capacitance determines the amount of charge stored in the capacitor under a given voltage. The larger the capacitance, the greater the amount of charge stored. The inductance is a physical quantity that measures the ability of a coil to generate a magnetic field. When current passes through the coil, it generates a magnetic field, and changes in the magnetic field generate an induced electromotive force, which hinders changes in current. The value of the inductance depends on the number of turns, length, and cross-sectional area of ​​the coil, as well as the medium surrounding the coil. The larger the inductance, the greater the ability to hinder current changes.

[0033] Step S200, determine multiple distribution monitoring nodes, and build a distribution twin topology network in combination with the system parameters and topology. Determine multiple distribution monitoring nodes, and build a distribution twin topology network in combination with system parameters (such as impedance, capacitance, inductance) and topology, aiming to simulate and optimize the operation of the actual distribution system by digital means to improve the stability, reliability and efficiency of the system. Specifically, select key locations in the distribution system to set monitoring nodes, which can be substations, important load points, line branch points, etc., to collect real-time voltage, current, power and other operating data. For each monitoring node, collect the electrical parameters of its location, such as impedance, capacitance, inductance, etc. By analyzing the topology of the distribution system, understand the connection mode and position relationship between the components, and build a twin topology network corresponding to the actual distribution system based on the collected system parameters and topology information, using advanced modeling and simulation technology, which can reflect the actual system operation status in real time.

[0034] In a possible implementation, step S200 further includes step S210, based on digital twin technology, combining the system parameters and topology structure to simulate and model the target distribution system, and construct an initial distribution twin topology network. The virtual model obtained by simulation modeling is the initial distribution twin topology network, which contains all the key information and data of the distribution system and can fully reflect the operating status and performance of the actual system. It also includes step S220, determining multiple distribution monitoring nodes and multiple auxiliary branches, wherein the auxiliary branch is a branch without a distribution monitoring node. A distribution monitoring node refers to a device or apparatus installed at a specific location in the distribution system for real-time monitoring and collection of system operating status data; an auxiliary branch refers to a branch in the distribution system that does not have a distribution monitoring node. In the distribution network, current and voltage are distributed to each user or device through transformers, switchgear and lines at various levels. The auxiliary branches are part of these distribution paths, but due to certain reasons (such as cost, technical limitations or operating requirements), distribution monitoring nodes are not installed on these branches. It also includes step S230, embedding the multiple distribution monitoring nodes and multiple subsidiary branches into the initial distribution twin topology network, and positioning and identifying them to generate the distribution twin topology network. The multiple distribution monitoring nodes and subsidiary branches in the actual distribution system are embedded into the distribution twin topology network. Each distribution monitoring node and subsidiary branch needs to be accurately positioned and identified. Specifically, they are assigned unique identifiers (such as ID numbers) and their positions in the network are clearly defined to generate the final distribution twin topology network, which not only includes all the key components and branches of the distribution system, but also accurately reflects the connection relationship and electrical characteristics between them through virtual nodes and lines.

[0035] Step S300, calling historical power distribution records, training the power distribution twin topology network based on the historical power distribution records, and generating a converged power distribution twin network. Call historical distribution records, and train the distribution twin topology network based on these records to generate a converged distribution twin network. Specifically, the distribution twin topology network is trained based on historical distribution records. The distribution twin topology network will continuously adjust its internal parameters and structure according to the input historical distribution records and the corresponding output results (usually the predicted values ​​of system status or performance indicators) to minimize the error between the predicted value and the actual value. When the network's prediction ability for training data reaches a certain accuracy and its generalization ability for new data is also good, the network is considered to have converged, that is, a converged distribution twin network is generated to estimate the harmonic voltages and harmonic currents of unknown branches. The generation and distribution of harmonic currents are affected by various factors in the power system, such as system parameters such as impedance, capacitance, inductance, and grid structure, which may cause the propagation and distribution of harmonic currents in the system to become complicated, thereby increasing the difficulty of measurement. Among them, historical distribution records refer to the operating data of the distribution system over a period of time in the past, which may include voltage, current, power factor, fault records, equipment status, etc.

