Power grid instantaneous power fluctuation compensation control method, system, equipment and medium

Abstract

The invention relates to the field of power systems, in particular to a power grid instantaneous power fluctuation compensation control method, system and device and a medium. The method comprises the following steps: acquiring key operation data of a power grid in real time by using a high-precision synchronous acquisition technology; preprocessing the collected key operation data to obtain preprocessed data, and identifying key characteristics of power grid operation in the preprocessed data through a characteristic extraction technology to obtain a key characteristic set; a pre-trained deep learning model is adopted to learn and analyze the key feature set, instantaneous power fluctuation in the power grid is predicted, and a prediction result is obtained; and inputting the prediction result into a preset adaptive compensation control strategy, and dynamically adjusting control parameters according to the real-time state of the power grid. According to the method, the prediction accuracy of the instantaneous power fluctuation is improved. The control parameters are dynamically adjusted through the prediction result, the instantaneous power fluctuation in the power grid is effectively compensated, and the stability and reliability of the power grid are enhanced.

Description

Technical Field

[0001] The present invention relates to the field of electric power systems, and in particular to a method, system, equipment and medium for compensating and controlling instantaneous power fluctuations in a power grid. Background Art

[0002] In modern power systems, the stability of the power grid is crucial to ensuring the reliability and safety of power supply. With the rapid development of industrialization and informatization, the power grid is facing more and more challenges, especially the problem of instantaneous power fluctuations in the power grid. These problems are usually caused by factors such as rapid changes in grid load, unstable output of renewable energy, and sudden failures of grid equipment. Instantaneous power fluctuations may cause voltage sags, equipment overloads, and even grid collapse, seriously affecting the stability of the power system and the quality of power supply.

[0003] Existing grid power fluctuation compensation methods, such as the ramp-type gravity energy storage instantaneous power mirror compensation method described in patent document CN116706948A, mainly rely on physical energy storage systems to balance power fluctuations. However, these methods have problems such as slow response speed, limited compensation capacity, and high equipment cost. In addition, traditional compensation control strategies, such as the new energy power generation grid load fluctuation compensation method mentioned in patent document CN118117607A, are often based on simple feedback control logic, lacking in-depth understanding and prediction capabilities of the dynamic behavior of the grid, and are difficult to adapt to the complex and changeable grid operation environment.

[0004] With the development of smart grid technology, how to use advanced information and communication technology, data processing technology and artificial intelligence algorithms to achieve intelligent management and control of power grids has become a research hotspot in the field of power systems. In particular, the development of deep learning technology has provided new possibilities for the prediction and compensation of power grid power fluctuations. Deep learning models, such as convolutional neural networks (CNNs) and recurrent neural networks (RNNs), can learn the operation mode of the power grid from a large amount of historical data and predict future power fluctuations.

[0005] Therefore, how to effectively integrate deep learning technology into the power grid compensation control strategy to achieve real-time and accurate power fluctuation compensation. In addition, the nonlinear characteristics and uncertainty of the power grid require that the compensation control strategy must have a high degree of adaptability and flexibility. Existing control strategies often find it difficult to meet these requirements, and further technological innovation and improvement are needed. Therefore, a power grid instantaneous power fluctuation compensation control method, system, device and medium are urgently needed. Summary of the invention

[0006] The purpose of the present invention is to provide a method, system, device and medium for controlling instantaneous power fluctuation compensation of a power grid, so as to solve the problems of low efficiency and slow response speed of power fluctuation compensation of the power grid in the prior art.

[0007] To achieve the above object, the present invention adopts the following technical solutions:

[0008] A method for compensating and controlling instantaneous power fluctuations in a power grid comprises the following steps:

[0009] S1. Use high-precision synchronous acquisition technology to collect key operation data of the power grid in real time, including voltage, current, frequency and load changes;

[0010] S2. Preprocessing the collected key operation data, including applying wavelet transform and filtering algorithm to remove noise, obtaining preprocessed data, identifying key features of power grid operation in the preprocessed data through feature extraction technology, and obtaining a key feature set;

[0011] S3, using a pre-trained deep learning model to learn and analyze the key feature set, predict the instantaneous power fluctuation in the power grid, and obtain the prediction result;

[0012] S4. Input the prediction results into the preset adaptive compensation control strategy and dynamically adjust the control parameters according to the real-time status of the power grid.

[0013] Optionally, step S1 includes the following steps:

[0014] S1.1. Configure a group of synchronous acquisition modules, each module includes a voltage sensor, a current sensor, a frequency sensor, and a load monitor, all of which are connected to a high-precision clock synchronization device to synchronize timestamps;

[0015] S1.2, transmitting the output of the synchronous acquisition module to a central data processor via an optical fiber communication network;

[0016] S1.3, implement data acquisition management software in the central data processor to control the startup, synchronization and real-time collection of data from the data acquisition unit;

[0017] Optional, pre-training steps for deep learning models, including:

[0018] Constructing a deep learning model, including at least one convolutional neural network sub-model and one recurrent neural network sub-model, for processing the time series and spatial characteristics of the power grid operation data;

[0019] The convolutional neural network sub-model is used to extract the spatial features of the power grid operation data and obtain the spatial feature representation;

[0020] Use the recurrent neural network sub-model to extract the time series features of the data and obtain the time feature representation;

[0021] The obtained spatial feature representation and temporal feature representation are fused to form a comprehensive feature representation;

[0022] Using the fused comprehensive feature representation, feature integration is performed through the fully connected layer to obtain the integrated feature vector;

[0023] The integrated feature vector is input into the output layer, and the instantaneous power fluctuation in the power grid is predicted based on the mapping relationship learned during the training process to obtain the prediction result;

[0024] Based on the prediction results, the deep reinforcement learning method is used to pre-train the deep learning model, where the deep reinforcement learning algorithm optimizes the model's prediction strategy based on historical power fluctuation data;

[0025] During the pre-training process, the power grid operation experience provided by the expert system is used as a reward signal to guide the training direction of the deep reinforcement learning algorithm, so that the deep learning model can make accurate power fluctuation predictions under various power grid conditions and obtain a trained deep learning model.

