Power distribution network voltage sag compensation method, system, equipment and medium

Abstract

The invention relates to the technical field of power systems, in particular to a power distribution network voltage sag compensation method, system and device and a medium. The method comprises the following steps: monitoring the voltage level of a power distribution network in real time by using a high-precision voltage sensor to obtain voltage data; identifying a voltage sag event based on the acquired voltage data; performing feature extraction on the determined voltage sag event; according to the extracted characteristic quantity, a compensation strategy is dynamically formulated, and the compensation strategy comprises the step of determining the amplitude and duration of compensation; executing voltage compensation through the grid-connected inverter according to the determined compensation strategy, wherein the output of the grid-connected inverter is adjusted according to the impedance change of the power grid and the influence of voltage disturbance; and monitoring the compensated power grid voltage, evaluating a compensation effect, and feeding back an evaluation result to optimize a subsequent compensation strategy. According to the invention, the voltage sag event can be rapidly and accurately detected, and an effective compensation strategy is formulated for compensation.

Description

Technical Field

[0001] The present invention relates to the technical field of power systems, and in particular to a method, system, equipment and medium for compensating for voltage sag in a distribution network. Background Art

[0002] In modern power systems, distribution networks, as the last link in power transmission, undertake the important task of transmitting electrical energy from substations to end users. However, distribution networks often face voltage sag problems in actual operation, especially under non-ideal grid conditions, such as grid impedance uncertainty and voltage disturbances. These problems not only affect the stability of the grid, but also cause serious damage to users' power equipment, resulting in equipment downtime, production interruption and economic losses. Voltage sag refers to the phenomenon that the grid voltage is lower than the rated value in a short period of time, usually caused by factors such as the start and stop of high-power equipment, short-circuit faults or grid load fluctuations.

[0003] Existing voltage sag compensation technologies mainly rely on equipment such as static voltage compensation devices and dynamic voltage restorers. These traditional technologies often find it difficult to provide effective compensation when faced with complex power grid environments, especially when the grid impedance changes and voltage sags occur frequently. The compensation effect is not ideal, resulting in a decrease in the power quality of the power grid. For example, patent CN109390960A describes a real-time online comprehensive compensation device for voltage sag in a distribution network and a method thereof. The method dynamically adjusts the output voltage by installing equipment on the high-voltage side or the low-voltage side of the distribution transformer to address the ubiquitous voltage sag power quality problem in the distribution network. However, this technology may have problems such as slow response speed and inflexible compensation strategy in practical applications.

[0004] In addition, the existing technology also has deficiencies in voltage sag detection. Traditional detection methods often rely on fixed thresholds, lack flexibility, and cannot adapt to dynamic changes in the power grid. Even after a voltage sag event is detected, the formulation of compensation strategies is often based on experience and lacks scientific basis, resulting in improper selection of compensation amplitude and duration, further affecting the compensation effect. Patent CN113315133A provides a control method, device, and electronic device for a voltage sag control device. This method still has limitations in terms of adaptability under non-ideal power grid conditions and dynamic adjustment of compensation strategies.

[0005] Therefore, how to quickly and accurately detect voltage sag events under non-ideal grid conditions and formulate effective compensation strategies has become a technical problem that needs to be solved urgently in the current power system. Based on this, the present application proposes a distribution network voltage sag compensation method, system, device and medium. Summary of the invention

[0006] The purpose of the present invention is to provide a distribution network voltage sag compensation method, system, device and medium to solve the problem of how to quickly and accurately detect voltage sag events under non-ideal grid conditions and formulate effective compensation strategies for compensation.

[0007] To achieve the above purpose, the following technical solution is adopted.

[0008] A method for compensating for voltage sag in a distribution network comprises the following steps:

[0009] S1. Use high-precision voltage sensors to monitor the voltage level of the distribution network in real time and obtain voltage data;

[0010] S2. Based on the acquired voltage data, a voltage sag detection model pre-developed by combining statistics and machine learning algorithms is used to quickly identify voltage sag events;

[0011] S3. Extracting features of the determined voltage sag event, wherein the extracted features include the depth, duration and affected grid area of ​​the sag;

[0012] S4. Based on the extracted feature quantity, a compensation strategy is dynamically formulated using fuzzy logic or a neural network algorithm, wherein the compensation strategy includes determining the magnitude and duration of the compensation;

[0013] S5. Performing voltage compensation according to the determined compensation strategy through the grid-connected inverter, wherein the output of the grid-connected inverter is adjusted according to the influence of the grid impedance change and the voltage disturbance;

[0014] S6. Monitor the grid voltage after compensation, evaluate the compensation effect, and feed the evaluation result back to step S4 to optimize the subsequent compensation strategy.

[0015] Optionally, also include,

[0016] In step S5, a robust control algorithm is applied to enhance the performance of the grid-connected inverter under non-ideal grid conditions.