[0036] In a possible implementation, step S300 further includes step S310, extracting information from the historical distribution records based on the multiple distribution monitoring nodes and the multiple subsidiary branches, and obtaining multiple sample monitoring data sets and multiple sample data sets. The historical distribution records contain the operating status data of the distribution system in the past period of time, and record the historical operation of the distribution system. They can be real-time data from the distribution monitoring nodes, or statistical data after processing and analysis; extract data related to the distribution monitoring nodes and subsidiary branches from the historical distribution records, and obtain multiple sample monitoring data sets and multiple sample data sets, wherein the sample monitoring data set is mainly the real-time monitoring data of the distribution monitoring nodes, including data of one or more monitoring nodes, such as voltage, current and harmonic data, as well as metadata such as timestamps and device identifications of these data; the sample data set refers to a sample data set related to the distribution system obtained from other sources (such as equipment logs, fault records, etc.), which contains richer information, such as the operation history of the equipment, maintenance records, fault diagnosis results, etc. It also includes step S320, which, based on the principle of machine learning, uses the multiple sample monitoring data sets as training data and the multiple sample data sets as supervision data to perform supervised training and verification on the distribution twin topology network to obtain a converged distribution twin network that meets the expected training constraints.

[0037] Step S400, collect the real-time monitoring data sets of the multiple distribution monitoring nodes, and input the real-time monitoring data sets into the convergent distribution twin network for data prediction of the subsidiary branches, and obtain a monitoring data distribution map. Collect the real-time monitoring data sets of multiple distribution monitoring nodes, and input these data into the trained convergent distribution twin network for data prediction to obtain a monitoring data distribution map. Specifically, multiple monitoring nodes are deployed at key positions in the distribution system to collect data such as voltage, current and harmonics in real time, forming a real-time monitoring data set. The real-time monitoring data set is input into the trained convergent distribution twin network. The convergent distribution twin network uses the input real-time monitoring data to predict the data of the subsidiary branches in the distribution system, including the changing trends and possible problems of parameters such as voltage and current. Based on the prediction results of the network, a monitoring data distribution map can be generated to intuitively display the data distribution of each monitoring point in the distribution system, including the fluctuation range and changing trends of parameters such as voltage and current.

[0038] In a possible implementation, step S400 further includes step S410, configuring a data preprocessing strategy, wherein the data preprocessing strategy includes a data cleaning strategy and a Fourier transform filtering and denoising strategy. Data cleaning is the core step of data preprocessing, and its purpose is to identify and correct errors, anomalies or missing values ​​in a data set; Fourier transform is a widely used technology in signal processing, which can convert signals from the time domain to the frequency domain for analysis. In data preprocessing, Fourier transform can be used for filtering and denoising. It also includes step S420, configuring a data preprocessing scheme based on the data preprocessing strategy, and integrating and building a data preprocessing model. Based on the data preprocessing strategy, configure the data preprocessing scheme, and integrate and build the data preprocessing model. Specifically, after clarifying the data preprocessing strategy, configure the data preprocessing scheme according to the specific data set and analysis requirements, including determining the specific operations, parameter settings and sequence of each step. For example, when dealing with missing values, you may need to choose deletion, replacement or interpolation, and determine the corresponding thresholds and fill values; when integrating data, you need to consider how to deal with homonymy, synonymy and data redundancy; integrating and building a data preprocessing model is to integrate the above strategies and schemes into a unified model to automatically perform data preprocessing tasks. It also includes step S430, preprocessing the real-time monitoring data set through the data preprocessing model, obtaining a standard monitoring data set and inputting it into the converged distribution twin network for data prediction of the subsidiary branches.

[0039] Step S500, based on the monitoring data distribution diagram, combined with the grid load state, capacitor state and the distribution twin topology network, the inductance parameters of the multiple series reactors of the target distribution system are optimized to generate an inductance distribution control strategy. Based on the monitoring data distribution diagram, combined with the grid load state, capacitor state and the distribution twin topology network, the inductance parameters of the multiple series reactors of the target distribution system are optimized to generate an inductance distribution control strategy. Specifically, the monitoring data distribution diagram generated from the real-time monitoring data set collected from multiple distribution monitoring nodes is deeply analyzed. At the same time, the load state of the grid, including load fluctuations, load types, etc., is considered. It is also necessary to evaluate the working state of the capacitor, including its capacity, voltage, temperature and other parameters. The distribution twin topology network is used to simulate and predict the operating state of the system, and then the inductance parameters of the multiple series reactors are optimized, that is, the inductance parameters are adjusted to find the optimal combination of inductance parameters, and the corresponding inductance distribution control strategy is generated to guide how to adjust the inductance value of each series reactor to achieve the optimized operation of the grid.