[0026] Optionally, based on the model predictive control principle, an adaptive compensation control strategy is designed, which specifically includes the following steps:

[0027] Determine the control objectives and constraints for grid operation, including but not limited to standards for voltage stability, power balance, and safe operation of equipment;

[0028] Based on the determined control objectives and constraints, a predictive control mathematical model is constructed to describe the dynamic behavior of power grid operation;

[0029] The prediction results provided by the deep learning model are used as the input of the predictive control mathematical model to perform real-time optimization calculations and predict the changing trend of the future power grid state. Based on the optimization calculation results of the predictive control mathematical model, a series of compensating control actions are formulated, which are aimed at adjusting the operating parameters of the power grid equipment to cope with the predicted instantaneous power fluctuations. The formulated compensating control actions are converted into control signals and sent to the key control nodes of the power grid to perform real-time adjustments. The post-execution power grid operation data is collected through the real-time monitoring system to evaluate the effect of the compensating control actions, and the predictive control mathematical model is adjusted online according to the difference between the actual power grid state and the predicted state. This step is repeated to continuously compensate and control the instantaneous power fluctuations of the power grid to ensure the stable operation of the power grid.

[0030] Optionally, it also includes an adaptive compensation control strategy combined with a fuzzy logic controller for compensation control, specifically including:

[0031] Based on the historical operation data of the power grid and expert experience, the initial parameters of the fuzzy logic controller are set to define the fuzzy sets and inference rules;

[0032] Monitor the operating status of the power grid in real time, including voltage, current, frequency and load changes, and collect real-time data for real-time adjustment of the fuzzy logic controller; input the collected real-time data into the fuzzy logic controller, and convert the precise input data into fuzzy values ​​through fuzzification processing; use the fuzzy inference system in the fuzzy logic controller to perform inference calculations according to predefined fuzzy rules and fuzzy values ​​to obtain fuzzy control outputs; defuzzify the obtained fuzzy control outputs and convert them into specific control instructions; dynamically adjust the operating parameters of the power grid equipment according to the obtained control instructions, and implement adaptive compensation control strategies; after executing the control instructions, collect feedback data from the power grid, evaluate the control effect, and adaptively adjust the parameters of the fuzzy logic controller according to the actual power grid response; repeat this step to form a closed-loop control system that continuously adapts to the nonlinear characteristics and uncertainties of the power grid.

[0033] Optionally, it also includes deploying edge computing devices at key nodes of the power grid to perform localized processing and real-time analysis of data, and the decision results of the edge computing devices are fed back to the deep learning model to optimize the model prediction.

[0034] Optionally, the following steps are also included:

[0035] Collect real-time and historical data on power grid operation, as well as decision results provided by edge computing devices and prediction results of deep learning models;

[0036] Use edge computing devices to perform preliminary processing on real-time data, including data cleaning, formatting, and feature extraction, to generate data sets that can be used in intelligent decision support systems;

[0037] Input the processed data set into an intelligent decision support system that integrates expert systems and optimization algorithms;

[0038] In the intelligent decision support system, the knowledge base and reasoning engine of the expert system are applied to analyze the data set, identify the key characteristics of power grid operation and potential power fluctuation risks, and obtain the first analysis results;

[0039] Combined with the prediction results of the deep learning model, the intelligent decision support system uses optimization algorithms, including genetic algorithms or particle swarm optimization, to evaluate and select possible grid regulation solutions to obtain a second analysis result;

[0040] The intelligent decision support system generates an optimal power grid regulation plan based on the first analysis result and the second analysis result, including real-time data, historical data, prediction results and expert experience;

[0041] Output the optimal adjustment plan to the power grid control system to guide the real-time adjustment of power grid equipment, compensate for instantaneous power fluctuations and maintain power grid stability;

[0042] Monitor the implementation effect of the optimal adjustment plan, collect feedback data, and use the feedback data for model updating and optimization of the intelligent decision support system.

[0043] A power grid instantaneous power fluctuation compensation control system, comprising:

[0044] Data acquisition module, used to collect key operation data of the power grid in real time, including voltage, current, frequency and load changes, using high-precision synchronous acquisition technology;

[0045] The data processing module is used to pre-process the collected key operation data, including applying wavelet transform and filtering algorithm to remove noise to obtain pre-processed data, and identifying the key features of power grid operation in the pre-processed data through feature extraction technology to obtain a key feature set;

[0046] The model prediction module is used to learn and analyze the key feature set using a pre-trained deep learning model to predict the instantaneous power fluctuation in the power grid and obtain the prediction results;

[0047] The compensation strategy control module is used to input the prediction results into the preset adaptive compensation control strategy and dynamically adjust the control parameters according to the real-time status of the power grid.

[0048] An electronic device comprises a processor and a memory, wherein the processor is used to execute a computer program stored in the memory to implement a power grid instantaneous power fluctuation compensation control method.

[0049] A computer-readable storage medium stores at least one instruction. When the at least one instruction is executed by a processor, a method for compensating and controlling instantaneous power fluctuations in a power grid is implemented.

[0050] Compared with the prior art, the present invention has the following beneficial effects:

[0051] The present invention improves the prediction accuracy of instantaneous power fluctuations by real-time acquisition and preprocessing of power grid operation data, combined with the analytical capabilities of deep learning models. The control parameters are dynamically adjusted based on the prediction results, effectively compensating for instantaneous power fluctuations in the power grid, and enhancing the stability and reliability of the power grid. The adaptive compensation control strategy can adjust resource allocation according to the real-time status of the power grid, optimizing the operation efficiency and energy utilization of the power grid.