[0017] Optionally, the robust control algorithm includes the following steps, which are applied to a closed-loop system of grid-connected inverter control:

[0018] Through a low-pass filter, high-frequency noise and disturbance in the power grid are filtered out, and stable power grid voltage and current signals are extracted to form the input signal of the closed-loop system;

[0019] The obtained stable grid voltage and current signals are compared with predetermined reference signals, and an error signal reflecting the deviation between the actual output and the expected output of the closed-loop system is calculated;

[0020] Adopting adaptive control strategy, dynamically adjusting the control parameters of the inverter according to the size and change trend of the error signal, reducing the error signal, and optimizing the stability and responsiveness of the closed-loop system;

[0021] The robustness of the closed-loop system is analyzed using Lyapunov stability theory, and the analysis results are obtained, where the closed-loop system includes the grid-connected inverter, the grid-connected equipment, and the control algorithm;

[0022] Based on the obtained analysis results, the adaptive control strategy is adjusted to optimize the control parameters of the grid-connected inverter.

[0023] Optionally, the steps of pre-developing a voltage sag detection model by combining statistics and machine learning algorithms specifically include:

[0024] The voltage data obtained by real-time monitoring of the distribution network using high-precision voltage sensors is denoised using wavelet transform, and the data is Z-score standardized to eliminate noise and outliers to obtain a preprocessed voltage data set;

[0025] Based on the preprocessed voltage data set, the mutual information between each feature and the voltage sag event is calculated using a statistical method based on information theory, and the features with mutual information values ​​higher than a set threshold are selected as the features most relevant to the voltage sag event, including but not limited to the voltage amplitude change rate, frequency change rate and voltage imbalance, to obtain a relevant feature set;

[0026] The obtained relevant feature set is used to train a machine learning model for identifying characteristic patterns of voltage sags, thereby obtaining a trained voltage sag detection model.

[0027] Optionally, based on the acquired voltage data, a voltage sag detection model pre-developed by combining statistics and machine learning algorithms is used to quickly identify specific steps of voltage sag events, including:

[0028] S21, using a high-precision voltage sensor to monitor the voltage data obtained by the distribution network in real time, and importing it as input data into a pre-trained voltage sag detection model;

[0029] S22, the voltage sag detection model analyzes the real-time voltage data and outputs a predicted probability of whether each event is a voltage sag, wherein the predicted probability of each event indicates the possibility that the event is a voltage sag;

[0030] S23, performing confidence evaluation on the output prediction probability. If the prediction probability is lower than a preset confidence threshold, it is not confirmed as a voltage sag event, and the result is fed back to step S22 for re-event identification;

[0031] S24. If the predicted probability exceeds a preset confidence threshold, the event is confirmed to be a voltage sag event;

[0032] S25, outputting the confirmed voltage sag event and its characteristics as input data for feature extraction in step S3, and for formulating subsequent compensation strategies.

[0033] Optionally, the step of dynamically formulating a compensation strategy using fuzzy logic or a neural network algorithm based on the extracted feature quantity specifically includes:

[0034] The extracted features of voltage sag events, including the depth, duration and affected grid area of ​​the sag, are normalized to eliminate the dimensional influence between different features and obtain a standardized feature vector.

[0035] The obtained standardized feature vector is input into the fuzzy logic controller and fuzzified through the preset fuzzy rule base, where each feature corresponds to multiple linguistic variables and a membership function is assigned to each linguistic variable;

[0036] Based on the fuzzification results, the fuzzy control action reflecting the response strategy to the voltage sag event is obtained through the inference engine of the fuzzy logic controller.

[0037] The obtained fuzzy control action is converted into specific compensation parameters through a defuzzification process;

[0038] The obtained standardized feature vector is input into the trained neural network model, and the model predicts the optimal compensation strategy through weighted summation and activation function processing of multiple layers of neurons;

[0039] Compare the compensation strategies output by the fuzzy logic controller and the neural network model, select the optimal strategy or perform weighted fusion of the two results to obtain the final compensation strategy.

[0040] Optionally, the steps of using Lyapunov stability theory to analyze the robustness of the closed-loop system, obtaining analysis results, and adjusting the adaptive control strategy based on the obtained analysis results to optimize the control parameters of the grid-connected inverter include:

[0041] The closed-loop system includes a grid-connected inverter, a grid connection device, and a control algorithm, wherein the grid-connected inverter is the main execution unit of the closed-loop system, the grid connection device is used to interface with the grid, and the control algorithm is used to adjust the output of the grid-connected inverter to maintain grid stability;

[0042] Construct a Lyapunov function for the closed-loop system that reflects the stability and energy changes of the system state;

[0043] Compute the time derivative of the constructed Lyapunov function;

[0044] By calculating the time derivative, the stability conditions of the system are determined, and the key parameters and factors affecting the stability of the system are found, including the output voltage, frequency and phase of the inverter;

[0045] Based on the determined stability conditions, sensitivity analysis is performed on the grid-connected inverter control parameters to identify the parameters that have the greatest impact on system stability;

[0046] According to the results of sensitivity analysis, formulate parameter optimization strategy, determine the direction and amplitude of adjusting control parameters, and improve the robustness and stability of the system;

[0047] Implement the formulated parameter optimization strategy, dynamically adjust the control parameters of the grid-connected inverter, adapt to grid impedance changes and voltage disturbances, and ensure the performance and stability of the closed-loop system.