[0040] In a possible implementation, step S500 further includes step S510, wherein the capacitor state is a capacitance parameter of the capacitor, wherein the capacitor and the series inductor correspond one to one, and there is a first mapping relationship between the capacitance parameter and the inductance parameter. In certain specific circuit configurations, the capacitor and the series inductor are designed to have a one-to-one correspondence, meaning that each capacitor may be associated with a specific series inductor, or they may be used in pairs in some way, and the main parameter of the capacitor is capacitance, which indicates the amount of charge that the capacitor can store; the main parameter of the inductor is inductance, which indicates the resistance of the inductor to current changes. Also included is step S520, wherein the first mapping relationship expression is:

[0041]

[0042] Among them, L is the inductance parameter, f1 is the harmonic frequency, and C is the capacitance parameter. It also includes step S530, based on the monitoring data distribution map, the distribution twin topology network is assigned in combination with the grid load state and multiple capacitance parameters to generate a real-time distribution topology network. Map information such as real-time operating data, load status and capacitance parameters to the corresponding nodes and branches of the twin topology network, that is, process and analyze massive data, extract valuable information, and convert it into node attributes and branch parameters in the topology network, and finally generate a real-time distribution topology network. It also includes step S540, for the purpose of improving the power quality, based on the power quality evaluation function and the real-time distribution topology network, the inductance parameters of multiple series reactors are optimized to generate the inductance distribution control strategy. According to the information of the power quality evaluation function and the real-time distribution topology network, the inductance parameters of multiple series reactors are optimized and adjusted to obtain a set of optimal inductance parameter combinations, and finally an inductance distribution control strategy is generated, including the target inductance value, adjustment step, adjustment time and other parameters of each series reactor, which are used to guide the adjustment operation of the series reactor in the actual power system. Specifically, the power quality evaluation function is a mathematical function used to evaluate the power quality of the power system. It is constructed based on power quality indicators (such as voltage fluctuation, harmonic content, three-phase imbalance, etc.) to quantitatively describe the quality of power. The real-time distribution topology network is the real-time state of the distribution network in the power system at a certain moment. The series reactor is an important device in the power system for suppressing harmonics, regulating voltage fluctuations, etc., and the inductance parameters directly affect its ability to suppress harmonics and regulate voltage.

[0043] In a possible implementation, step S540 further includes step S541, obtaining multiple rated inductance parameter thresholds of the multiple series reactors. The rated inductance parameter threshold is the rated inductance value of the series reactor. It also includes step S542, randomly generating multiple inductance parameter combinations based on the multiple rated inductance parameter thresholds, and setting them as multiple initial solutions. It also includes step S543, using the multiple initial solutions to simulate harmonic control of the real-time distribution topology network, and generating multiple simulated harmonic control results, wherein the simulated harmonic control results include voltage stability coefficient, frequency stability coefficient, and harmonic content. Based on the real-time distribution topology network and the selected initial solution, a simulation analysis of harmonic control is performed, the power system is mathematically modeled, and simulation software is used to simulate the operation of the system under given parameters, evaluate the harmonic level of the system under different conditions, and the impact of these harmonics on system stability and power quality. The simulated harmonic control results include voltage stability coefficient, frequency stability coefficient, and harmonic content. Specifically, the initial solution represents different combinations of series reactor inductance parameters. The voltage stability coefficient is an indicator for quantitatively evaluating the stability of the power supply voltage. It is usually calculated by comparing the fluctuation amplitude of the power supply voltage when the load changes with the rated voltage value. The smaller the voltage stability coefficient, the more stable the power supply voltage; the frequency stability coefficient is an indicator for evaluating the frequency stability of the system, reflecting the fluctuation of the system frequency in a long or short period of time; the harmonic content is usually expressed as the proportion of harmonic voltage or harmonic current to fundamental voltage or fundamental current. The higher the harmonic content, the more serious the harmonic pollution in the system. It also includes step S544, using the power quality evaluation function to evaluate the multiple simulated harmonic control results and determine multiple quality fitness. In the power system, in order to measure the degree of improvement of power quality by different harmonic governance strategies or parameter adjustment schemes, a predefined power quality evaluation function is used to quantitatively evaluate various simulated harmonic governance results. Specifically, by comparing the quality fitness of different simulated harmonic governance results, the degree of improvement of power quality by various harmonic governance strategies or parameter adjustment schemes can be evaluated. The higher the quality fitness, the better the improvement effect of the scheme on power quality. It also includes step S545, arranging the multiple initial solutions from large to small according to the quality fitness, and determining N head solutions and M tail solutions based on the initial solution sequence, wherein N is less than M, and the sum of N and M is the number of initial solutions. It also includes step S546, randomly clustering the M tail solutions based on the N head solutions to generate N domains, and optimizing and iteratively updating the N domains according to the preset optimization strategy in the N domains, wherein the preset optimization strategy includes a predetermined optimization step size and a predetermined update strategy, and the predetermined update strategy includes a taboo update strategy and an allowed update strategy. If the updated tail solution does not meet the corresponding rated inductance parameter threshold, the tail solution will not be updated; if the fitness of the updated tail solution is greater than the fitness of the head solution in the same field, a replacement update will be performed.The method further includes step S547, until the preset optimization number threshold is met, the current N fields are output, and the head solution of the field with the largest comprehensive fitness among the current N fields is selected and set as the inductance distribution control strategy. After reaching the preset optimization number threshold, the algorithm stops iterating and outputs the evaluation results of the current N fields. From these fields, the algorithm selects the field with the largest comprehensive fitness and sets its head solution (i.e., the optimal inductance parameter setting in the field) as the final inductance distribution control strategy, which is used to guide the adjustment of the inductance parameters of the series reactor in the actual power system to optimize the power quality.

