A reactive power compensation device and control method for power supply network

Abstract

The invention discloses a reactive power compensation device and a control method for a power supply network, which belongs to the technical field of power supply networks, and solves the problem that the existing method determines the reactive power compensation phase difference of the inductive load only by the power factor, and cannot evaluate and analyze the real-time status of the inductive load at the AC end and the DC end. The method comprises acquiring the operation data of the power supply network, executing a pre-built power compensation model, obtaining the power health of the inductive load in the dynamic monitoring network, determining the compensation mode of the inductive load by a compensation priority model, and generating a reactive power compensation instruction; the embodiment of the invention extracts the operation data of the inductive load in the dynamic monitoring network by constructing a power compensation model that accurately evaluates the inductive load of the power supply network, and can obtain the accurate power health of the inductive load in the dynamic monitoring network, thereby accurately compensating for the harmonics and reactive current of the inductive load in the power supply network whose frequency and amplitude both change.

Description

Technical Field

[0001] The present invention belongs to the technical field of power supply networks, and in particular relates to a reactive power compensation device and a control method for a power supply network. Background Art

[0002] With the continuous development of cross-regional interconnection of my country's power grid, the demand for reactive power for large power supply equipment such as large transmission power and long-distance transmission lines has also increased dramatically. Therefore, reactive power compensation is becoming more and more important for these large power supply equipment. Reactive power compensation, the full name of which is reactive power compensation, is a technology used to improve the power factor of the power grid in the power supply system. Its main function is to reduce the loss of power supply transformers and transmission lines, improve power supply efficiency, and thus improve the power supply environment. Reactive power compensation devices play a vital role in power supply systems. Reasonable selection of compensation devices can minimize the loss of the power grid, thereby improving the quality of the power grid.

[0003] Chinese patent CN104377710A discloses a method for reactive power compensation of a transformer, comprising: detecting the voltage and current at the DC and AC ends of a converter transformer, calculating the phase difference between the voltage at the AC end and the current at the AC end, calculating the phase difference between the voltage at the DC end and the current at the DC end, taking the phase difference between the voltage at the AC end and the current at the AC end, and the voltage at the AC end and the current at the AC end as parameters, calculating the power at the AC end, and taking the phase difference between the voltage at the DC end and the current at the DC end, and the voltage at the DC end and the current at the DC end as parameters, calculating the power at the DC end, and taking the difference between the power at the AC end and the power at the DC end as reactive power compensation power and outputting it; however, the existing method only determines the reactive power compensation phase difference of the inductive load by the power factor, and cannot evaluate and analyze the real-time status of the inductive load at the AC end and the DC end, resulting in a waste of power resources and poor reactive power compensation effect. In order to solve the above problems, we propose a reactive power compensation device and control method for a power supply network. Summary of the invention

[0004] The purpose of the present invention is to address the deficiencies of the prior art and provide a power supply network reactive power compensation device and control method, which solves the problem that the prior method only determines the reactive power compensation phase difference of the inductive load by the power factor and cannot evaluate and analyze the real-time status of the inductive load at the AC end and the DC end.

[0005] The existing method only determines the reactive compensation phase difference of the inductive load by the power factor, and is unable to evaluate and analyze the real-time status of the inductive load at the AC end and the DC end, resulting in a waste of power resources and poor reactive compensation effect. In order to solve the above problems, we propose a reactive compensation control method for a power supply network. The control method includes obtaining the operation data of the power supply network, constructing a power supply network infrastructure based on the power supply network database and a dynamic monitoring network of the power supply network based on the operation data of the power supply network, and then constructing a power compensation model based on the historical operation data of the power supply network, extracting the operation data of the inductive load in the dynamic monitoring network, and executing the pre-constructed power compensation model with the operation data of the inductive load as input to obtain the power health of the inductive load in the dynamic monitoring network, determining the compensation mode of the inductive load based on the power health of the inductive load in the dynamic monitoring network and the pre-constructed compensation priority model, and generating reactive compensation instructions. The embodiment of the present invention constructs a power compensation model that accurately evaluates the inductive load of the power supply grid to extract the operating data of the inductive load in the dynamic monitoring network, and can obtain accurate power health of the inductive load in the dynamic monitoring network, thereby accurately compensating for the harmonics and reactive currents in the power supply grid whose frequency and amplitude both change. Compared with the prior art, it avoids the waste of power resources, has lower energy consumption, and can also avoid damage to the inductive load during reactive compensation.

[0006] The present invention is implemented as follows: a reactive power compensation control method for a power supply network, the reactive power compensation control method for a power supply network comprising:

[0007] Acquire power grid operation data, upload the power grid operation data to the power grid database, build a power grid infrastructure based on the power grid database and a dynamic monitoring network for the power grid based on the power grid operation data, wherein the power grid operation data includes the operation data of the power supply end, the distribution end, the load end, the inductive load, the capacitive load and the resistive load in the power supply line;

[0008] Obtaining historical operation data of the power supply grid in the power supply grid database, building a power compensation model based on the historical operation data of the power supply grid, extracting the operation data of the inductive load in the dynamic monitoring network, and executing the pre-built power compensation model with the operation data of the inductive load as input to obtain the power health of the inductive load in the dynamic monitoring network;

[0009] Based on the power health of the inductive load in the dynamic monitoring network and the pre-built compensation priority model, the inductive load compensation mode is determined and a reactive power compensation instruction is generated.

[0010] Preferably, it also includes:

[0011] In response to the reactive power compensation instruction, the operating data of the capacitive load in the dynamic monitoring network is retrieved, the capacitive load is controlled to be turned on and off, and reactive power compensation is performed on the inductive load;

[0012] The operation data of the power supply grid after reactive power compensation is obtained, and the operation data of the inductive load is used as input to execute the pre-built power compensation model to obtain the power health of the inductive load in the dynamic monitoring network, and determine whether the power health of the inductive load is greater than the compensation threshold. If it is greater than the compensation threshold, a control completion instruction is generated; if it is less than the compensation threshold, an early warning instruction is triggered.

