Electric energy quality management method, system, equipment, medium and product

Abstract

The invention discloses an electric energy quality management method, system and device, a medium and a product, and relates to the field of electric energy quality management.The method comprises the steps that power grid state parameters of photovoltaic and wind power access points are obtained; determining a power grid state parameter data set according to the power grid state parameter and the historical power grid state parameter; according to the power grid state parameter data set, using an improved fast Fourier transform algorithm and wavelet transform to determine an abnormal feature vector; determining a fault source type and a fault source corresponding to traceability based on a probability reasoning method of a Bayesian network; according to the traceable fault source, generating a corresponding optimization scheduling parameter; adjusting a photovoltaic and wind power dispatching strategy and a reactive power compensation strategy by utilizing the optimized dispatching parameters; according to the invention, electric energy quality optimization after the new energy is accessed to the power grid can be realized, and the safety, stability and economy of the power grid are improved.

Description

Technical Field

[0001] The present application relates to the field of power quality management, and in particular to a power quality management method, system, equipment, medium and product. Background Art

[0002] To address the power quality issues brought about by renewable energy generation, technicians have proposed a variety of solutions, such as synchronized phasor measurement technology (PMU), power quality monitoring and evaluation technology, intelligent reactive compensation devices (static var generators (SVG) and devices used to dynamically adjust reactive power in power systems (static var compensators, SVC)), distributed energy storage systems, and grid optimization and scheduling methods based on big data and artificial intelligence. Although these technical means have improved the stability of renewable energy power grids to a certain extent, there are still problems such as monitoring lag, insufficient anomaly detection accuracy, difficulty in fault tracing, and slow dispatch optimization response. Therefore, it is difficult to adapt to the higher requirements for power quality management in power grids with a high proportion of renewable energy.

[0003] Existing power quality management technologies still have many shortcomings in scenarios where a high proportion of renewable energy is connected. First, in terms of power quality monitoring, traditional monitoring methods rely on data acquisition and supervisory control systems (SCADA) or sampling equipment at fixed measurement points. The data acquisition cycle is long, making it difficult to achieve millisecond-level real-time monitoring, resulting in a lag in the detection of abnormal power quality events. Second, in terms of voltage anomaly identification, the conventional Fast Fourier Transform (FFT) method is limited by frequency domain resolution and has difficulty accurately capturing short-term voltage fluctuation signals. Time domain analysis methods have limited accuracy when dealing with complex disturbances, resulting in insufficient ability to identify voltage fluctuations caused by renewable energy generation.

[0004] Furthermore, traditional rule-based approaches to fault source tracing and dispatch optimization struggle to adapt to the nonlinear characteristics and dynamic changes of renewable energy grids, resulting in low fault tracing accuracy and, in turn, difficulty adjusting dispatch strategies quickly. Finally, in terms of reactive power compensation control, existing strategies often rely on preset parameter thresholds, resulting in slow response times and difficulty meeting the demand for rapid compensation in response to high-frequency fluctuations in renewable energy, thus impacting grid stability.

[0005] Therefore, based on the above problems, there is an urgent need to provide a new power quality management method to optimize the power quality after new energy is connected to the power grid and improve the safety, stability and economy of the power grid. Summary of the Invention

[0006] The purpose of this application is to provide a power quality management method, system, equipment, medium and product that can optimize the power quality after new energy is connected to the power grid, and improve the safety, stability and economy of the power grid.

[0007] To achieve the above objectives, this application provides the following solutions:

[0008] In a first aspect, the present application provides a power quality management method, the power quality management method comprising:

[0009] Obtaining grid status parameters of photovoltaic and wind power access points; the grid status parameters include: voltage, current, frequency and phase angle;

[0010] Determining a grid state parameter data set based on the grid state parameters and historical grid state parameters; the historical grid state parameters include: historical voltage, historical current, historical frequency and historical phase angle;

[0011] Determining an abnormal feature vector using an improved fast Fourier transform algorithm and wavelet transform according to the power grid state parameter data set;

[0012] According to the abnormal feature vector, the fault source type and the corresponding fault source are determined based on the probabilistic reasoning method of the Bayesian network;

[0013] Generate corresponding optimized scheduling parameters based on the traced fault source;

[0014] The photovoltaic and wind power power dispatch strategies and reactive power compensation strategies are adjusted using optimized dispatch parameters.

[0015] Optionally, determining the abnormal feature vector based on the grid state parameter data set by using an improved fast Fourier transform algorithm and wavelet transform specifically includes:

[0016] determining the main frequency components of the voltage fluctuations using an improved fast Fourier transform algorithm based on the grid state parameter data set;

[0017] Performing time-frequency analysis using wavelet transform according to the power grid state parameter data set;

[0018] An abnormal feature vector is determined according to the main frequency components and the time-frequency analysis results.