[0048] A distribution network voltage sag compensation system, comprising:

[0049] Voltage monitoring unit, used to monitor the voltage level of the distribution network in real time and obtain voltage data;

[0050] A voltage sag detection unit for quickly identifying voltage sag events using a pre-developed voltage sag detection model that combines statistics and machine learning algorithms;

[0051] A feature extraction unit, used for extracting features of the determined voltage sag event, wherein the extracted features include the depth, duration and affected grid area of ​​the sag;

[0052] A compensation strategy generating unit, used for dynamically formulating a compensation strategy based on the extracted feature quantity by using fuzzy logic or neural network algorithm, wherein the compensation strategy includes determining the magnitude and duration of compensation;

[0053] a voltage compensation execution unit, configured to execute voltage compensation according to a determined compensation strategy, wherein the output of the grid-connected inverter is adjusted according to the influence of grid impedance changes and voltage disturbances;

[0054] Compensation effect evaluation unit: used to monitor the compensated grid voltage, evaluate the compensation effect, and feed back the evaluation result to the compensation strategy generation unit to optimize the subsequent compensation strategy;

[0055] Robust control algorithm unit: used to apply robust control algorithms to enhance the performance of grid-connected inverters under non-ideal grid conditions, including filtering out high-frequency noise and disturbances through low-pass filters, calculating error signals, dynamically adjusting grid-connected inverter control parameters, and using Lyapunov stability theory to analyze the robustness of closed-loop systems and optimize control parameters.

[0056] 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 distribution network voltage sag compensation method.

[0057] A computer-readable storage medium The computer-readable storage medium stores at least one instruction, and when the at least one instruction is executed by a processor, the method for compensating for voltage sag in a distribution network is implemented.

[0058] Compared with the prior art, the present invention has the following beneficial effects:

[0059] The method of the present application can quickly identify voltage sag events by combining statistics and machine learning algorithms. This feature effectively improves the accuracy and response speed of detection, ensuring that compensation measures can be taken in a timely manner when a voltage sag occurs, reducing the impact on user equipment. By extracting the characteristic quantity of the voltage sag event and dynamically formulating a compensation strategy using fuzzy logic or a neural network algorithm, the present invention can flexibly adjust the compensation amplitude and duration according to the real-time status and historical data of the power grid. This dynamic adjustment capability makes the compensation strategy more scientific and reasonable, and more adaptable. The present invention can quickly and accurately detect voltage sag events and formulate an effective compensation strategy for compensation.

[0060] The present invention also applies a robust control algorithm, which uses Lyapunov stability theory to analyze the robustness of the closed-loop system, ensuring that the system can maintain stable operation under conditions of grid impedance changes and voltage disturbances. This significantly improves the performance of the grid-connected inverter under non-ideal grid conditions and enhances the overall stability and reliability of the grid.

[0061] Through effective voltage sag compensation, the present invention can significantly improve the power quality of the distribution network, reduce power outages and equipment damage caused by voltage sag, thereby reducing users' economic losses and improving their power consumption experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 It is a schematic diagram of the steps of a method for compensating for voltage sag in a distribution network according to an embodiment of the present invention;

[0063] Figure 2 It is a module schematic diagram of an embodiment of a distribution network voltage sag compensation system according to the present invention;

[0064] Figure 3 is a structural block diagram of an electronic device according to an embodiment of the present invention; DETAILED DESCRIPTION

[0065] 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.

[0066] 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.

[0067] like Figure 1 As shown, a method for compensating for voltage sag in a distribution network is provided. The method involves multiple steps including real-time monitoring, event detection, feature extraction, strategy formulation, compensation execution, and effect evaluation, as follows:

[0068] First, a high-precision voltage sensor is used to monitor the voltage level of the distribution network in real time to obtain voltage data. The sensor can accurately capture small changes in voltage and provide an accurate data basis for subsequent voltage sag event detection.

[0069] Next, based on the acquired voltage data, a pre-developed voltage sag detection model that combines statistics and machine learning algorithms is used to quickly identify voltage sag events. The model is able to analyze abnormal patterns in the voltage data and trigger an alarm when a voltage sag is detected.