[0013] Preferably, the method for constructing a power supply grid infrastructure based on a power supply grid database and a dynamic monitoring network of a power supply grid based on power supply grid operation data specifically includes:

[0014] Traversing the power supply network database, extracting three-dimensional coordinate parameters of the inductive load, the capacitive load and the resistive load in the power supply network database;

[0015] Perform three-dimensional modeling based on three-dimensional coordinate parameters to obtain an initial dynamic monitoring framework;

[0016] Traversing the power grid database, extracting historical operation data of the inductive load in the power grid database, and building and training a compensation priority model based on the historical operation data;

[0017] The compensation priority model determines the compensation priority coefficient of the inductive load based on the historical operation data of the inductive load, and renders the compensation priority coefficient to the initial dynamic monitoring architecture to obtain the dynamic monitoring network of the power supply grid.

[0018] Preferably, the method for determining the compensation priority coefficient of the inductive load based on the historical operation data of the inductive load by the compensation priority model specifically includes:

[0019] Acquire historical operation data of the inductive load, and extract configuration information of the inductive load from the historical operation data of the inductive load;

[0020] Based on the configuration information of the inductive load and the improved hierarchical analysis method, the weight distribution of the initial dynamic monitoring architecture is determined, and the initial weight is assigned to the inductive load;

[0021] Obtain the historical power factor in the historical operation data of the inductive load, and determine the compensation rate and compensation level of the inductive load based on the historical power factor;

[0022] The calculation formula of the compensation rate is:

[0023] (1)

[0024] Among them, the is the compensation rate of the inductive load, and Data collection time period for inductive loads The number of inductive load acquisitions within It is the number of inductive load acquisitions when the historical power factor of the inductive load is 1.

[0025] Preferably, the method for determining the compensation priority coefficient of the inductive load based on the historical operation data of the inductive load by the compensation priority model specifically further includes:

[0026] Obtain the initial weight, compensation rate and compensation level assigned to the inductive load, normalize the compensation rate and compensation level based on the dimensional effect and normalization function to eliminate the index, and obtain the normalized priority index value ;

[0027] (2)

[0028] Among them, the is a normalized vector, and They represent the pre-coefficient and post-coefficient of the normalization function respectively. is the compensation rate of the inductive load, Assign initial weights to the inductive loads;

[0029] (3)

[0030] in, They represent the pre-coefficient and post-coefficient of the normalization function respectively. is the normalized curvature of the normalized function;

[0031] Acquire multiple groups of normalized priority index values, and calculate the inductive load compensation comparison value based on the normalized priority index values ​​and the compensation comparison function;

[0032] (4)

[0033] in, represents the normalized priority index value, and is the inductive load compensation comparison value, is the number of inductive loads, Then it represents the comparison coefficient of the compensation comparison function;

[0034] Calculating a compensation priority coefficient based on a comparison value of inductive load compensation;

[0035] (5);

[0036] Among them, the represents the compensation priority factor of the inductive load, and is the number of inductive loads, Compare the mean value of load compensation for inductive loads;

[0037] (6)

[0038] Among them, the represents the compensation priority factor of the inductive load, and is the number of inductive loads.

[0039] Preferably, the method for constructing the power compensation model specifically includes:

[0040] Obtain historical operation data of the power supply network in the power supply network database, select the historical operation data using a non-uniform extraction method, and divide the historical operation data into a training set and a test set;

[0041] Retrieving an initial model of a power compensation model, wherein the initial model of the power compensation model is an MLP neural network model, and setting an input layer, an output layer, and initial hyperparameters of the MLP neural network model;

[0042] Taking the training set as input, executing the initial model of the power compensation model, adjusting the hyperparameters of the initial model of the power compensation model based on the activation function, and finding multiple groups of power compensation models based on the grid optimization method;

[0043] Iteratively train the power compensation model for multiple rounds until convergence, stop the training, and select multiple groups of converged power compensation models using an early stopping method;

[0044] Obtain a validation set to determine whether the power compensation model is overfitting, and determine the accuracy of the power compensation model based on the preset validation accuracy. If the validation accuracy is met, complete the construction of the power compensation model and output the power compensation model.

[0045] Preferably, the method of executing the pre-built power compensation model to obtain the method of dynamically monitoring the power health of the inductive load in the network specifically includes:

[0046] Extract and input the operating data of the inductive load in the dynamic monitoring network;

[0047] The power compensation model calculates the power factor of the inductive load and calculates the abnormal radiation value of the inductive load based on the operating data of the inductive load and the power factor. ;

[0048] (7)

[0049] in, is the number of attributes of the inductive load operating data, and For the input inductive load operating data, is the power factor of the inductive load, is the radiation correlation coefficient, which is obtained through the radiation function;

[0050] (8)

[0051] Among them, the is the number of inductive loads, and is the capacitive load quantity, is the number of resistive loads, is the load influence coefficient;

[0052] Correcting the power factor based on the abnormal radiation value to obtain a power factor correction value;

[0053] (9)

[0054] in, is the power factor of the inductive load, is the radiation correlation coefficient.

[0055] Preferably, the method of executing the pre-built power compensation model to obtain the method of dynamically monitoring the power health of the inductive load in the network specifically includes:

[0056] Loading the power factor correction value and the abnormal radiation value of the inductive load, taking the power factor correction value and the abnormal radiation value of the inductive load as input, executing the power compensation model, and obtaining the power health of the inductive load in the dynamic monitoring network;

[0057] (10)

[0058] in, is the number of attributes of the inductive load operating data, and For the input inductive load operating data, Abnormal radiation value.

[0059] Preferably, determining the inductive load compensation mode based on the power health of the inductive load in the dynamic monitoring network and the pre-built compensation priority model specifically includes:

[0060] Obtain the power health of inductive loads in the dynamic monitoring network;

[0061] Determine whether the inductive load needs reactive power compensation and generate a queue to be compensated;

[0062] The compensation priority model identifies the running data of the queue to be compensated;

[0063] The priority coefficients of the inductive loads in the dynamic monitoring network are retrieved, and a compensation priority queue is generated based on the priority coefficients.