[0044] In a possible implementation, step S540 further includes step S548, and the expression of the power quality evaluation function is:

[0045]

[0046] Among them, F represents the adaptability of power quality, Q is the number of series reactors, and v i is the weight value of the ith series reactor, where the weight value is positively correlated with the circuit importance of the ith series reactor, w1 is the weight of the voltage stability coefficient, V ei is the voltage fluctuation variance of the circuit associated with the i-th series reactor, w2 is the weight of the frequency stability coefficient, Z ei is the frequency fluctuation variance of the circuit associated with the i-th series reactor, w3 is the weight of the harmonic content, X i is the harmonic content of the circuit associated with the ith series reactor. The greater the importance of the circuit of the ith series reactor, the higher its weight value v i Also bigger.

[0047] Step S600: constructing a reactance distribution topology based on the multiple series reactors, and mapping and controlling the reactance distribution topology according to the inductance distribution control strategy. A reactance distribution topology is built based on multiple series reactors, and the reactance distribution topology is mapped and controlled according to the inductance distribution control strategy. Specifically, multiple series reactors are arranged at different positions and lines in the distribution system to improve the voltage stability of the system, suppress harmonics, etc. The reactance distribution topology refers to the spatial distribution and connection relationship of these series reactors in the system. Building a reactance distribution topology means clarifying the position, capacity, connection method and other information of each series reactor in the system, and integrating them into a complete network model. The inductance distribution control strategy is applied to the reactance distribution topology for mapping control. Specifically, the calculated inductance value of each series reactor is allocated to the corresponding reactor according to the position and connection relationship in the reactance distribution topology, ensuring that each reactor can accurately receive its corresponding inductance value and can automatically adjust according to the value. The inductance value of each series reactor is automatically calculated and adjusted according to the inductance distribution control strategy to improve the stability and reliability of the power quality.

[0048] In the above, refer to Figure 1 The intelligent control method of a series reactor for improving power quality according to an embodiment of the present invention is described in detail. Figure 2 An intelligent control system for a series reactor for improving power quality according to an embodiment of the present invention is described.

[0049] The intelligent control system of series inductors for improving power quality according to the embodiment of the present invention is used to solve the technical problem that the existing series inductor control cannot monitor and accurately adjust the parameters of the reactor in real time, which makes it difficult to achieve rapid response and effective compensation for power quality problems, resulting in poor stability and reliability of power quality. The intelligent control of series inductors is realized, and the technical effect of improving the stability and reliability of power quality is achieved. The intelligent control system of series inductors for improving power quality includes: a parameter and topology structure identification module 10, a distribution twin topology network construction module 20, a convergent distribution twin network generation module 30, a monitoring data distribution map acquisition module 40, an inductance parameter optimization module 50, and a reactance distribution topology control module 60.