[0070] Then, feature extraction is performed on the identified voltage sag events, including the depth, duration, and affected grid area of ​​the sag. These features are crucial for the subsequent formulation of compensation strategies because they directly affect the magnitude and duration of compensation.

[0071] Next, based on the extracted features, compensation strategies are dynamically formulated using fuzzy logic or neural network algorithms. The fuzzy logic controller converts the input features into compensation actions through a preset fuzzy rule base, while the neural network model predicts the optimal compensation strategy by learning historical data. These strategies include determining the magnitude and duration of compensation to ensure that the grid voltage can quickly return to normal levels.

[0072] Subsequently, voltage compensation is performed by the grid-connected inverter according to the determined compensation strategy. The output of the grid-connected inverter is adjusted according to the influence of grid impedance changes and voltage disturbances to achieve accurate voltage support.

[0073] Finally, the compensated grid voltage is monitored, the compensation effect is evaluated, and the evaluation results are fed back to the compensation strategy formulation step to optimize the subsequent compensation strategy. This closed-loop control ensures that the compensation strategy can be dynamically adjusted according to the real-time status of the grid, improving the efficiency and accuracy of voltage sag compensation.

[0074] In specific implementation, the following technical means can be used:

[0075] Use a high-precision voltage sensor, such as a Hall effect sensor, to achieve accurate monitoring of the voltage.

[0076] Wavelet transform is used to denoise voltage data and improve the accuracy of voltage sag detection.

[0077] Support vector machine (SVM) or deep learning algorithms, such as convolutional neural network (CNN), are used to extract features and recognize patterns of voltage sag events.

[0078] Using fuzzy logic controller or neural network model, the compensation strategy is dynamically adjusted according to the real-time status of the power grid.

[0079] The grid-connected inverter adopts advanced control strategies, such as virtual synchronous machine (VSG) control, to achieve fast response and accurate compensation to the grid voltage.

[0080] The compensation effect is evaluated through real-time data acquisition system and data analysis software, and the compensation strategy is adjusted according to the evaluation results.

[0081] The present invention can achieve rapid response and effective compensation for temporary voltage sag in the distribution network, thereby improving the stability and reliability of the power grid.

[0082] As a specific example, the steps of pre-developing a voltage sag detection model by combining statistics and machine learning algorithms specifically include:

[0083] The voltage data obtained by real-time monitoring of the distribution network using high-precision voltage sensors is first denoised using wavelet transform to eliminate the influence of noise and outliers. Wavelet transform can effectively separate the high-frequency noise part of the signal and retain the main feature information. Then, the data is normalized by Z-score, which can eliminate the dimensional influence between different features, making the data have zero mean and unit variance, providing a unified data standard for subsequent feature extraction and model training.

[0084] Based on the preprocessed voltage data set, the mutual information between each feature and the voltage sag event is calculated using a statistical method based on information theory. Mutual information can quantify the mutual dependence between two random variables, thereby evaluating the amount of relevant information of the feature for the voltage sag event. Features with mutual information values ​​higher than the set threshold are selected as the features most relevant to the voltage sag event. These features include but are not limited to the voltage amplitude change rate, frequency change rate and voltage imbalance, thereby obtaining a relevant feature set.

[0085] The obtained relevant feature set is used to train a machine learning model that can identify the characteristic patterns of voltage sags. The model can be a support vector machine (SVM), a random forest (RF), a neural network, etc. By learning the characteristic patterns in historical data, the model can identify voltage sag events. During the training process, methods such as cross-validation and grid search are used to optimize the model's hyperparameters to ensure that the model has good generalization ability and prediction accuracy.

[0086] Through the above steps, the present invention can pre-develop a voltage sag detection model, which can accurately identify and predict voltage sag events, and provide scientific basis and technical support for voltage sag compensation in distribution networks. This method combining statistics and machine learning improves the accuracy and efficiency of voltage sag detection, and provides a strong guarantee for the stable operation of the power grid.

[0087] As a specific example, based on the acquired voltage data, using a pre-developed voltage sag detection model that combines statistics and machine learning algorithms, the specific steps for quickly identifying voltage sag events include:

[0088] First, the distribution network is monitored in real time using high-precision voltage sensors to obtain voltage data. This data is used as input to a pre-trained voltage sag detection model. The model is developed using statistical and machine learning algorithms and is able to identify voltage sag events.

[0089] The voltage sag detection model receives real-time voltage data and analyzes it. The model evaluates each monitored event through its built-in algorithm and outputs a predicted probability, which represents the likelihood that the event is a voltage sag.

[0090] The predicted probability output by the model is then subjected to a confidence assessment. This step involves comparing the predicted probability with a preset confidence threshold. If the predicted probability is lower than this threshold, the event is not confirmed as a voltage sag event. In this case, the system will feed the results back to the event analysis step to re-identify the event to improve the accuracy of the detection.