[0064] On the other hand, the present invention further provides a power supply network reactive power compensation device, the power supply network reactive power compensation device specifically comprising:

[0065] A data acquisition module, wherein the data acquisition module is used to obtain power grid operation data, upload the power grid operation data to a power grid database, and construct a power grid infrastructure based on the power grid database and a dynamic monitoring network for the power grid based on the power grid operation data;

[0066] A health calculation module, which is used to obtain historical operation data of the power supply network in the power supply network database, build a power compensation model based on the historical operation data of the power supply network, extract the operation data of the inductive load in the dynamic monitoring network, and use the operation data of the inductive load as input to execute the pre-built power compensation model to obtain the power health of the inductive load in the dynamic monitoring network;

[0067] A priority compensation determination module, wherein the priority compensation determination module determines the inductive load compensation mode based on the power health of the inductive load in the dynamic monitoring network and a pre-built compensation priority model, and generates a reactive compensation instruction;

[0068] A compensation response module, which responds to a reactive power compensation instruction, retrieves the capacitive load operation data in the dynamic monitoring network, controls the capacitive load on and off, and performs reactive power compensation for the inductive load;

[0069] An early warning judgment module is used to obtain the operating data of the power supply network after reactive power compensation, take the operating data of the inductive load as input, execute the pre-built power compensation model, obtain the power health of the inductive load in the dynamic monitoring network, and judge whether the power health of the inductive load is greater than the compensation threshold. If it is greater than the compensation threshold, a control completion instruction is generated; if it is less than the compensation threshold, a early warning instruction is triggered.

[0070] Compared with the prior art, the embodiments of the present application have the following beneficial effects:

[0071] The embodiment of the present invention constructs a power compensation model that accurately evaluates the inductive load of the power supply grid to extract the operating data of the inductive load in the dynamic monitoring network, and can obtain accurate power health of the inductive load in the dynamic monitoring network, thereby accurately compensating for the harmonics and reactive currents in the power supply grid whose frequency and amplitude both change. Compared with the prior art, it avoids the waste of power resources, has lower energy consumption, and can also avoid damage to the inductive load during reactive compensation.

[0072] The embodiment of the present invention constructs a dynamic monitoring network of the power supply grid through the power supply grid operation data, which can facilitate real-time presentation of the operation status of the power supply grid and facilities in the background, and also facilitates real-time presentation of abnormal inductive loads when abnormal reactive compensation occurs. At the same time, compared with the existing three-dimensional modeling technology, the dynamic monitoring network renders the compensation priority coefficient of the inductive load, thereby providing decision support for reactive compensation.

[0073] In the embodiment of the present invention, the compensation priority coefficient of the inductive load can be determined through the compensation priority model, thereby solving the problem that there are many inductive loads in the power supply grid and the control and connection relationship of the inductive loads are complex. When performing reactive power compensation on multiple groups of inductive loads, it is necessary to determine the priority of the inductive loads. After determining the compensation priority coefficient of the inductive load, the reactive compensation operation can be performed in conjunction with the power compensation model, making the reactive compensation more accurate and efficient, and further avoiding the waste of power resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] Figure 1 It is a schematic diagram of the implementation flow of the reactive power compensation control method for the power supply network provided by the present invention.

[0075] Figure 2 It is a schematic diagram of the implementation process of the method for constructing a power supply grid infrastructure based on a power supply grid database and a dynamic monitoring network of a power supply grid based on power supply grid operation data provided by the present invention.

[0076] Figure 3 It is a schematic diagram of the implementation flow of a method for determining a compensation priority coefficient of an inductive load based on historical operating data of the inductive load by the compensation priority model provided by the present invention.

[0077] Figure 4 It is a schematic diagram of the implementation flow of the method for constructing a power compensation model provided by the present invention.

[0078] Figure 5 The present invention provides a schematic diagram of the implementation process of a method for dynamically monitoring the power health of inductive loads in a network by executing a pre-built power compensation model.

[0079] Figure 6 It is a schematic diagram of an implementation flow of determining a compensation method for an inductive load based on the power health of the inductive load in the dynamic monitoring network and a pre-built compensation priority model provided by the present invention.

[0080] Figure 7 It is a structural schematic diagram of the reactive power compensation control device for a power supply network provided by the present invention. DETAILED DESCRIPTION

[0081] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by technicians in the technical field of this application; the terms used in the specification of the application herein are only for the purpose of describing specific embodiments and are not intended to limit this application; the terms "including" and "having" and any variations thereof in the specification and claims of this application and the above-mentioned drawings are intended to cover non-exclusive inclusions. The terms "first", "second", etc. in the specification and claims of this application or the above-mentioned drawings are used to distinguish different objects, not to describe a specific order.

[0082] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0083] The existing method only determines the reactive compensation phase difference of the inductive load by the power factor, and is unable to evaluate and analyze the real-time status of the inductive load at the AC end and the DC end, resulting in a waste of power resources and poor reactive compensation effect. In order to solve the above problems, we propose a reactive compensation control method for a power supply network. The control method includes obtaining the operation data of the power supply network, constructing a power supply network infrastructure based on the power supply network database and a dynamic monitoring network of the power supply network based on the operation data of the power supply network, and then constructing a power compensation model based on the historical operation data of the power supply network, extracting the operation data of the inductive load in the dynamic monitoring network, and executing the pre-constructed power compensation model with the operation data of the inductive load as input to obtain the power health of the inductive load in the dynamic monitoring network, determining the compensation mode of the inductive load based on the power health of the inductive load in the dynamic monitoring network and the pre-constructed compensation priority model, and generating reactive compensation instructions. The embodiment of the present invention constructs a power compensation model that accurately evaluates the inductive load of the power supply grid to extract the operating data of the inductive load in the dynamic monitoring network, and can obtain accurate power health of the inductive load in the dynamic monitoring network, thereby accurately compensating for the harmonics and reactive currents in the power supply grid whose frequency and amplitude both change. Compared with the prior art, it avoids the waste of power resources, has lower energy consumption, and can also avoid damage to the inductive load during reactive compensation.

[0084] The embodiment of the present invention provides a reactive power compensation control method for a power supply network. Figure 1 The schematic diagram of the implementation flow of the reactive power compensation control method of the power supply network is shown, and the reactive power compensation control method of the power supply network specifically includes:

[0085] Step S10, obtaining the power supply grid operation data, uploading the power supply grid operation data to the power supply grid database, building the power supply grid infrastructure based on the power supply grid database and building a dynamic monitoring network for the power supply grid based on the power supply grid operation data, wherein the power supply grid operation data includes the operation data of the power supply end, the distribution end, the load end, the inductive load, the capacitive load and the resistive load in the power supply line.