[0050] A parameter and topology identification module 10, wherein the parameter and topology identification module 10 is used to identify system parameters and topology of a target power distribution system;

[0051] A distribution twin topology network building module 20, wherein the distribution twin topology network building module 20 is used to determine a plurality of distribution monitoring nodes and build a distribution twin topology network in combination with the system parameters and the topology structure;

[0052] A convergent distribution twin network generation module 30, wherein the convergent distribution twin network generation module 30 is used to call historical distribution records, train the distribution twin topology network based on the historical distribution records, and generate a convergent distribution twin network;

[0053] A monitoring data distribution map acquisition module 40, wherein the monitoring data distribution map acquisition module 40 is used to collect real-time monitoring data sets of the multiple distribution monitoring nodes, and input the real-time monitoring data sets into the converged distribution twin network to perform data prediction of the subsidiary branches, and obtain a monitoring data distribution map;

[0054] An inductance parameter optimization module 50, which is used to optimize the inductance parameters of multiple series reactors of the target distribution system based on the monitoring data distribution diagram, combined with the grid load state, the capacitor state and the distribution twin topology network, and generate an inductance distribution control strategy;

[0055] The reactance distribution topology control module 60 is used to build a reactance distribution topology based on the multiple series reactors, and map and control the reactance distribution topology according to the inductance distribution control strategy.

[0056] The specific configuration of the distribution twin topology network building module 20 will be described in detail below. The distribution twin topology network building module 20 further includes: based on the digital twin technology, combining the system parameters and topology structure to simulate and model the target distribution system, and construct an initial distribution twin topology network; determine multiple distribution monitoring nodes and multiple subsidiary branches, wherein the subsidiary branches are branches without distribution monitoring nodes; embed the multiple distribution monitoring nodes and multiple subsidiary branches into the initial distribution twin topology network, and perform positioning and identification to generate the distribution twin topology network.

[0057] The specific configuration of the convergent distribution twin network generation module 30 will be described in detail below. The convergent distribution twin network generation module 30 may further include: extracting information from the historical distribution records based on the multiple distribution monitoring nodes and the multiple subsidiary branches to obtain multiple sample monitoring data sets and multiple sample data sets; based on the principle of machine learning, using the multiple sample monitoring data sets as training data and the multiple sample data sets as supervision data, supervised training and verification of the distribution twin topology network are performed to obtain a convergent distribution twin network that meets the expected training constraints.

[0058] The specific configuration of the monitoring data distribution map acquisition module 40 will be described in detail below. The monitoring data distribution map acquisition module 40 may further include: configuring a data preprocessing strategy, the data preprocessing strategy including a data cleaning strategy and a Fourier transform filtering and denoising strategy; configuring a data preprocessing scheme based on the data preprocessing strategy, integrating and building a data preprocessing model; preprocessing the real-time monitoring data set through the data preprocessing model to obtain a standard monitoring data set that is input into the converged distribution twin network for data prediction of the subsidiary branches.

[0059] The specific configuration of the inductance parameter optimization module 50 will be described in detail below. The inductance parameter optimization module 50 further includes: the capacitor state is a capacitance parameter of the capacitor, wherein the capacitor and the series reactor correspond one to one, and there is a first mapping relationship between the capacitance parameter and the inductance parameter;

[0060] The first mapping relationship expression is:

[0061]

[0062] Among them, L is the inductance parameter, f1 is the harmonic frequency, and C is the capacitance parameter; based on the monitoring data distribution diagram, the distribution twin topology network is assigned values ​​in combination with the grid load state and multiple capacitance parameters to generate a real-time distribution topology network; for the purpose of improving the power quality, the inductance parameters of multiple series reactors are optimized based on the power quality evaluation function and the real-time distribution topology network to generate the inductance distribution control strategy.

[0063] The specific configuration of the inductance parameter optimization module 50 will be described in detail below. The inductance parameter optimization module 50 further includes: obtaining multiple rated inductance parameter thresholds of the multiple series reactors; randomly generating multiple inductance parameter combinations based on the multiple rated inductance parameter thresholds, and setting them as multiple initial solutions; using the multiple initial solutions to simulate harmonic governance of the real-time distribution topology network, and generating multiple simulated harmonic governance results, wherein the simulated harmonic governance results include voltage stability coefficient, frequency stability coefficient, and harmonic content; using the power quality evaluation function to evaluate the multiple simulated harmonic governance results, and determine multiple quality fitness; and sorting the multiple initial solutions from large to small according to the quality fitness. The method comprises the following steps: determining N head solutions and M tail solutions based on the initial solution sequence, wherein N is less than M, and the sum of N and M is the number of initial solutions; randomly clustering the M tail solutions based on the N head solutions to generate N domains, and optimizing and iteratively updating the N domains according to a preset optimization strategy within the N domains, wherein the preset optimization strategy includes a predetermined optimization step length and a predetermined update strategy, and the predetermined update strategy includes a taboo update strategy and an allowed update strategy; until a preset optimization number threshold is met, outputting the current N domains, and selecting the head solution of the domain with the largest comprehensive fitness among the current N domains as the inductance distribution control strategy.