[0091] On the contrary, if the predicted probability exceeds the preset confidence threshold, the event is confirmed as a voltage sag event. This confirmation provides a quantitative basis for judgment based on the predicted probability of the model.

[0092] Once an event is confirmed as a voltage sag event, the system will output these events and their related features. These data will be used as input for the feature extraction step and used to formulate the subsequent compensation strategy. This step ensures that the compensation strategy can be formulated based on accurate voltage sag event data, improving the effectiveness and pertinence of the compensation strategy.

[0093] Through the above steps, the present invention provides a voltage sag detection model based on statistics and machine learning algorithms, which can quickly and accurately identify voltage sag events and provide scientific and accurate data support for distribution network voltage sag compensation. This method not only improves the efficiency of voltage sag detection, but also enhances the adaptability and effectiveness of the compensation strategy, thereby improving the overall stability and reliability of the distribution network.

[0094] As a specific example, the steps of dynamically formulating a compensation strategy using fuzzy logic or a neural network algorithm according to the extracted feature quantity specifically include:

[0095] The extracted features of voltage sag events, including the depth, duration and affected grid area of ​​the sag, are normalized to eliminate the dimensional effects between different features and obtain standardized feature vectors. This step ensures that different features are compared at the same scale, which facilitates subsequent fuzzy logic or neural network algorithm processing.

[0096] The obtained standardized feature vector is input into the fuzzy logic controller and fuzzified through the preset fuzzy rule base. Each feature corresponds to multiple linguistic variables, and a membership function is assigned to each linguistic variable. The fuzzy rule base is built based on expert experience and historical data to convert the precise feature value into the membership value in the fuzzy set.

[0097] Based on the fuzzification results, the inference engine of the fuzzy logic controller is used to infer the fuzzy control action that reflects the response strategy to the voltage sag event. This step maps the input feature vector to the specific control action through the "if-then" rule of fuzzy logic.

[0098] The obtained fuzzy control action is converted into a specific compensation parameter through the defuzzification process. The defuzzification process usually uses the centroid method or other methods to convert the fuzzy set into a specific output value, that is, the parameter of the compensation strategy.

[0099] The obtained standardized feature vector is input into the trained neural network model, which predicts the optimal compensation strategy through weighted summation and activation function processing of multiple layers of neurons. The neural network model can automatically learn the relationship between features and compensation strategies without explicit programming.

[0100] Compare the compensation strategies output by the fuzzy logic controller and the neural network model, select the optimal strategy or perform weighted fusion of the two results to obtain the final compensation strategy. This step ensures the diversity and robustness of the compensation strategy and improves the accuracy and effectiveness of the compensation.

[0101] The present invention can dynamically formulate the most appropriate voltage sag compensation strategy according to the real-time status and historical data of the power grid, thereby improving the response speed and compensation effect of the distribution network when facing voltage sag, thereby improving the stability and reliability of the power grid.

[0102] Example 2

[0103] A distribution network voltage sag compensation method applies a robust control algorithm in the voltage compensation step to enhance the performance of a grid-connected inverter under non-ideal grid conditions.

[0104] The robust control algorithm is as follows:

[0105] At the input of the closed-loop system, a low-pass filter is set to filter out high-frequency noise and transient disturbances in the power grid. The filter is designed to pass low-frequency signals and block high-frequency signals, thereby extracting stable power grid voltage and current signals, which are then used as input signals for the closed-loop system.

[0106] The filtered grid voltage and current signals are compared with the predetermined reference signals, and an error signal is calculated from the comparison results. The error signal quantitatively reflects the deviation between the actual output and the expected output of the closed-loop system, providing a basis for the control algorithm to adjust the inverter output.

[0107] According to the size and change trend of the error signal, the control parameters of the inverter are dynamically adjusted. This step involves real-time monitoring of the error signal and dynamically adjusting the control parameters of the inverter according to its changes to reduce the error signal and optimize the stability and responsiveness of the closed-loop system. The control parameters can be adjusted using a proportional-integral-derivative (PID) controller or other advanced control algorithms.

[0108] The robustness of the closed-loop system is analyzed using Lyapunov stability theory. A Lyapunov function is constructed that reflects the stability and energy change of the system state. The stability of the system in the face of disturbances is proved by calculating the time derivative of the Lyapunov function and ensuring that it is negative definite or negative semidefinite.

[0109] Based on the analysis results of Lyapunov stability theory, the control parameters in the adaptive control strategy are adjusted. This step involves performing sensitivity analysis on the control parameters of the inverter, identifying the parameters that have the greatest impact on system stability, and optimizing these parameters based on the analysis results to improve the robustness and stability of the system.