[0086] It should be noted that the power supply end, distribution end, load end, and power supply line constitute the power supply network described in this embodiment, and the power supply end, distribution end, and load end are electrically connected through the power supply line, and the power supply end includes but is not limited to power station boilers, steam turbines, gas turbines, turbines, generators and transformers, and the distribution end includes but is not limited to transformers, incoming line cabinets, outgoing line cabinets, capacitor cabinets, compensation cabinets, feeder cabinets, UPS uninterruptible power supplies, UPS input cabinets, UPS output cabinets, drawer cabinets and ancillary facilities, and the inductive load in this application can be an electric motor, a transformer, a motor, and an electromagnetic valve.

[0087] The power grid operation data includes but is not limited to the operating current, load voltage, parameters of the main equipment in the grid, topological relationships and measurement data of the equipment during operation, etc.

[0088] Step S20, obtaining historical operation data of the power supply grid in the power supply grid database, constructing a power compensation model based on the historical operation data of the power supply grid, extracting the operation data of the inductive load in the dynamic monitoring network, taking the operation data of the inductive load as input, executing the pre-constructed power compensation model, and obtaining the power health of the inductive load in the dynamic monitoring network;

[0089] Step S30, determining the inductive load compensation method based on the power health of the inductive load in the dynamic monitoring network and the pre-built compensation priority model, and generating a reactive power compensation instruction.

[0090] Step S40, in response to the reactive power compensation instruction, retrieve the capacitive load operation data in the dynamic monitoring network, control the capacitive load on and off, and perform reactive power compensation for the inductive load. It should be noted that the capacitive load in this embodiment can be a capacitor.

[0091] Step S50, obtain the operating data of the power supply grid after reactive power compensation, take the operating data of the inductive load as input, execute the pre-built power compensation model, obtain the power health of the inductive load in the dynamic monitoring network, and determine whether the power health of the inductive load is greater than the compensation threshold. If it is greater than the compensation threshold, generate a control completion instruction; if it is less than the compensation threshold, trigger an early warning instruction.

[0092] In this embodiment, the compensation threshold may be set to 0.95, which is used to judge the operating state of the inductive load after reactive power compensation of the inductive load.

[0093] The embodiment of the present invention constructs a power compensation model that accurately evaluates the inductive load of the power supply grid to extract the operating data of the inductive load in the dynamic monitoring network, and can obtain accurate power health of the inductive load in the dynamic monitoring network, thereby accurately compensating for the harmonics and reactive currents in the power supply grid whose frequency and amplitude both change. Compared with the prior art, it avoids the waste of power resources, has lower energy consumption, and can also avoid damage to the inductive load during reactive compensation.

[0094] The embodiment of the present invention provides a method for constructing a power supply grid infrastructure based on a power supply grid database and a method for constructing a dynamic monitoring network of a power supply grid based on power supply grid operation data. Figure 2 The present invention shows a schematic diagram of the implementation process of a method for constructing a power supply grid infrastructure based on a power supply grid database and a method for constructing a dynamic monitoring network of a power supply grid based on power supply grid operation data. The method for constructing a power supply grid infrastructure based on a power supply grid database and a method for constructing a dynamic monitoring network of a power supply grid based on power supply grid operation data specifically includes:

[0095] Step S101, traversing the power supply network database, extracting three-dimensional coordinate parameters of the inductive load, capacitive load and resistive load in the power supply network database;

[0096] It should be noted that the power supply grid database stores the three-dimensional coordinate parameters of inductive loads, capacitive loads and resistive loads. At the same time, the power supply grid database can store multi-source heterogeneous power supply grid operation data. The data format of the power supply grid database can be text files, web josh, XML, table data, images, etc.

[0097] Step S102, performing three-dimensional modeling based on the three-dimensional coordinate parameters to obtain an initial dynamic monitoring framework;

[0098] When doing 3D modeling, you can choose parametric modeling software and non-parametric modeling software for 3D modeling.

[0099] Step S103, traversing the power grid database, extracting historical operation data of the inductive load in the power grid database, and constructing and training a compensation priority model based on the historical operation data;

[0100] Step S104 , the compensation priority model determines the compensation priority coefficient of the inductive load based on the historical operation data of the inductive load, and renders the compensation priority coefficient to the initial dynamic monitoring framework to obtain a dynamic monitoring network of the power supply grid.

[0101] It should be noted that the compensation priority model can specifically be a deep network based on deep learning. Compared with the existing technology, the compensation priority model constructed based on deep learning can solve the defects of traditional machine learning methods. It takes less time to determine the compensation priority coefficient of the inductive load, and automatically extracts high-level features, which greatly reduces the problem of feature engineering that traditional machine learning relies on. The compensation priority model is composed of a recurrent neural network with multiple hidden layers, a convolutional layer and a multi-layer perceptron. When constructing the compensation priority model, the weights can be initialized using He or Xavier, and the forward propagation algorithm can be used to establish the neural network input, output, network connection method, number of network layers and number of nodes in each layer.

[0102] The embodiment of the present invention constructs a dynamic monitoring network of the power supply grid through the power supply grid operation data, which can facilitate real-time presentation of the operation status of the power supply grid and facilities in the background, and also facilitates real-time presentation of abnormal inductive loads when abnormal reactive compensation occurs. At the same time, compared with the existing three-dimensional modeling technology, the dynamic monitoring network renders the compensation priority coefficient of the inductive load, thereby providing decision support for reactive compensation.

[0103] Considering that there are many inductive loads in the power supply grid and the control and connection relationship of the inductive loads are complicated, when performing reactive power compensation for multiple groups of inductive loads, it is necessary to determine the priority of the inductive loads, so as to avoid the problem of insufficient power or failure of inductive loads with high importance and weight. In order to solve the above problem, an embodiment of the present invention provides a method for determining the compensation priority coefficient of the inductive load based on the historical operation data of the inductive load by the compensation priority model. Figure 3 The schematic diagram of the implementation flow of the method for determining the compensation priority coefficient of the inductive load based on the historical operation data of the inductive load by the compensation priority model is shown. The method for determining the compensation priority coefficient of the inductive load based on the historical operation data of the inductive load by the compensation priority model specifically includes:

[0104] Step S1041, acquiring historical operation data of the inductive load, and extracting configuration information of the inductive load from the historical operation data of the inductive load;

[0105] Step S1042, determining the weight distribution of the initial dynamic monitoring architecture based on the configuration information of the inductive load and the improved analytic hierarchy process, and assigning an initial weight to the inductive load;

[0106] It should be noted that the improved analytic hierarchy process (AHP) is a decision-making method, the core of which is to decompose the elements related to the decision into levels such as goals, criteria, and plans, and then conduct qualitative and quantitative analysis. This method usually includes four steps: constructing a hierarchical structure of the indicator system, determining the scale and constructing the judgment matrix, calculating the weight vector and consistency test, and in this embodiment, the improved analytic hierarchy process is used to determine the weight distribution of the initial dynamic monitoring architecture. Specifically, expert scoring and AHP Toolbox scoring can be introduced to determine the weight distribution of the initial dynamic monitoring architecture, and the initial weight is assigned to the perceptual load.