[0052] The present invention ensures that the data collected from multiple sensors have time consistency through a high-precision clock synchronization device, providing an accurate data basis for subsequent analysis. The use of optical fiber communication networks to transmit data reduces the delay of data during transmission and improves the real-time nature of data processing. The data acquisition management software in the central data processor can effectively control the startup, synchronization and real-time collection of data acquisition units, thereby improving the overall efficiency of data processing.

[0053] The present invention improves the generalization ability of the model for predicting power fluctuations under different power grid conditions by combining historical data and expert knowledge through transfer learning technology. The deep reinforcement learning method optimizes the prediction strategy of the model based on historical power fluctuation data, so that the model can more accurately predict future power fluctuations. The power grid operation experience provided by the expert system is used as a reward signal, so that the model training process can better reflect the complexity of actual power grid operation.

[0054] The present invention dynamically sets the control target according to the control target and constraint conditions of the power grid operation, making the control strategy more flexible and adaptable. The prediction results of the deep learning model are used for real-time optimization calculation to predict the changing trend of the future power grid state, thereby improving the foresight and effectiveness of the control strategy. Through real-time monitoring and online adjustment, continuous compensation and control of instantaneous power fluctuations of the power grid are achieved, ensuring the long-term stable operation of the power grid.

[0055] The fuzzy logic controller of the present invention can adapt to the nonlinear characteristics of the power grid and improve the effectiveness of the control strategy in the face of complex power grid environments. By real-time monitoring of the power grid operation status and dynamically adjusting the parameters of the fuzzy logic controller, the control strategy can quickly respond to real-time changes in the power grid. A closed-loop control system is formed, and through continuous feedback and adjustment, the instantaneous power fluctuation compensation effect of the power grid is optimized.

[0056] The edge computing device of the present invention processes data locally, reducing the delay of data transmission and processing and improving the response speed. The edge computing device provides real-time data analysis capabilities, enabling the control strategy to make decisions based on the latest data. The decision results of the edge computing device are fed back to the deep learning model, optimizing the prediction performance of the model and improving the accuracy of the prediction.

[0057] The present invention integrates expert systems and optimization algorithms to automate the decision-making process, reduce human intervention, and improve the efficiency and accuracy of decision-making. Comprehensive analysis is performed using the results of edge computing devices and deep learning models to provide the optimal grid regulation solution, enhancing the scientificity and effectiveness of grid regulation. By collecting feedback data on implementation effects for model updating and optimization of the intelligent decision support system, the system can continuously adapt to changes in grid operation, improving the system's adaptability and long-term stability. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 The present invention is a flowchart of an embodiment of a method for compensating for instantaneous power fluctuations in a power grid.

[0059] Figure 2 The present invention is a schematic diagram of a module of a power grid instantaneous power fluctuation compensation control system embodiment of the present invention.

[0060] Figure 3 A schematic flow chart of an adaptive compensation control strategy is designed for an instantaneous power fluctuation compensation control system embodiment of the present invention.

[0061] Figure 4 The present invention is a schematic diagram of a combined control process of an instantaneous power fluctuation compensation control system for a power grid.

[0062] Figure 5 The present invention is a structural block diagram of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0063] The present invention will be described in detail below with reference to the accompanying drawings and in combination with embodiments. It should be noted that the embodiments and features in the embodiments of the present application can be combined with each other without conflict.

[0064] The following detailed description is an exemplary description, which is intended to provide further detailed description of the present invention. Unless otherwise specified, all technical terms used in the present invention have the same meaning as those generally understood by those skilled in the art to which the present application belongs. The terms used in the present invention are only for describing specific embodiments, and are not intended to limit exemplary embodiments according to the present invention.

[0065] Example 1

[0066] like Figure 1 As shown, a method for compensating and controlling instantaneous power fluctuations in a power grid comprises the following steps:

[0067] S1. Use high-precision synchronous acquisition technology to collect key operation data of the power grid in real time, including voltage, current, frequency and load changes;

[0068] S2. Preprocessing the collected key operation data, including applying wavelet transform and filtering algorithm to remove noise, obtaining preprocessed data, identifying key features of power grid operation in the preprocessed data through feature extraction technology, and obtaining a key feature set;

[0069] S3, using a pre-trained deep learning model to learn and analyze the key feature set, predict the instantaneous power fluctuation in the power grid, and obtain the prediction result;

[0070] S4. Input the prediction results into the preset adaptive compensation control strategy and dynamically adjust the control parameters according to the real-time status of the power grid.

[0071] Specifically, an optical fiber-based communication system and a high-precision clock synchronization device, such as a GPS receiver, are used to ensure accurate synchronization of the timestamps of all acquisition points. By configuring a set of synchronous acquisition modules, each module includes a voltage sensor, a current sensor, a frequency sensor, and a load monitor, real-time acquisition of key operating parameters of the power grid is achieved.

[0072] Wavelet transform and filtering algorithms are used to denoise the collected data to improve data quality. Through feature extraction techniques such as principal component analysis (PCA) or independent component analysis (ICA), key features of power grid operation are identified from pre-processed data to form a key feature set.

[0073] A composite deep learning model is built, combining convolutional neural networks (CNN) and recurrent neural networks (RNN), especially long short-term memory networks (LSTM), to learn and analyze key feature sets. Model training uses historical data sets and optimizes weights through back-propagation algorithms to improve prediction accuracy.

[0074] The trained deep learning model is applied to real-time data to predict instantaneous power fluctuations in the power grid. The prediction results are input into an adaptive compensation control strategy, which is designed based on the principle of model predictive control (MPC) and dynamically adjusts control parameters to achieve a fast response to instantaneous power fluctuations.

[0075] Combined with the fuzzy logic controller, the adaptive compensation control strategy is refined and optimized. The parameters of the fuzzy logic controller are adaptively adjusted according to the real-time data and historical operation data of the power grid to adapt to the nonlinear characteristics and uncertainties of the power grid.