[0110] During the process of voltage compensation performed by the grid-connected inverter, the real-time performance of the grid is continuously monitored. Based on the monitoring results, the control parameters are adjusted in real time to ensure the effectiveness of the compensation strategy and the optimal operation of the inverter. This step ensures that the system can adapt to the dynamic changes in grid conditions and maintain the accuracy and timeliness of voltage compensation.

[0111] The present invention can improve the performance of grid-connected inverters under non-ideal grid conditions, such as grid impedance uncertainty and voltage disturbance, and ensure the effectiveness of voltage sag compensation and grid stability. The application of this robust control algorithm makes the voltage compensation method more adaptable to complex grid environments and improves the power quality and power supply reliability of the distribution network.

[0112] As a specific example of the above example,

[0113] The closed-loop system involved in this method includes a grid-connected inverter, a grid connection device, and a control algorithm. The grid-connected inverter, as the main execution unit of the closed-loop system, is responsible for adjusting its output according to the control signal to maintain the stability of the grid voltage; the grid connection device is used to interface with the grid to ensure the effective transmission of electric energy; and the control algorithm is used to adjust the output of the grid-connected inverter to maintain the stability of the grid.

[0114] Construct a Lyapunov function for the closed-loop system, which can reflect the stability and energy change of the system state. Usually a positive definite function is selected as the Lyapunov candidate function, such as V(x) = xTPx, where x represents the system state vector and P is a positive definite matrix.

[0115] Compute the time derivative of the constructed Lyapunov function This derivative should be negative definite or negative semidefinite to prove the stability of the system state, that is, to ensure that for all nonzero x,

[0116] By calculating the time derivative, the stability conditions of the system are determined, and the key parameters and factors affecting the stability of the system are found, including the output voltage, frequency and phase of the inverter.

[0117] Based on the determined stability conditions, sensitivity analysis of the grid-connected inverter control parameters is performed to identify the parameters that have the greatest impact on system stability.

[0118] According to the results of sensitivity analysis, a parameter optimization strategy is formulated to determine the direction and amplitude of adjusting the control parameters to improve the robustness and stability of the system.

[0119] Implement the formulated parameter optimization strategy, dynamically adjust the control parameters of the grid-connected inverter, adapt to grid impedance changes and voltage disturbances, and ensure the performance and stability of the closed-loop system.

[0120] Through the above steps, the present invention can improve the performance of the grid-connected inverter under non-ideal grid conditions, such as grid impedance uncertainty and voltage disturbance, and ensure the effectiveness of voltage sag compensation and the stability of the grid. The application of this robust control algorithm makes the voltage compensation method more adaptable to complex grid environments and improves the power quality and power supply reliability of the distribution network.

[0121] Example 3

[0122] like Figure 2 As shown, a distribution network voltage sag compensation system is specifically as follows:

[0123] The voltage monitoring unit is equipped with a high-precision voltage sensor to monitor the voltage level of the distribution network in real time and obtain voltage data. The sensor can capture voltage signals at a high sampling rate to ensure the real-time and accuracy of the data.

[0124] The voltage sag detection unit contains a pre-developed voltage sag detection model that combines statistical and machine learning algorithms to quickly identify voltage sag events. The model is trained to detect characteristic patterns of voltage sags from monitoring data.

[0125] The feature extraction unit, once a voltage sag event is detected, extracts features of the event, including the depth, duration, and affected grid area of ​​the sag. These features provide key information for formulating compensation strategies.

[0126] The compensation strategy generation unit dynamically formulates the compensation strategy using fuzzy logic or neural network algorithm according to the feature quantity provided by the feature extraction unit. The fuzzy logic controller converts the features into compensation actions through the fuzzy rule base, while the neural network model predicts the optimal compensation strategy.

[0127] The voltage compensation execution unit includes a grid-connected inverter, which executes voltage compensation according to the strategy determined by the compensation strategy generation unit. The output of the grid-connected inverter is adjusted in real time according to the influence of grid impedance changes and voltage disturbances to ensure the accuracy of compensation.

[0128] The compensation effect evaluation unit is responsible for monitoring the grid voltage after compensation and evaluating the compensation effect. The evaluation results will be fed back to the compensation strategy generation unit to optimize the subsequent compensation strategy and form a closed-loop control system.

[0129] The robust control algorithm unit applies a robust control algorithm to enhance the performance of the grid-connected inverter under non-ideal grid conditions. The algorithm includes filtering out high-frequency noise and disturbances through a low-pass filter, calculating the error signal, dynamically adjusting the grid-connected inverter control parameters, and using the Lyapunov stability theory to analyze the robustness of the closed-loop system and optimize the control parameters.

[0130] Example 4

[0131] like Figure 3 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 method for compensating for voltage sag in a distribution network.