[0107] Step S1043, obtaining the historical power factor in the historical operation data of the inductive load, and determining the compensation rate and compensation level of the inductive load based on the historical power factor;

[0108] In this embodiment, the compensation level is determined by the historical power factor. For example, if the historical power factor is 0.9-0.95, the compensation level is 5, if the historical power factor is 0.8-0.9, the compensation level is 4, if the historical power factor is 0.6-0.8, the compensation level is 3, if the historical power factor is 0.3-0.6, the compensation level is 2, and when the historical power factor is less than 0.2, the compensation level is 1.

[0109] It should be noted that the calculation and acquisition methods of the historical power factor are both existing technologies, but they only determine the reactive compensation phase difference of the inductive load through the power factor, and cannot evaluate and analyze the real-time status of the inductive load, which can easily cause waste of power resources and poor reactive compensation effect. The present application can obtain the historical power factor in the historical operation data of the inductive load, and determine the compensation rate and compensation level of the inductive load based on the size of the historical power factor, so as to determine the reactive compensation priority and the phase difference of the reactive compensation, thereby improving the reactive compensation effect.

[0110] The calculation formula of the compensation rate is:

[0111] (1)

[0112] Among them, the is the compensation rate of the inductive load, and Data collection time period for inductive loads The number of inductive load acquisitions within It is the number of inductive load collections when the historical power factor of the inductive load is 1. It should be noted that the historical power factor of the inductive load is 1, which means that the inductive load is operating normally and does not need reactive power compensation.

[0113] Step S1044, obtaining the initial weight, compensation rate and compensation level assigned to the inductive load, performing normalization processing on the compensation rate and compensation level to eliminate the index based on the dimensional effect and the normalization function, and obtaining the normalized priority index value ;

[0114] (2)

[0115] Among them, the is a normalized vector, and Respectively represent the pre-coefficient and post-coefficient of the normalization function. In this embodiment The value range of can be 0.1-0.3, is the compensation rate of the inductive load, Assign initial weights to the inductive loads;

[0116] (3)

[0117] in, They represent the pre-coefficient and post-coefficient of the normalization function respectively. is the normalized curvature of the normalized function. In this embodiment, the normalized curvature has a value range of 0.1-0.6, and may be 0.15.

[0118] It should be noted that the normalized priority index value It is an intermediate value used to calculate the comparative value of inductive load compensation. It is calculated based on dimensional effect and normalized function. It can be understood as an indicator of inductive load status that combines compensation rate and compensation level.

[0119] Step S1045, obtaining multiple groups of normalized priority index values, and calculating the inductive load compensation comparison value based on the normalized priority index values ​​and the compensation comparison function;

[0120] (4)

[0121] in, represents the normalized priority index value, and is the inductive load compensation comparison value, is the number of inductive loads, represents the comparison coefficient of the compensation comparison function. In this embodiment, the comparison coefficient of the compensation comparison function may be 0.15;

[0122] Step S1046, calculating a compensation priority coefficient based on the inductive load compensation comparison value;

[0123] (5);

[0124] Among them, the represents the compensation priority factor of the inductive load, and is the number of inductive loads, Compare the mean value of load compensation for inductive loads;

[0125] (6)

[0126] Among them, the represents the compensation priority factor of the inductive load, and is the number of inductive loads.

[0127] In the embodiment of the present invention, the compensation priority coefficient of the inductive load can be determined through the compensation priority model, thereby solving the problem that there are many inductive loads in the power supply grid and the control and connection relationship of the inductive loads are complex. When performing reactive power compensation on multiple groups of inductive loads, it is necessary to determine the priority of the inductive loads. After determining the compensation priority coefficient of the inductive load, the reactive compensation operation can be performed in conjunction with the power compensation model, making the reactive compensation more accurate and efficient, and further avoiding the waste of power resources.

[0128] The embodiment of the present invention provides a method for constructing a power compensation model. Figure 4 A schematic diagram of the implementation process of a method for constructing a power compensation model is shown, and the method for constructing a power compensation model specifically includes:

[0129] Step S201, obtain the historical operation data of the power supply network in the power supply network database, select the historical operation data using a non-uniform extraction method, and divide the historical operation data into a training set and a test set; it should be noted that the ratio of the training set to the test set can be 7:3, and the use of a non-uniform extraction method can ensure the randomness of the sampling, so that the constructed model evaluation accuracy is higher.

[0130] Step S202, calling an initial model of a power compensation model, wherein the initial model of the power compensation model is an MLP neural network model, and setting an input layer, an output layer and initial hyperparameters of the MLP neural network model;

[0131] In this embodiment, the input layer and the output layer can both be 3 layers; and the number of nodes in the input layer and the output layer is 34, 58, and 60.

[0132] Step S203, using the training set as input, executing the initial model of the power compensation model, adjusting the hyperparameters of the initial model of the power compensation model based on the activation function, and finding multiple groups of power compensation models based on a grid optimization method;

[0133] Step S204, iteratively training the power compensation model for multiple rounds until convergence, stopping the training, and selecting multiple groups of converged power compensation models using an early stopping method;

[0134] In this embodiment, the iterative training rounds may be 100-1000 times.

[0135] Step S205, obtaining a verification set, and determining whether the power compensation model is overfitted. In this embodiment, the criterion for determining overfitting may be that the result accuracy of the verification data is greater than the actual result accuracy.