[0076] Edge computing devices are deployed at key nodes of the power grid to achieve local processing and real-time analysis of data. The decision results of edge computing devices are fed back to the deep learning model to optimize model predictions.

[0077] Develop an intelligent decision support system that integrates expert systems and optimization algorithms to automate and intelligentize the decision-making process. The system uses the decision results of edge computing devices and the prediction results of deep learning models for comprehensive analysis and decision-making.

[0078] Build a unified power grid management platform that integrates data collection, processing, prediction, control and decision support modules. The platform can automatically adapt to the dynamic changes of the power grid and the influence of the external environment, and guide the overall operation and maintenance of the power grid.

[0079] Through the above-mentioned specific implementation methods, the present invention can achieve accurate prediction and effective compensation for instantaneous power fluctuations in the power grid, thereby improving the stability and power supply quality of the power grid.

[0080] Example 2

[0081] As a preferred example, step S1 includes the following steps:

[0082] S1.1. Configure a group of synchronous acquisition modules, each module includes a voltage sensor, a current sensor, a frequency sensor, and a load monitor, all of which are connected to a high-precision clock synchronization device to synchronize the timestamp to ensure the time consistency of data acquisition;

[0083] S1.2, transmitting the output of the synchronous acquisition module to the central data processor through the optical fiber communication network; utilizing the high bandwidth and low latency characteristics of optical fiber to ensure the real-time and reliability of data transmission;

[0084] S1.3. Implement data acquisition management software in the central data processor to control the startup, synchronization and real-time data collection of the data acquisition unit; the software manages the work of each acquisition module through a preset synchronization algorithm to ensure the synchronization and integrity of the data; and performs preliminary processing on the data received by the central data processor, including calibration of the timestamp and unification of the data format, so as to carry out subsequent data analysis and feature extraction;

[0085] A high-precision clock synchronization device is used to accurately control the sampling time of each acquisition module. Through the synchronization of timestamps, high-precision synchronous acquisition of multi-channel data is achieved, where the synchronization accuracy reaches the microsecond level to meet the needs of high-precision synchronization of power grid data.

[0086] Example 3

[0087] As a preferred example, the pre-training steps of the deep learning model specifically include:

[0088] Constructing a deep learning model, including at least one convolutional neural network sub-model and one recurrent neural network sub-model, for processing the time series and spatial characteristics of the power grid operation data;

[0089] The convolutional neural network sub-model is used to extract the spatial features of the power grid operation data and obtain the spatial feature representation;

[0090] Use the recurrent neural network sub-model to extract the time series features of the data and obtain the time feature representation;

[0091] The obtained spatial feature representation and temporal feature representation are fused to form a comprehensive feature representation;

[0092] Using the fused comprehensive feature representation, feature integration is performed through the fully connected layer to obtain the integrated feature vector;

[0093] The integrated feature vector is input into the output layer, and the instantaneous power fluctuation in the power grid is predicted based on the mapping relationship learned during the training process to obtain the prediction result;

[0094] Based on the prediction results, the deep reinforcement learning method is used to pre-train the deep learning model, where the deep reinforcement learning algorithm optimizes the model's prediction strategy based on historical power fluctuation data;

[0095] During the pre-training process, the power grid operation experience provided by the expert system is used as a reward signal to guide the training direction of the deep reinforcement learning algorithm, so that the deep learning model can make accurate power fluctuation predictions under various power grid conditions and obtain a trained deep learning model.

[0096] Specifically, a deep learning model is constructed, including at least one convolutional neural network (CNN) sub-model and a recurrent neural network (RNN) sub-model, wherein the CNN sub-model is used to process the spatial characteristics of the power grid operation data, and the RNN sub-model is used to process the time series characteristics, so as to form a comprehensive representation of the power grid operation status;

[0097] The CNN sub-model is used to extract the spatial features of the power grid operation data. The spatial feature representation is obtained through the operation of the convolution layer and the pooling layer. The weight matrix W of the convolution layer is c and the bias vector b c By formula W c *x+b c Perform feature mapping, and the pooling layer reduces the dimension of the features;

[0098] The RNN sub-model, especially the long short-term memory network (LSTM), is used to extract the time series features of the data and obtain the time feature representation, where the state update formula of the LSTM unit is ht = LSTM (x t ,h t-1 ,c t-1 ), where x t is the input, h t is the output, c t is the cell state;

[0099] The obtained spatial feature representation and temporal feature representation are fused to form a comprehensive feature representation, and the weight matrix W of the feature fusion layer is used to f and the bias vector bf using the formula W f *[spatial_features,temporal_features]+b f Perform feature integration;

[0100] Using the fused comprehensive feature representation, the features are integrated through the fully connected layer to obtain the integrated feature vector, where the output formula of the fully connected layer is z = W d *features+b d , W d is the weight matrix of the fully connected layer, b d is the bias vector;

[0101] The integrated feature vector is input into the output layer, and the instantaneous power fluctuation in the power grid is predicted based on the mapping relationship learned during the training process to obtain the prediction result, where the prediction formula of the output layer is y^=softmax(W o *z+b o ), W o is the weight matrix of the output layer, b o is the bias vector;

[0102] Based on the prediction results, the deep reinforcement learning method is used to pre-train the deep learning model, where the deep reinforcement learning algorithm optimizes the model's prediction strategy based on historical power fluctuation data;

[0103] During the pre-training process, the power grid operation experience provided by the expert system is used as a reward signal to guide the training direction of the deep reinforcement learning algorithm, so that the deep learning model can make accurate power fluctuation predictions under various power grid conditions and obtain a trained deep learning model.