[0132] The present invention also provides an electronic device 100 for implementing a method for compensating for a voltage sag in a distribution network according to the above embodiment; the electronic device 100 includes a memory 101, at least one processor 102, a computer program 103 stored in the memory 101 and executable on the at least one processor 102, and at least one communication bus 104. The memory 101 can be used to store the computer program 103, and the processor 102 implements the steps of a method for compensating for a voltage sag in a distribution network according to embodiment 1 by running or executing the computer program stored in the memory 101 and calling the data stored in the memory 101. The memory 101 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 by 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 100, etc. In addition, the memory 101 may include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart media card (SMC), a 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 102 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 gates or transistor logic devices, discrete hardware components, etc. The processor 102 may be a microprocessor or the processor 102 may also be any conventional processor, etc. The processor 102 is the control center of the electronic device 100, and various interfaces and lines are used to connect various parts of the entire electronic device 100. The memory 101 in the electronic device 100 stores multiple instructions to implement a distribution network voltage sag compensation method.

[0133] Example 5

[0134] A 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 for voltage sag in a distribution network is implemented.

[0135] 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 1These 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 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0136] Through the coordinated work of these units, the distribution network voltage sag compensation system can monitor in real time, quickly respond to voltage sag events, and dynamically adjust the compensation strategy to adapt to changes in grid conditions, ensuring grid voltage stability and power quality. The implementation of this system provides an efficient and reliable voltage sag compensation solution that is suitable for various grid environments, especially under non-ideal grid conditions.

[0137] It is known from common technical knowledge that the present invention can be implemented by other embodiments that do not deviate from its spirit or essential features. Therefore, the above disclosed embodiments 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 for voltage sag in a distribution network, characterized in that: The following steps are included: S1. Use high-precision voltage sensors to monitor the voltage level of the distribution network in real time and obtain voltage data; S2. Based on the acquired voltage data, a voltage sag detection model pre-developed by combining statistics and machine learning algorithms is used to quickly identify voltage sag events; S3. Extracting features of the determined voltage sag event, wherein the extracted features include the depth, duration and affected grid area of ​​the sag; S4. Based on the extracted feature quantity, a compensation strategy is dynamically formulated using fuzzy logic or a neural network algorithm, wherein the compensation strategy includes determining the magnitude and duration of the compensation; S5. Performing voltage compensation according to the determined compensation strategy through the grid-connected inverter, wherein the output of the grid-connected inverter is adjusted according to the influence of the grid impedance change and the voltage disturbance; S6. Monitor the grid voltage after compensation, evaluate the compensation effect, and feed the evaluation result back to step S4 to optimize the subsequent compensation strategy.

2. A distribution network voltage sag compensation method according to claim 1, characterized in that: Also includes, In step S5, a robust control algorithm is applied to enhance the performance of the grid-connected inverter under non-ideal grid conditions.

3. A distribution network voltage sag compensation method according to claim 2, characterized in that: The robust control algorithm includes the following steps, which are applied to a closed-loop system of grid-connected inverter control: Through a low-pass filter, high-frequency noise and disturbance in the power grid are filtered out, and stable power grid voltage and current signals are extracted to form the input signal of the closed-loop system; The obtained stable grid voltage and current signals are compared with predetermined reference signals, and an error signal reflecting the deviation between the actual output and the expected output of the closed-loop system is calculated; Adopting adaptive control strategy, dynamically adjusting the control parameters of the inverter according to the size and change trend of the error signal, reducing the error signal, and optimizing the stability and responsiveness of the closed-loop system; The robustness of the closed-loop system is analyzed using Lyapunov stability theory, and the analysis results are obtained, where the closed-loop system includes the grid-connected inverter, the grid-connected equipment, and the control algorithm; Based on the obtained analysis results, the adaptive control strategy is adjusted to optimize the control parameters of the grid-connected inverter.

4. A distribution network voltage sag compensation method according to claim 1, characterized in that: The steps to pre-develop a voltage sag detection model by combining statistics and machine learning algorithms include: The voltage data obtained by real-time monitoring of the distribution network using high-precision voltage sensors is denoised using wavelet transform, and the data is Z-score standardized to eliminate noise and outliers to obtain a preprocessed voltage data set; Based on the preprocessed voltage data set, the mutual information between each feature and the voltage sag event is calculated using a statistical method based on information theory, and the features with mutual information values ​​higher than a set threshold are selected as the features most relevant to the voltage sag event, including but not limited to the voltage amplitude change rate, frequency change rate and voltage imbalance, to obtain a relevant feature set; The obtained relevant feature set is used to train a machine learning model for identifying characteristic patterns of voltage sags, thereby obtaining a trained voltage sag detection model.