[0019] Optionally, determining the abnormal feature vector according to the main frequency component and the time-frequency analysis result specifically includes:

[0020] Using the formula F(Q)={f dom , E w ,σ(U),Δ2 U, PF} determine the abnormal feature vector F(Q); where f dom is the main frequency component of voltage fluctuation, E w is the local wavelet energy, σ(U) is the voltage standard deviation, Δ 2 U is the second-order voltage change rate, and PF is the power factor.

[0021] Optionally, the adjusting of photovoltaic and wind power dispatch strategies and reactive power compensation strategies by optimizing dispatch parameters specifically includes:

[0022] Based on the optimized dispatch parameters, a short-term load forecasting model is used to adjust the photovoltaic and wind power dispatch strategies based on historical load data, real-time load data, and weather data, including sunlight intensity, sunshine duration, wind direction, and wind speed.

[0023] According to the optimized dispatching parameters, the dynamic fuzzy decision-making method is used to adjust the reactive power compensation strategy.

[0024] Optionally, the optimizing scheduling parameters, based on historical load data, real-time load data, and weather data, uses a short-term load forecasting model to adjust the photovoltaic and wind power scheduling strategies, specifically including:

[0025] According to the optimized dispatch parameters, based on historical load data and real-time load data, a short-term load forecasting model is used to predict future load changes;

[0026] Predict future changes in photovoltaic and wind power output based on weather data;

[0027] The photovoltaic and wind power power dispatch strategies are adjusted based on the forecast results of future load changes and the forecast results of future photovoltaic and wind power output changes.

[0028] Optionally, the method of adjusting the reactive power compensation strategy by using a dynamic fuzzy decision-making method according to the optimized scheduling parameters specifically includes:

[0029] Using the formula ΔQ=k1·f dom +k2·E w +k3·σ(U)+k4·Δ 2 U+k5·PF determines the reactive power adjustment ΔQ;

[0030] Using the formula Determine the compensation current I comp ;

[0031] Among them, k1 is the first weight, k2 is the second weight, k3 is the third weight, k4 is the fourth weight, k5 is the fifth weight, f dom is the main frequency component of voltage fluctuation, Ew is the local wavelet energy, σ(U) is the voltage standard deviation, Δ 2 U is the second-order voltage change rate, PF is the power factor, and U is the voltage at the compensation device access point.

[0032] In a second aspect, the present application provides a power quality management system, the power quality management system comprising:

[0033] A grid state parameter acquisition module is used to obtain grid state parameters of photovoltaic and wind power access points; the grid state parameters include: voltage, current, frequency and phase angle;

[0034] A power grid state parameter data set determination module is used to determine a power grid state parameter data set based on the power grid state parameter and historical power grid state parameters; the historical power grid state parameters include: historical voltage, historical current, historical frequency and historical phase angle;

[0035] an abnormal feature vector determination module, configured to determine an abnormal feature vector based on the power grid state parameter data set by utilizing an improved fast Fourier transform algorithm and a wavelet transform;

[0036] A fault source tracing module is used to determine the fault source type and the corresponding fault source based on the abnormal feature vector and the probabilistic reasoning method of the Bayesian network;

[0037] The optimization scheduling parameter determination module is used to generate the corresponding optimization scheduling parameters according to the traced fault source;

[0038] The strategy adjustment module is used to adjust the photovoltaic and wind power scheduling strategies and reactive power compensation strategies by optimizing scheduling parameters.

[0039] In a third aspect, the present application provides a computer device comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of any one of the above-described power quality management methods.

[0040] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any one of the above-mentioned power quality management methods.

[0041] In a fifth aspect, the present application provides a computer program product, comprising a computer program, which, when executed by a processor, implements the steps of any one of the above-mentioned power quality management methods.

[0042] According to the specific embodiments provided in this application, this application has the following technical effects:

[0043] The present application provides a power quality management method, system, device, medium and product. Through the grid state parameter data set, an improved fast Fourier transform algorithm and wavelet transform are used to determine the abnormal feature vector. That is, the present application uses an improved fast Fourier transform (FFT) algorithm to extract voltage fluctuation characteristics, and combines it with wavelet transform (WT) to analyze short-term voltage change trends to generate abnormal feature vectors, thereby improving the accuracy of voltage anomaly identification and the safety of grid operation; based on the abnormal feature vector, the fault source type and the corresponding fault source are backtracked based on the Bayesian network to improve the intelligence level of grid anomaly analysis; and the photovoltaic and wind power power scheduling strategies and reactive compensation strategies are adjusted by optimizing scheduling parameters. Both strategies are adjusted at the same time to form a complementary relationship, thereby further improving the safety of grid operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0045] Figure 1 This is a flow chart of a power quality management method in one embodiment of the present application;

[0046] Figure 2 This is a structural diagram of a computer device in one embodiment of the present application. DETAILED DESCRIPTION

[0047] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0048] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0049] In an exemplary embodiment, Figure 1 As shown, a power quality management method is provided, including the following S1-S6. The power quality management method includes:

[0050] S1: Obtain grid status parameters of photovoltaic and wind power access points.