[0104] Example 4

[0105] like Figure 3 As shown, as a preferred example, based on the model predictive control principle, an adaptive compensation control strategy is designed, which specifically includes the following steps:

[0106] Determine the control objectives and constraints for grid operation, including but not limited to standards for voltage stability, power balance, and safe operation of equipment;

[0107] Based on the determined control objectives and constraints, a predictive control mathematical model is constructed to describe the dynamic behavior of power grid operation;

[0108] The prediction results provided by the deep learning model are used as the input of the predictive control mathematical model to perform real-time optimization calculations and predict the changing trend of the future power grid state. Based on the optimization calculation results of the predictive control mathematical model, a series of compensating control actions are formulated, which are aimed at adjusting the operating parameters of the power grid equipment to cope with the predicted instantaneous power fluctuations. The formulated compensating control actions are converted into control signals and sent to the key control nodes of the power grid to perform real-time adjustments. The post-execution power grid operation data is collected through the real-time monitoring system to evaluate the effect of the compensating control actions, and the predictive control mathematical model is adjusted online according to the difference between the actual power grid state and the predicted state. This step is repeated to continuously compensate and control the instantaneous power fluctuations of the power grid to ensure the stable operation of the power grid.

[0109] Specifically, the control objectives and constraints of grid operation are determined, including but not limited to standards for voltage stability, power balance and safe operation of equipment, to ensure the stability and reliability of the grid under various operating conditions.

[0110] Based on the determined control objectives and constraints, a predictive control mathematical model is constructed, which uses mathematical formulas and algorithms to describe the dynamic behavior of the power grid, including but not limited to the load flow, voltage changes and frequency response of the power grid.

[0111] The prediction results provided by the deep learning model are used as the input of the predictive control mathematical model to perform real-time optimization calculations to predict the changing trend of the future power grid state. The optimization calculation uses an iterative algorithm, such as the gradient descent method, to solve the optimal control strategy. According to the optimization calculation results of the predictive control mathematical model, a series of compensation control actions are formulated. These actions are aimed at adjusting the operating parameters of the power grid equipment, such as the transformer tap position, generator output and load distribution, to cope with the predicted instantaneous power fluctuations. The formulated compensation control actions are converted into control signals and sent to the key control nodes of the power grid to perform real-time adjustments. The control signals are transmitted to the execution devices through digital communication protocols, such as IEC 61850. The power grid operation data after execution is collected through the real-time monitoring system to evaluate the effect of the compensation control actions. According to the difference between the actual power grid state and the predicted state, the predictive control mathematical model is adjusted online to improve the accuracy and response speed of the control strategy. Repeat this step to continuously compensate and control the instantaneous power fluctuations of the power grid to ensure the stable operation of the power grid. This cycle adopts an adaptive learning mechanism so that the control strategy can be continuously optimized according to the actual operation of the power grid.

[0112] The adaptive compensation control strategy based on the model predictive control principle includes the steps of determining the control target, building a mathematical model, real-time optimization calculation, formulating compensation control actions, converting control signals, real-time monitoring and online adjustment, etc. These steps together constitute a complete control process, ensuring the effective compensation of instantaneous power fluctuations in the power grid and the stable operation of the power grid.

[0113] Example 5

[0114] like Figure 4 As shown, as a preferred example, it also includes an adaptive compensation control strategy combined with a fuzzy logic controller to perform compensation control, specifically including:

[0115] Based on the historical operation data of the power grid and expert experience, the initial parameters of the fuzzy logic controller are set to define the fuzzy sets and inference rules;

[0116] Monitor the operating status of the power grid in real time, including voltage, current, frequency and load changes, and collect real-time data for real-time adjustment of the fuzzy logic controller; input the collected real-time data into the fuzzy logic controller, and convert the precise input data into fuzzy values ​​through fuzzification processing; use the fuzzy inference system in the fuzzy logic controller to perform inference calculations according to predefined fuzzy rules and fuzzy values ​​to obtain fuzzy control outputs; defuzzify the obtained fuzzy control outputs and convert them into specific control instructions; dynamically adjust the operating parameters of the power grid equipment according to the obtained control instructions, and implement adaptive compensation control strategies; after executing the control instructions, collect feedback data from the power grid, evaluate the control effect, and adaptively adjust the parameters of the fuzzy logic controller according to the actual power grid response; repeat this step to form a closed-loop control system that continuously adapts to the nonlinear characteristics and uncertainties of the power grid.

[0117] Specifically,

[0118] Based on the historical operation data of the power grid and expert experience, the initial parameters of the fuzzy logic controller are set to define fuzzy sets and inference rules, where the membership function of the fuzzy set is adjusted according to historical data to adapt to the nonlinear characteristics and uncertainty of the power grid;

[0119] Real-time monitoring of the operating status of the power grid, including voltage, current, frequency and load changes, and collection of real-time data for real-time adjustment of the fuzzy logic controller. These data are obtained through high-precision synchronous acquisition technology to ensure the accuracy and timeliness of the data;

[0120] The collected real-time data is input into the fuzzy logic controller, and the precise input data is converted into fuzzy values ​​through fuzzification. The fuzzification process adopts a specific fuzzification method, such as Zadeh fuzzification or Mamdani fuzzification.

[0121] Using the fuzzy inference system in the fuzzy logic controller, according to the predefined fuzzy rules and fuzzy values, the inference calculation is performed to obtain the fuzzy control output. The fuzzy inference system adopts the Mamdani or Takagi-Sugeno model.

[0122] The obtained fuzzy control output is defuzzified and converted into specific control instructions. The defuzzification process adopts the central average method or the maximum membership principle;

[0123] According to the obtained control instructions, the operating parameters of the power grid equipment are dynamically adjusted to implement the adaptive compensation control strategy. The control instructions are sent to the execution equipment through the digital communication protocol, such as transformer tap position, generator output and load distribution;

[0124] After executing the control command, the feedback data of the power grid is collected to evaluate the control effect, and the changes in the operation status of the power grid are monitored through the real-time monitoring system. The feedback data is used to evaluate the effectiveness of the control strategy and the stability of the power grid;

[0125] The parameters of the fuzzy logic controller are adaptively adjusted according to the actual power grid response. The adjustment process is based on the evaluation of the control effect and the adaptive algorithm is used to optimize the fuzzy sets and inference rules.