5. A distribution network voltage sag compensation method according to claim 1, characterized in that: Based on the acquired voltage data, a voltage sag detection model pre-developed by combining statistics and machine learning algorithms is used to quickly identify the specific steps of voltage sag events. include, S21, using a high-precision voltage sensor to monitor the voltage data obtained by the distribution network in real time, and importing it as input data into a pre-trained voltage sag detection model; S22, the voltage sag detection model analyzes the real-time voltage data and outputs a predicted probability of whether each event is a voltage sag, wherein the predicted probability of each event indicates the possibility that the event is a voltage sag; S23, performing confidence evaluation on the output prediction probability. If the prediction probability is lower than a preset confidence threshold, it is not confirmed as a voltage sag event, and the result is fed back to step S22 for re-event identification; S24. If the predicted probability exceeds a preset confidence threshold, the event is confirmed to be a voltage sag event; S25, outputting the confirmed voltage sag event and its characteristics as input data for feature extraction in step S3, and for formulating subsequent compensation strategies.

6. A distribution network voltage sag compensation method according to claim 1, characterized in that: Based on the extracted features, the steps of dynamically formulating compensation strategies using fuzzy logic or neural network algorithms include: The extracted features of voltage sag events, including the depth, duration and affected grid area of ​​the sag, are normalized to eliminate the dimensional influence between different features and obtain a standardized feature vector. The obtained standardized feature vector is input into the fuzzy logic controller and fuzzified through the preset fuzzy rule base, where each feature corresponds to multiple linguistic variables and a membership function is assigned to each linguistic variable; Based on the fuzzification results, the fuzzy control action reflecting the response strategy to the voltage sag event is obtained through the inference engine of the fuzzy logic controller. The obtained fuzzy control action is converted into specific compensation parameters through a defuzzification process; The obtained standardized feature vector is input into the trained neural network model, and the model predicts the optimal compensation strategy through weighted summation and activation function processing of multiple layers of neurons; Compare the compensation strategies output by the fuzzy logic controller and the neural network model, select the optimal strategy or perform weighted fusion of the results of the two to obtain the final compensation strategy.

7. A distribution network voltage sag compensation method according to claim 3, characterized in that: The robustness of the closed-loop system is analyzed using Lyapunov stability theory to obtain analysis results. Based on the obtained analysis results, the adaptive control strategy is adjusted to optimize the control parameters of the grid-connected inverter. Specifically, The closed-loop system includes a grid-connected inverter, a grid connection device, and a control algorithm, wherein the grid-connected inverter is the main execution unit of the closed-loop system, the grid connection device is used to interface with the grid, and the control algorithm is used to adjust the output of the grid-connected inverter to maintain grid stability; Construct a Lyapunov function for the closed-loop system that reflects the stability and energy changes of the system state; Compute the time derivative of the constructed Lyapunov function; By calculating the time derivative, the stability conditions of the system are determined, and the key parameters and factors affecting the stability of the system are found, including the output voltage, frequency and phase of the inverter; Based on the determined stability conditions, sensitivity analysis is performed on the grid-connected inverter control parameters to identify the parameters that have the greatest impact on system stability; According to the results of sensitivity analysis, formulate parameter optimization strategy, determine the direction and amplitude of adjusting control parameters, and improve the robustness and stability of the system; Implement the formulated parameter optimization strategy, dynamically adjust the control parameters of the grid-connected inverter, adapt to grid impedance changes and voltage disturbances, and ensure the performance and stability of the closed-loop system.

8. A distribution network voltage sag compensation system, based on a distribution network voltage sag compensation method according to any one of claims 1 to 7, characterized in that: include, Voltage monitoring unit, used to monitor the voltage level of the distribution network in real time and obtain voltage data; A voltage sag detection unit for quickly identifying voltage sag events using a pre-developed voltage sag detection model that combines statistics and machine learning algorithms; A feature extraction unit, used for extracting features of the determined voltage sag event, wherein the extracted features include the depth, duration and affected grid area of ​​the sag; A compensation strategy generating unit, used for dynamically formulating a compensation strategy based on the extracted feature quantity by using fuzzy logic or neural network algorithm, wherein the compensation strategy includes determining the magnitude and duration of compensation; a voltage compensation execution unit, configured to execute voltage compensation according to a determined compensation strategy, wherein the output of the grid-connected inverter is adjusted according to the influence of grid impedance changes and voltage disturbances; Compensation effect evaluation unit: used to monitor the grid voltage after compensation, evaluate the compensation effect, and feed back the evaluation result to the compensation strategy generation unit to optimize the subsequent compensation strategy; Robust control algorithm unit: used to apply robust control algorithms to enhance the performance of grid-connected inverters under non-ideal grid conditions, including filtering out high-frequency noise and disturbances through low-pass filters, calculating error signals, dynamically adjusting grid-connected inverter control parameters, and using Lyapunov stability theory to analyze the robustness of closed-loop systems and optimize control parameters.

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 distribution network voltage sag compensation 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 the processor, a distribution network voltage sag compensation method as described in any one of claims 1 to 7 is implemented.

Images