[0136] Step S206, if there is no overfitting phenomenon, determine the accuracy of the power compensation model based on the preset verification accuracy; if there is overfitting phenomenon, return to step S202;

[0137] Step S207, if it meets the verification accuracy, complete the construction of the power compensation model and output the power compensation model. If it does not meet the verification accuracy, return to step S202. In this embodiment, the verification accuracy can be 0.6-0.7.

[0138] The embodiment of the present invention can accurately evaluate and measure the state of the inductive load by constructing an accurate power compensation model, thereby providing data support for reactive power compensation, and feedback the operating status of the power supply network equipment through health status, so as to facilitate timely intervention and processing by technical personnel when abnormalities or failures occur in the power supply network equipment.

[0139] The embodiment of the present invention provides a method for executing a pre-built power compensation model to obtain a method for dynamically monitoring the power health of an inductive load in a network. Figure 5 The present invention shows a schematic diagram of a method for dynamically monitoring the power health of an inductive load in a network by executing a pre-built power compensation model. The method for dynamically monitoring the power health of an inductive load in a network by executing a pre-built power compensation model specifically includes:

[0140] Step S301, extracting and inputting the operating data of the inductive load in the dynamic monitoring network;

[0141] The power compensation model calculates the power factor of the inductive load and calculates the abnormal radiation value of the inductive load based on the operating data of the inductive load and the power factor. It should be noted that the abnormal radiation value The impact on other power equipment when the inductive load has a phase difference and needs reactive power compensation;

[0142] (7)

[0143] in, is the number of attributes of the inductive load operation data. In this embodiment, the number of attributes of the data may be 3-8, and the data attributes may be current data, voltage data, load data, etc., and For the input inductive load operating data, is the power factor of the inductive load, is the radiation correlation coefficient, which is obtained through the radiation function;

[0144] (8)

[0145] Among them, the is the number of inductive loads, and is the capacitive load quantity, is the number of resistive loads, is the load influence coefficient. In this embodiment, the load influence coefficient ranges from 0.2 to 0.9;

[0146] Step S302, correcting the power factor based on the abnormal radiation value to obtain a power factor correction value;

[0147] (9)

[0148] in, is the power factor of the inductive load, is the radiation correlation coefficient.

[0149] Step S303, loading the power factor correction value and the abnormal radiation value of the inductive load, taking the power factor correction value and the abnormal radiation value of the inductive load as input, executing the power compensation model, and obtaining the power health of the inductive load in the dynamic monitoring network;

[0150] (10)

[0151] in, is the number of attributes of the inductive load operating data, and For the input inductive load operating data, Abnormal radiation value.

[0152] The embodiment of the present invention provides a method for determining a compensation mode for an inductive load based on the power health of the inductive load in the dynamic monitoring network and a pre-built compensation priority model. Figure 6 The schematic diagram of the implementation process of determining the compensation mode of the inductive load based on the power health of the inductive load in the dynamic monitoring network and the pre-built compensation priority model is shown. The inductive load compensation mode is determined based on the power health of the inductive load in the dynamic monitoring network and the pre-built compensation priority model, specifically including:

[0153] Step S401, obtaining the power health of the inductive load in the dynamic monitoring network;

[0154] Step S402, determining whether the inductive load needs reactive power compensation;

[0155] Step S403, if the inductive load needs to be compensated for reactive power, a queue to be compensated is generated; if reactive power compensation is not needed, a normal operation instruction of the inductive load is fed back;

[0156] Step S404, the compensation priority model identifies the operation data of the queue to be compensated;

[0157] Step S405 , retrieve the priority coefficient of the inductive load in the dynamic monitoring network, and generate a compensation priority queue based on the priority coefficient.

[0158] The embodiment of the present invention provides a reactive power compensation control device for a power supply network. Figure 7 The structure diagram of the reactive power compensation control device of the power supply network is shown, and the reactive power compensation control device of the power supply network specifically comprises:

[0159] A data acquisition module 100, the data acquisition module 100 is used to obtain power grid operation data, upload the power grid operation data to a power grid database, build a power grid infrastructure based on the power grid database and build a dynamic monitoring network for the power grid based on the power grid operation data;

[0160] A health calculation module 200, which is used to obtain historical operation data of the power supply network in the power supply network database, build a power compensation model based on the historical operation data of the power supply network, extract the operation data of the inductive load in the dynamic monitoring network, and use the operation data of the inductive load as input to execute the pre-built power compensation model to obtain the power health of the inductive load in the dynamic monitoring network;

[0161] A priority compensation determination module 300, wherein the priority compensation determination module 300 determines a compensation mode for an inductive load based on the power health of the inductive load in the dynamic monitoring network and a pre-built compensation priority model, and generates a reactive compensation instruction;

[0162] The compensation response module 400 responds to the reactive power compensation instruction, retrieves the capacitive load operation data in the dynamic monitoring network, controls the capacitive load on and off, and performs reactive power compensation for the inductive load;

[0163] The early warning judgment module 500 is used to obtain the operating data of the power supply network after reactive power compensation, take the operating data of the inductive load as input, execute the pre-built power compensation model, obtain the power health of the inductive load in the dynamic monitoring network, and judge whether the power health of the inductive load is greater than the compensation threshold. If it is greater than the compensation threshold, a control completion instruction is generated; if it is less than the compensation threshold, a early warning instruction is triggered.

[0164] During operation, the data acquisition module 100 obtains the operation data of the power supply network, uploads the operation data of the power supply network to the power supply network database, constructs the power supply network infrastructure and the dynamic monitoring network of the power supply network based on the power supply network database and the operation data of the power supply network, and then the health calculation module 200 obtains the historical operation data of the power supply network in the power supply network database, constructs a power compensation model based on the historical operation data of the power supply network, extracts the operation data of the inductive load in the dynamic monitoring network, takes the operation data of the inductive load as input, executes the pre-constructed power compensation model, and obtains the power health of the inductive load in the dynamic monitoring network. The priority compensation determination module 300 determines the compensation method of the inductive load based on the power health of the inductive load in the dynamic monitoring network and the pre-constructed compensation priority model, and generates a reactive compensation instruction. The compensation response module 400 responds to the reactive compensation instruction, retrieves the operation data of the capacitive load in the dynamic monitoring network, controls the on and off of the capacitive load, and performs reactive compensation for the inductive load.