[0126] Repeat steps 2 to 8 above to form a closed-loop control system that continuously adapts to the nonlinear characteristics and uncertainties of the power grid. The closed-loop control system adopts iterative learning control or adaptive control strategy to achieve continuous optimization and adaptive adjustment.

[0127] The specific implementation method of adaptive compensation control strategy combined with fuzzy logic controller for compensation control includes the steps of initial parameter setting of fuzzy logic controller, real-time data collection, fuzzification processing, fuzzy reasoning calculation, defuzzification processing, control instruction execution, feedback data collection and adaptive adjustment. These steps together constitute a complete control process, ensuring the effective compensation of instantaneous power fluctuations of the power grid and the stable operation of the power grid.

[0128] Example 6

[0129] As a preferred example, it also includes, specifically, deploying edge computing devices at key nodes of the power grid to perform localized processing and real-time analysis of data, and the decision results of the edge computing devices are fed back to the deep learning model to optimize model predictions.

[0130] Specifically, edge computing devices are deployed at key nodes of the power grid, such as substations, distribution rooms, or user-side load access points. These devices have data processing and analysis capabilities and can execute real-time control strategies;

[0131] The edge computing device receives the real-time collected power grid operation data through the communication interface with the synchronous acquisition module, and performs local processing on the data, including data cleaning, formatting and preliminary analysis;

[0132] Utilize the computing power of edge computing devices to analyze the received data in real time, identify potential problems in power grid operation, and generate control instructions based on preset control logic;

[0133] Based on the real-time analysis results, the edge computing device directly controls the connected grid equipment through its control interface, such as adjusting the transformer tap position, controlling the generator output, or adjusting the load distribution to compensate for the predicted instantaneous power fluctuations;

[0134] Edge computing devices feed back their decision results and executed control actions to the central data processor for retraining and optimization of deep learning models to improve prediction accuracy and control effects;

[0135] Through a continuous feedback loop, edge computing devices continuously adjust their control strategies to adapt to changes in the grid's operating status and ensure stable grid operation.

[0136] These devices not only perform local processing and real-time analysis of data, but also directly participate in the real-time control of the power grid, forming a closed-loop control system with deep learning models.

[0137] Example 7

[0138] As a preferred example, the method further includes the following steps:

[0139] Collect real-time and historical data on power grid operation, as well as decision results provided by edge computing devices and prediction results of deep learning models;

[0140] Use edge computing devices to perform preliminary processing on real-time data, including data cleaning, formatting, and feature extraction, to generate data sets that can be used in intelligent decision support systems;

[0141] Input the processed data set into an intelligent decision support system that integrates expert systems and optimization algorithms;

[0142] In the intelligent decision support system, the knowledge base and reasoning engine of the expert system are applied to analyze the data set, identify the key characteristics of power grid operation and potential power fluctuation risks, and obtain the first analysis results;

[0143] Combined with the prediction results of the deep learning model, the intelligent decision support system uses optimization algorithms, including genetic algorithms or particle swarm optimization, to evaluate and select possible grid regulation solutions to obtain a second analysis result;

[0144] The intelligent decision support system generates an optimal power grid regulation plan based on the first analysis result and the second analysis result, including real-time data, historical data, prediction results and expert experience;

[0145] Output the optimal adjustment plan to the power grid control system to guide the real-time adjustment of power grid equipment, compensate for instantaneous power fluctuations and maintain power grid stability;

[0146] Monitor the implementation effect of the optimal adjustment plan, collect feedback data, and use the feedback data for model updating and optimization of the intelligent decision support system.

[0147] Specifically, real-time and historical data of power grid operation are collected, including but not limited to parameters such as voltage, current, frequency, and load changes, as well as decision results provided by edge computing devices and prediction results of deep learning models;

[0148] Use edge computing devices to perform preliminary processing on real-time data, including data cleaning, formatting, and feature extraction, to generate data sets that can be used in intelligent decision support systems. The data cleaning step includes removing outliers and filling missing data, the formatting step includes converting data into a unified data format, and the feature extraction step includes applying statistical analysis and signal processing techniques to extract key information from the data.

[0149] The processed data set is input into an intelligent decision support system that integrates an expert system and an optimization algorithm, wherein the expert system includes a knowledge base that stores expert knowledge about power grid operation and control, and the optimization algorithm includes a genetic algorithm or a particle swarm optimization for solving the optimal control strategy;

[0150] In the intelligent decision support system, the knowledge base and reasoning engine of the expert system are applied to analyze the data set, identify the key characteristics of the power grid operation and the potential power fluctuation risk, and obtain the first analysis result, wherein the reasoning engine performs logical reasoning according to the rules in the knowledge base to identify the power grid operation status;

[0151] Combined with the prediction results of the deep learning model, the intelligent decision support system uses an optimization algorithm to evaluate and select possible grid regulation solutions to obtain a second analysis result, where the optimization algorithm determines the optimal solution through iterative search;

[0152] The intelligent decision support system generates an optimal grid regulation plan based on the first analysis result and the second analysis result, including real-time data, historical data, prediction results and expert experience, wherein the optimal regulation plan takes into account the stability, economy and environmental impact of the grid;

[0153] Outputting the optimal adjustment scheme to the power grid control system to guide the real-time adjustment of power grid equipment, compensate for instantaneous power fluctuations and maintain power grid stability, wherein the output step includes converting the adjustment scheme into control instructions and sending them to the execution device;

[0154] Monitor the implementation effect of the optimal regulation plan, collect feedback data, and use the feedback data for model updating and optimization of the intelligent decision support system. The monitoring step includes real-time monitoring of the power grid status and evaluating the effect of the regulation plan, and the model updating and optimization step includes adjusting the model parameters according to the feedback data.