[0064] The specific configuration of the inductance parameter optimization module 50 will be described in detail below. The inductance parameter optimization module 50 may further include: The expression of the power quality evaluation function is:

[0065]

[0066] Among them, F represents the adaptability of power quality, Q is the number of series reactors, and v i is the weight value of the ith series reactor, where the weight value is positively correlated with the circuit importance of the ith series reactor, w1 is the weight of the voltage stability coefficient, V ei is the voltage fluctuation variance of the circuit associated with the i-th series reactor, w2 is the weight of the frequency stability coefficient, Z ei is the frequency fluctuation variance of the circuit associated with the i-th series reactor, w3 is the weight of the harmonic content, X i is the harmonic content of the circuit associated with the i-th series reactor.

[0067] The intelligent control system for series inductors for improving power quality provided in the embodiments of the present invention can execute the intelligent control method for series inductors for improving power quality provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0068] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules may be used and run on the user terminal and / or server, and the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0069] The above specific implementations do not constitute a limitation on the protection scope of this application. It should be understood by those skilled in the art that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of this application should be included in the protection scope of this application.

Claims

1. An intelligent control method for series reactors for improving power quality, characterized in that: The method comprises: Identify system parameters and topology of target power distribution system; Determine multiple power distribution monitoring nodes, and build a power distribution twin topology network based on the system parameters and topology structure; Calling historical power distribution records, training the power distribution twin topology network based on the historical power distribution records, and generating a converged power distribution twin network; Collecting real-time monitoring data sets of the multiple distribution monitoring nodes, and inputting the real-time monitoring data sets into the converged distribution twin network to perform data prediction of the subsidiary branches, and obtaining a monitoring data distribution map; Based on the monitoring data distribution diagram, combined with the grid load state, the capacitor state and the distribution twin topology network, the inductance parameters of multiple series reactors of the target distribution system are optimized to generate an inductance distribution control strategy; Building a reactance distribution topology based on the multiple series reactors, and mapping and controlling the reactance distribution topology according to the inductance distribution control strategy; Based on the monitoring data distribution diagram, combined with the grid load state, the capacitor state and the distribution twin topology network, the inductance parameters of multiple series reactors of the target distribution system are optimized to generate an inductance distribution control strategy, including: The capacitor state is a capacitance parameter of the capacitor, wherein the capacitor and the series reactor correspond one to one, and there is a first mapping relationship between the capacitance parameter and the inductance parameter; The first mapping relationship expression is: ； in, is the inductance parameter, is the harmonic frequency, C is the capacitance parameter; Based on the monitoring data distribution diagram, the distribution twin topology network is assigned a value in combination with the grid load state and multiple capacitance parameters to generate a real-time distribution topology network; For the purpose of improving power quality, inductance parameters of multiple series reactors are optimized based on the power quality evaluation function and the real-time distribution topology network to generate the inductance distribution control strategy.

2. The intelligent control method for series reactor for improving power quality according to claim 1, characterized in that: Determine multiple power distribution monitoring nodes, and build a power distribution twin topology network based on the system parameters and topology structure, including: Based on the digital twin technology, the target power distribution system is simulated and modeled in combination with the system parameters and topology structure to construct an initial power distribution twin topology network; Determine a plurality of power distribution monitoring nodes and a plurality of auxiliary branches, wherein the auxiliary branches are branches that do not have power distribution monitoring nodes; The multiple power distribution monitoring nodes and multiple subsidiary branches are embedded in the initial power distribution twin topology network, and are positioned and identified to generate the power distribution twin topology network.