[0165] The embodiment of the present invention constructs a power compensation model that accurately evaluates the inductive load of the power supply grid to extract the operating data of the inductive load in the dynamic monitoring network, and can obtain accurate power health of the inductive load in the dynamic monitoring network, thereby accurately compensating for the harmonics and reactive currents in the power supply grid whose frequency and amplitude both change. Compared with the prior art, it avoids the waste of power resources, has lower energy consumption, and can also avoid damage to the inductive load during reactive compensation.

[0166] On the other hand, an embodiment of the present invention further provides a computer-readable storage medium, which stores computer program instructions, and the computer program instructions can be executed by a processor. When the computer program instructions are executed, the method of any one of the above embodiments is implemented.

[0167] In another aspect of the present invention, a computer device is provided. The computer device includes a memory and a processor. The memory stores a computer program. When the computer program is executed by the processor, the method of any one of the above embodiments is implemented.

[0168] As a non-volatile computer-readable storage medium, the memory can be used to store non-volatile software programs, non-volatile computer executable programs and modules, such as the program instructions / modules corresponding to the power supply network reactive power compensation control method in the embodiment of the present application. The memory may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created by the use of the power supply network reactive power compensation control method, etc. In addition, the memory may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some embodiments, the memory may optionally include a memory remotely arranged relative to the processor, and these remote memories may be connected to the local module via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0169] Finally, it should be noted that the computer-readable storage medium (e.g., memory) herein may be a volatile memory or a non-volatile memory, or may include both a volatile memory and a non-volatile memory. As an example and not by way of limitation, a non-volatile memory may include a read-only memory (ROM), a programmable ROM (PROM), an electrically programmable ROM (EPROM), an electrically erasable programmable ROM (EEPROM), or a flash memory. A volatile memory may include a random access memory (RAM), which may act as an external cache memory. As an example and not by way of limitation, RAM may be obtained in a variety of forms, such as synchronous RAM (DRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), and direct Rambus RAM (DRRAM). The storage devices of the disclosed aspects are intended to include, but are not limited to, these and other suitable types of memory.

[0170] The various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure herein may be implemented or executed using the following components designed to perform the functions herein: a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination of these components. A general-purpose processor may be a microprocessor, but alternatively, the processor may be any conventional processor, controller, microcontroller, or state machine. The processor may also be implemented as a combination of computing devices, for example, a combination of a DSP and a microprocessor, a plurality of microprocessors, one or more microprocessors in conjunction with a DSP, and / or any other such configuration.

[0171] In summary, the present invention provides a reactive power compensation device and control method for a power supply network. The embodiment of the present invention extracts the operating data of the inductive load in the dynamic monitoring network by constructing a power compensation model that accurately evaluates the inductive load of the power supply network, and can obtain accurate power health of the inductive load in the dynamic monitoring network, thereby accurately compensating for the harmonics and reactive currents in which the frequency and amplitude of the inductive load in the power supply network change. Compared with the prior art, it avoids the waste of power resources, has lower energy consumption, and can also avoid damage to the inductive load during reactive power compensation.

[0172] It should be noted that, for the above-mentioned embodiments, for the sake of simplicity, they are all described as a series of action combinations, but those skilled in the art should know that the present invention is not limited by the described order of actions, because according to the present invention, some steps may be performed in other orders or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present invention.

[0173] In the several embodiments provided in the present application, it should be understood that the disclosed device can be implemented in other ways. For example, the device embodiments described above are only schematic, such as the division of the above-mentioned units, which is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or communication connection shown or discussed can be through some interfaces, and the indirect coupling or communication connection between devices or units can be in the form of telecommunication or other forms.

[0174] The units described above as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0175] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the scope of protection of the invention. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on these embodiments, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in this field can still combine, add, delete or make other adjustments to the features in the various embodiments of the present invention according to the circumstances without conflict, without making creative work, so as to obtain different other technical solutions that do not deviate from the concept of the present invention in essence, and these technical solutions also belong to the scope of protection of the present invention.

Claims

1. A reactive power compensation control method for a power supply network, characterized in that: The power supply network reactive power compensation control method comprises: Acquire power grid operation data, upload the power grid operation data to the power grid database, build a power grid infrastructure based on the power grid database and a dynamic monitoring network for the power grid based on the power grid operation data, wherein the power grid operation data includes the operation data of the power supply end, the distribution end, the load end, the inductive load, the capacitive load and the resistive load in the power supply line; Obtaining historical operation data of the power supply grid in the power supply grid database, building a power compensation model based on the historical operation data of the power supply grid, extracting the operation data of the inductive load in the dynamic monitoring network, and executing the pre-built power compensation model with the operation data of the inductive load as input to obtain the power health of the inductive load in the dynamic monitoring network; Determine the inductive load compensation mode based on the power health of the inductive load in the dynamic monitoring network and the pre-built compensation priority model, and generate a reactive power compensation instruction; In response to the reactive power compensation instruction, the operating data of the capacitive load in the dynamic monitoring network is retrieved, the capacitive load is controlled to be turned on and off, and reactive power compensation is performed on the inductive load; Obtain the operation data of the power supply grid after reactive power compensation, take the operation data of the inductive load as input, execute the pre-built power compensation model, obtain the power health of the inductive load in the dynamic monitoring network, and determine whether the power health of the inductive load is greater than the compensation threshold. If it is greater than the compensation threshold, generate a control completion instruction; if it is less than the compensation threshold, trigger an early warning instruction; The method for constructing the power compensation model specifically includes: Obtain historical operation data of the power supply network in the power supply network database, select the historical operation data using a non-uniform extraction method, and divide the historical operation data into a training set and a test set; Retrieving an initial model of a power compensation model, wherein the initial model of the power compensation model is an MLP neural network model, and setting an input layer, an output layer, and initial hyperparameters of the MLP neural network model; Taking the training set as input, executing the initial model of the power compensation model, adjusting the hyperparameters of the initial model of the power compensation model based on the activation function, and finding multiple groups of power compensation models based on the grid optimization method; Iteratively train the power compensation model for multiple rounds until convergence, stop the training, and select multiple groups of converged power compensation models using an early stopping method; Obtain a validation set to determine whether the power compensation model is overfitting, and determine the accuracy of the power compensation model based on a preset validation accuracy. If the validation accuracy is met, complete the construction of the power compensation model and output the power compensation model. The method of executing the pre-built power compensation model to obtain the method of dynamically monitoring the power health of the inductive load in the network specifically includes: Extract and input the operating data of the inductive load in the dynamic monitoring network; The power compensation model calculates the power factor of the inductive load and calculates the abnormal radiation value of the inductive load based on the operating data of the inductive load and the power factor. ; （7） in, is the number of attributes of the inductive load operating data, and For the input inductive load operating data, is the power factor of the inductive load, is the radiation correlation coefficient, which is obtained through the radiation function; （8） Among them, the is the number of inductive loads, and is the capacitive load quantity, is the number of resistive loads, is the load influence coefficient; Correcting the power factor based on the abnormal radiation value to obtain a power factor correction value; （9） in, is the power factor of the inductive load, is the radiation correlation coefficient; Loading the power factor correction value and the abnormal radiation value of the inductive load, taking the power factor correction value and the abnormal radiation value of the inductive load as input, executing the power compensation model, and obtaining the power health of the inductive load in the dynamic monitoring network; （10） in, is the number of attributes of the inductive load operating data, and For the input inductive load operating data, Abnormal radiation value.