[0155] Example 8

[0156] like Figure 2 As shown, a power grid instantaneous power fluctuation compensation control system comprises:

[0157] A data acquisition module is used to collect key operation data of the power grid in real time, including voltage, current, frequency and load changes, using high-precision synchronous acquisition technology; wherein the synchronous acquisition technology includes but is not limited to global positioning system (GPS) time synchronization to ensure the time consistency of the data;

[0158] A data processing module is used to pre-process the collected key operation data, including applying wavelet transform and filtering algorithm to remove noise to obtain pre-processed data, and identifying key features of power grid operation in the pre-processed data through feature extraction technology to obtain a key feature set; wherein the feature extraction technology includes but is not limited to principal component analysis (PCA) or independent component analysis (ICA);

[0159] A model prediction module, used to learn and analyze key feature sets using a pre-trained deep learning model, predict instantaneous power fluctuations in the power grid, and obtain prediction results; wherein the deep learning model includes but is not limited to convolutional neural networks (CNN) and recurrent neural networks (RNN), and model training includes but is not limited to the use of historical data sets and transfer learning technology;

[0160] The compensation strategy control module is used to input the prediction results into a preset adaptive compensation control strategy and dynamically adjust the control parameters according to the real-time status of the power grid. The adaptive compensation control strategy includes but is not limited to model predictive control (MPC) and fuzzy logic controller, and the adjustment of the control strategy is based on the real-time data of the power grid and the historical operation data.

[0161] The system realizes real-time monitoring, prediction and compensation of instantaneous power fluctuations in the power grid by integrating modules such as data acquisition, data processing, model prediction and compensation strategy control. Each module has clear functions and implementation details, ensuring the coordinated operation and efficient performance of the entire system.

[0162] Example 9

[0163] like Figure 5 As shown, an electronic device includes a processor and a memory, wherein the processor is used to execute a computer program stored in the memory to implement a power grid instantaneous power fluctuation compensation control method.

[0164] The present invention also provides an electronic device for implementing the instantaneous power fluctuation compensation control method of the above-mentioned embodiment; the electronic device includes a memory, at least one processor, a computer program stored in the memory and executable on the at least one processor, and at least one communication bus. The memory can be used to store the computer program, and the processor implements the instantaneous power fluctuation compensation control method of the embodiment by running or executing the computer program stored in the memory and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area can store data (such as audio data) created according to the use of the electronic device, etc. In addition, the memory can include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (SmartMedia Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other non-volatile solid-state storage devices. At least one processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the electronic device, and various interfaces and lines are used to connect the various parts of the entire electronic device. The memory in the electronic device stores multiple instructions to implement a control method for instantaneous power fluctuation compensation of a power grid.

[0165] Example 10

[0166] A computer-readable storage medium stores at least one instruction. When the at least one instruction is executed by a processor, a method for compensating and controlling instantaneous power fluctuations in a power grid is implemented.

[0167] If the module / unit integrated in the electronic device 100 is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the process in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor, the steps of the above-mentioned various method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device that can carry computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory and read-only memory (ROM, Read-Only Memory). It should be understood by those skilled in the art that the embodiments of the present invention can be provided as a method, system, or computer program product. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code. The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 These computer program instructions can also be stored in a computer-readable memory that can guide a computer or other programmable data processing device to work in a specific way, so that the instructions stored in the computer-readable memory produce a product including an instruction device, which implements the functions specified in the process. Figure 1 A process or multiple processes and / or boxes Figure 1 These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to produce a computer-implemented process, so that the instructions executed on the computer or other programmable device provide for implementing the process in the process. Figure 1A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0168] It is known from common technical knowledge that the present invention can be implemented by embodiments thereof without departing from its spirit or essential features. Therefore, the embodiments disclosed above are only illustrative in all respects and are not exclusive. All changes within the scope of the present invention or within the scope equivalent to the present invention are included in the present invention.

Claims

1. A method for compensating and controlling instantaneous power fluctuations in a power grid, characterized in that: The following steps are involved: S1. Use high-precision synchronous acquisition technology to collect key operation data of the power grid in real time, including voltage, current, frequency and load changes; S2. Preprocessing the collected key operation data, including applying wavelet transform and filtering algorithm to remove noise, obtaining preprocessed data, identifying key features of power grid operation in the preprocessed data through feature extraction technology, and obtaining a key feature set; S3, using a pre-trained deep learning model to learn and analyze the key feature set, predict the instantaneous power fluctuation in the power grid, and obtain the prediction result; S4. Input the prediction results into the preset adaptive compensation control strategy and dynamically adjust the control parameters according to the real-time status of the power grid.

2. A method for compensating and controlling instantaneous power fluctuations in a power grid according to claim 1, characterized in that: The S1 step includes the following steps: S1.

1. Configure a group of synchronous acquisition modules, each module includes a voltage sensor, a current sensor, a frequency sensor, and a load monitor, all of which are connected to a high-precision clock synchronization device to synchronize timestamps; S1.2, transmitting the output of the synchronous acquisition module to a central data processor via an optical fiber communication network; S1.

3. Implement data acquisition management software in the central data processor to control the startup, synchronization and real-time collection of data from the data acquisition unit.