3. The intelligent control method for series reactor for improving power quality according to claim 2, characterized in that: Training the distribution twin topology network based on the historical distribution records to generate a converged distribution twin network includes: Extracting information from the historical power distribution records based on the multiple power distribution monitoring nodes and the multiple subsidiary branches to obtain multiple sample monitoring data sets and multiple sample data sets; Based on the principle of machine learning, the multiple sample monitoring data sets are used as training data, and the multiple sample data sets are used as supervision data to perform supervised training and verification on the distribution twin topology network to obtain a converged distribution twin network that meets the expected training constraints.

4. The intelligent control method for series reactor for improving power quality according to claim 1, characterized in that: Inputting the real-time monitoring data set into the converged distribution twin network to perform data prediction of the subsidiary branches, the method also includes: Configure a data preprocessing strategy, wherein the data preprocessing strategy includes a data cleaning strategy and a Fourier transform filtering and denoising strategy; Configure a data preprocessing solution based on the data preprocessing strategy, and integrate and build a data preprocessing model; The real-time monitoring data set is preprocessed by the data preprocessing model to obtain a standard monitoring data set which is input into the converged distribution twin network for data prediction of the subsidiary branches.

5. The intelligent control method for series reactor for improving power quality according to claim 1, characterized in that: Based on the power quality evaluation function and the real-time power distribution topology network, inductance parameters of multiple series reactors are optimized to generate the inductance distribution control strategy, including: Acquire multiple rated inductance parameter thresholds of the multiple series reactors; Randomly generate a plurality of inductance parameter combinations based on the plurality of rated inductance parameter thresholds and set them as a plurality of initial solutions; Using the multiple initial solutions to perform simulated harmonic control on the real-time distribution topology network, respectively, to generate multiple simulated harmonic control results, wherein the simulated harmonic control results include a voltage stability coefficient, a frequency stability coefficient, and a harmonic content; Using the power quality evaluation function to evaluate the multiple simulated harmonic control results, and determine multiple quality fitness; Arrange the multiple initial solutions in descending order according to quality fitness, and determine N head solutions and M tail solutions based on the initial solution sequence, where N is less than M, and the sum of N and M is the number of initial solutions; Based on the N head solutions, the M tail solutions are randomly clustered to generate N domains, and within the N domains, the N domains are optimized and iteratively updated according to a preset optimization strategy, wherein the preset optimization strategy includes a predetermined optimization step length and a predetermined update strategy, and the predetermined update strategy includes a taboo update strategy and an allowed update strategy; Until the preset optimization number threshold is met, the current N fields are output, and the head solution of the field with the largest comprehensive fitness among the current N fields is selected and set as the inductance distribution control strategy.

6. The intelligent control method for series reactor for improving power quality according to claim 5, characterized in that: The method further comprises: The expression of the power quality evaluation function is: ； in, Characterizes the adaptability of power quality, Q is the number of series reactors, is the weight value of the i-th series reactor, where the weight value is positively correlated with the circuit importance of the i-th series reactor. is the weight of the voltage stability factor, is the voltage fluctuation variance of the circuit associated with the i-th series reactor, is the weight of the frequency stability coefficient, is the frequency fluctuation variance of the circuit associated with the i-th series reactor, is the weight of the harmonic content, is the harmonic content of the circuit associated with the i-th series reactor.

7. An intelligent control system for series reactors for improving power quality, characterized in that: The system is used to implement the intelligent control method for a series reactor for improving power quality according to any one of claims 1 to 6, and the system comprises: A parameter and topology identification module, wherein the parameter and topology identification module is used to identify system parameters and topology of a target power distribution system; A distribution twin topology network building module, which is used to determine multiple distribution monitoring nodes and build a distribution twin topology network in combination with the system parameters and topology structure; A convergent distribution twin network generation module, which is used to call historical distribution records, train the distribution twin topology network based on the historical distribution records, and generate a convergent distribution twin network; A monitoring data distribution map acquisition module, wherein the monitoring data distribution map acquisition module is used to collect real-time monitoring data sets of the multiple distribution monitoring nodes, and input the real-time monitoring data sets into the converged distribution twin network to perform data prediction of the subsidiary branches, and obtain a monitoring data distribution map; An inductance parameter optimization module, which is used to optimize the inductance parameters of multiple series reactors of the target distribution system based on the monitoring data distribution diagram, combined with the grid load state, the capacitor state and the distribution twin topology network, and generate an inductance distribution control strategy; A reactance distribution topology control module is used to build a reactance distribution topology based on the multiple series reactors, and to map and control the reactance distribution topology according to the inductance distribution control strategy.

Images