2. The reactive power compensation control method for a power supply network according to claim 1, characterized in that: The method for constructing a power supply grid infrastructure based on a power supply grid database and a power supply grid operation data based on a power supply grid dynamic monitoring network specifically includes: Traversing the power supply network database, extracting three-dimensional coordinate parameters of the inductive load, the capacitive load and the resistive load in the power supply network database; Perform three-dimensional modeling based on three-dimensional coordinate parameters to obtain an initial dynamic monitoring framework; Traversing the power grid database, extracting historical operation data of the inductive load in the power grid database, and building and training a compensation priority model based on the historical operation data; The compensation priority model determines the compensation priority coefficient of the inductive load based on the historical operation data of the inductive load, and renders the compensation priority coefficient to the initial dynamic monitoring architecture to obtain the dynamic monitoring network of the power supply grid.

3. The reactive power compensation control method of the power supply network according to claim 2, characterized in that: The method for determining the compensation priority coefficient of the inductive load based on the historical operation data of the inductive load by the compensation priority model specifically includes: Acquire historical operation data of the inductive load, and extract configuration information of the inductive load from the historical operation data of the inductive load; Based on the configuration information of the inductive load and the improved hierarchical analysis method, the weight distribution of the initial dynamic monitoring architecture is determined, and the initial weight is assigned to the inductive load; Obtain the historical power factor in the historical operation data of the inductive load, and determine the compensation rate and compensation level of the inductive load based on the historical power factor; The calculation formula of the compensation rate is: （1） Among them, the is the compensation rate of the inductive load, and Data collection time period for inductive loads The number of inductive load acquisitions within It is the number of inductive load acquisitions when the historical power factor of the inductive load is 1.

4. The reactive power compensation control method for a power supply network as claimed in claim 3, characterized in that: The method for determining the compensation priority coefficient of the inductive load based on the historical operation data of the inductive load by the compensation priority model specifically includes: Obtain the initial weight, compensation rate and compensation level assigned to the inductive load, normalize the compensation rate and compensation level based on the dimensional effect and normalization function to eliminate the index, and obtain the normalized priority index value ; （2） Among them, the is a normalized vector, and They represent the pre-coefficient and post-coefficient of the normalization function respectively. is the compensation rate of the inductive load, Assign initial weights to the inductive loads; （3） in, They represent the pre-coefficient and post-coefficient of the normalization function respectively. is the normalized curvature of the normalized function; Acquire multiple groups of normalized priority index values, and calculate the inductive load compensation comparison value based on the normalized priority index values ​​and the compensation comparison function; （4） in, represents the normalized priority index value, and is the inductive load compensation comparison value, is the number of inductive loads, Then it represents the comparison coefficient of the compensation comparison function; Calculating a compensation priority coefficient based on a comparison value of inductive load compensation; （5）； Among them, the represents the compensation priority factor of the inductive load, and is the number of inductive loads, Compare the mean value of load compensation for inductive loads; （6） Among them, the represents the compensation priority factor of the inductive load, and is the number of inductive loads.

5. The reactive power compensation control method of the power supply network according to claim 2, characterized in that: Determining the inductive load compensation method based on the power health of the inductive load in the dynamic monitoring network and the pre-built compensation priority model specifically includes: Obtain the power health of inductive loads in the dynamic monitoring network; Determine whether the inductive load needs reactive power compensation and generate a queue to be compensated; The compensation priority model identifies the running data of the queue to be compensated; The priority coefficients of the inductive loads in the dynamic monitoring network are retrieved, and a compensation priority queue is generated based on the priority coefficients.

6. A reactive power compensation device for a power supply network, implemented by the reactive power compensation control method for a power supply network as claimed in any one of claims 2 to 5, characterized in that: The power supply network reactive power compensation device specifically comprises: A data acquisition module, wherein the data acquisition module is used to obtain power grid operation data, upload the power grid operation data to a power grid database, and construct a power grid infrastructure based on the power grid database and a dynamic monitoring network for the power grid based on the power grid operation data; A health calculation module, which is used to obtain historical operation data of the power supply network in the power supply network database, build a power compensation model based on the historical operation data of the power supply network, extract the operation data of the inductive load in the dynamic monitoring network, and use the operation data of the inductive load as input to execute the pre-built power compensation model to obtain the power health of the inductive load in the dynamic monitoring network; A priority compensation determination module, wherein the priority compensation determination module determines the inductive load compensation mode based on the power health of the inductive load in the dynamic monitoring network and a pre-built compensation priority model, and generates a reactive compensation instruction; A compensation response module, which responds to a reactive power compensation instruction, retrieves the capacitive load operation data in the dynamic monitoring network, controls the capacitive load on and off, and performs reactive power compensation for the inductive load; An early warning judgment module is used to obtain the operating data of the power supply network after reactive power compensation, take the operating data of the inductive load as input, execute the pre-built power compensation model, obtain the power health of the inductive load in the dynamic monitoring network, and judge whether the power health of the inductive load is greater than the compensation threshold. If it is greater than the compensation threshold, a control completion instruction is generated; if it is less than the compensation threshold, a early warning instruction is triggered.

Images