3. A method for compensating and controlling instantaneous power fluctuations in a power grid according to claim 1, characterized in that: The pre-training steps of deep learning models include: Constructing a deep learning model, including at least one convolutional neural network sub-model and one recurrent neural network sub-model, for processing the time series and spatial characteristics of the power grid operation data; The convolutional neural network sub-model is used to extract the spatial features of the power grid operation data and obtain the spatial feature representation; Use the recurrent neural network sub-model to extract the time series features of the data and obtain the time feature representation; The obtained spatial feature representation and temporal feature representation are fused to form a comprehensive feature representation; Using the fused comprehensive feature representation, the features are integrated through the fully connected layer to obtain the integrated feature vector; The integrated feature vector is input into the output layer, and the instantaneous power fluctuation in the power grid is predicted based on the mapping relationship learned during the training process to obtain the prediction result; Based on the prediction results, the deep reinforcement learning method is used to pre-train the deep learning model, where the deep reinforcement learning algorithm optimizes the model's prediction strategy based on historical power fluctuation data; During the pre-training process, the power grid operation experience provided by the expert system is used as a reward signal to guide the training direction of the deep reinforcement learning algorithm, so that the deep learning model can make accurate power fluctuation predictions under various power grid conditions and obtain a trained deep learning model.

4. A method for compensating and controlling instantaneous power fluctuations in a power grid according to claim 1, characterized in that: Based on the model predictive control principle, an adaptive compensation control strategy is designed. The following steps are included: Determine the control objectives and constraints for grid operation, including but not limited to standards for voltage stability, power balance, and safe operation of equipment; Based on the determined control objectives and constraints, a predictive control mathematical model is constructed to describe the dynamic behavior of power grid operation; The prediction results provided by the deep learning model are used as the input of the predictive control mathematical model to perform real-time optimization calculations and predict the changing trend of the future power grid state. Based on the optimization calculation results of the predictive control mathematical model, a series of compensating control actions are formulated, which are aimed at adjusting the operating parameters of the power grid equipment to cope with the predicted instantaneous power fluctuations. The formulated compensating control actions are converted into control signals and sent to the key control nodes of the power grid to perform real-time adjustments. The post-execution power grid operation data is collected through the real-time monitoring system to evaluate the effect of the compensating control actions, and the predictive control mathematical model is adjusted online according to the difference between the actual power grid state and the predicted state. This step is repeated to continuously compensate and control the instantaneous power fluctuations of the power grid to ensure the stable operation of the power grid.

5. A method for compensating and controlling instantaneous power fluctuations in a power grid according to claim 1, characterized in that: It also includes adaptive compensation control strategy combined with fuzzy logic controller for compensation control, specifically including: Based on the historical operation data of the power grid and expert experience, the initial parameters of the fuzzy logic controller are set to define the fuzzy sets and inference rules; Real-time monitoring of the operating status of the power grid, including voltage, current, frequency and load changes, and collection of real-time data for real-time adjustment of the fuzzy logic controller; inputting the collected real-time data into the fuzzy logic controller, and converting the precise input data into fuzzy values ​​through fuzzification processing; The fuzzy inference system in the fuzzy logic controller is used to perform inference calculations according to predefined fuzzy rules and fuzzy values ​​to obtain fuzzy control outputs; the obtained fuzzy control outputs are defuzzified and converted into specific control instructions; according to the obtained control instructions, the operating parameters of the power grid equipment are dynamically adjusted and an adaptive compensation control strategy is implemented; after executing the control instructions, feedback data from the power grid is collected, the control effect is evaluated, and the parameters of the fuzzy logic controller are adaptively adjusted according to the actual power grid response; this step is repeated to form a closed-loop control system that continuously adapts to the nonlinear characteristics and uncertainties of the power grid.

6. A method for compensating and controlling instantaneous power fluctuations in a power grid according to claim 1, characterized in that: It also includes deploying edge computing devices at key nodes of the power grid to perform localized processing and real-time analysis of data, and the decision results of the edge computing devices are fed back to the deep learning model to optimize model predictions.

7. A method for controlling instantaneous power fluctuation compensation of a power grid according to claim 6, characterized in that: The following steps are also included: Collect real-time and historical data on power grid operation, as well as decision results provided by edge computing devices and prediction results of deep learning models; Use edge computing devices to perform preliminary processing on real-time data, including data cleaning, formatting, and feature extraction, to generate data sets that can be used in intelligent decision support systems; Input the processed data set into an intelligent decision support system that integrates expert systems and optimization algorithms; In the intelligent decision support system, the knowledge base and reasoning engine of the expert system are applied to analyze the data set, identify the key characteristics of power grid operation and potential power fluctuation risks, and obtain the first analysis results; Combined with the prediction results of the deep learning model, the intelligent decision support system uses optimization algorithms, including genetic algorithms or particle swarm optimization, to evaluate and select possible grid regulation solutions to obtain a second analysis result; The intelligent decision support system generates an optimal power grid regulation plan based on the first analysis result and the second analysis result, including real-time data, historical data, prediction results and expert experience; Output the optimal adjustment plan to the power grid control system to guide the real-time adjustment of power grid equipment, compensate for instantaneous power fluctuations and maintain power grid stability; Monitor the implementation effect of the optimal adjustment plan, collect feedback data, and use the feedback data for model updating and optimization of the intelligent decision support system.

8. A power grid instantaneous power fluctuation compensation control system, characterized in that: include: Data acquisition module, used to collect key operation data of the power grid in real time, including voltage, current, frequency and load changes, using high-precision synchronous acquisition technology; The data processing module is used to pre-process the collected key operation data, including applying wavelet transform and filtering algorithm to remove noise to obtain pre-processed data, and identifying the key features of power grid operation in the pre-processed data through feature extraction technology to obtain a key feature set; The model prediction module is used to learn and analyze the key feature set using a pre-trained deep learning model to predict the instantaneous power fluctuation in the power grid and obtain the prediction results; The compensation strategy control module is used to input the prediction results into the preset adaptive compensation control strategy and dynamically adjust the control parameters according to the real-time status of the power grid.

9. An electronic device, characterized in that: It comprises a processor and a memory, wherein the processor is used to execute a computer program stored in the memory to implement a power grid instantaneous power fluctuation compensation control method as claimed in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores at least one instruction, and when the at least one instruction is executed by a processor, a method for compensating and controlling instantaneous power fluctuations in a power grid as described in any one of claims 1 to 7 is implemented.

Images