[0051] Real-time acquisition of grid status parameters at photovoltaic and wind power access points can ensure the coordinated regulation of photovoltaic and wind power, and can respond accurately and quickly to changes in grid operation.

[0052] In an exemplary embodiment, synchronized phasor measurement units (PMUs) and edge computing terminals are used for data collection and preliminary processing to ensure the real-time and accuracy of the data.

[0053] Among them, the role of the PMU device is to obtain the grid state parameters of the photovoltaic and wind power access points with a high-precision synchronous time reference (such as GPS signal). The grid state parameters include voltage U(t), current I(t), frequency f(t), phase angle The PMU device continuously samples grid state parameters at a high sampling rate (e.g., 30-60Hz). It provides microsecond-level time synchronization, ensuring precise alignment of data from different measurement points, laying the foundation for subsequent data analysis, fusion, and strategy optimization.

[0054] Edge computing terminals are primarily deployed at grid connection points at photovoltaic power plants, wind farms, and key substations. They receive data uploaded by PMUs and perform local preprocessing. Specifically, they perform signal noise reduction (such as noise suppression based on Kalman filtering or wavelet transforms) and analyze short-term trends in grid state parameters using the Short-Time Fourier Transform (STFT) method to identify abnormal fluctuations.

[0055] S2: Determine a grid state parameter data set Q(t) based on the grid state parameters and historical grid state parameters.

[0056] When generating a grid state parameter dataset, the collected grid state parameters may contain noise, as photovoltaic and wind farms are significantly affected by external environmental factors (such as weather changes and wind speed fluctuations). Therefore, outliers are first removed and data smoothing is performed on the grid state parameters. For example, Kalman filtering can be used to remove short-term, high-frequency noise, or wavelet threshold denoising can be used to eliminate measurement errors.

[0057] In an exemplary embodiment, the grid state parameters collected by the PMU device and processed by the edge computing terminal are further parsed and fused to form a grid state parameter data set that can be used for subsequent analysis. The grid state parameter data set includes grid state parameters and historical grid state parameters. The grid state parameters include: voltage, current, frequency, and phase angle. The historical grid state parameters include: historical voltage, historical current, historical frequency, and historical phase angle. The grid state parameter data set is used for subsequent fault source tracing and strategy optimization and adjustment.

[0058] This application utilizes PMU devices and edge computing terminals to achieve high-precision data collection and real-time monitoring of photovoltaic and wind power access points. Specifically, this application can acquire voltage, current, frequency, and phase angle within milliseconds, and construct a grid state parameter dataset based on historical grid state parameters, further enabling efficient perception of grid operating status.

[0059] S3: Based on the grid state parameter data set, the abnormal feature vector is determined using the improved fast Fourier transform algorithm and wavelet transform.

[0060] In one exemplary embodiment, a grid state parameter dataset is processed to extract voltage fluctuation characteristics to accurately identify voltage anomalies caused by photovoltaic and PV fluctuations and construct an anomaly feature vector. To do this, voltage parameters are first extracted from the grid state parameter dataset and a feature matrix is ​​constructed for signal analysis.

[0061] Specifically, the voltage U(t) at the photovoltaic and wind power access points is calculated as follows:

[0062] U(t)={U1,U2,...,U n}.

[0063] Where n is the total number of sampling points, U i is the voltage value of the i-th sampling point, and i is the sampling point number.

[0064] The voltages at photovoltaic and wind power access points come from PMU devices. Since PMU devices provide GPS time synchronization signals, the data at all measurement points can be aligned based on timestamps for subsequent calculations.

[0065] By calculating the voltage statistical characteristics within the sliding time window, the characteristic matrix V(t) for signal analysis is formed. The calculation formula is as follows:

[0066]

[0067] Among them, ΔU i =U i -U i-1 is the voltage change rate of the i-th sampling point, that is, the change difference between adjacent voltage points (can be used to calculate voltage sudden change), Δ 2 U i =ΔU i -ΔU i-1 is the second-order derivative of the voltage at the i-th sampling point, that is, the second-order voltage change rate at the i-th sampling point, reflecting the sudden voltage abnormality, ΔU i-1 is the voltage change rate of the i-1th sampling point, σ(U) i For the i-th sampling point n tThe standard deviation of voltage within one second, used to measure voltage stability.

[0068] S3 specifically includes:

[0069] S31: Based on the grid state parameter data set, the main frequency components of the voltage fluctuation are determined using an improved fast Fourier transform algorithm.

[0070] In one exemplary embodiment, because voltage fluctuations typically contain different frequency components, this application employs an improved Fast Fourier Transform (FFT) algorithm to perform spectral analysis on the voltage fluctuation signal and identify the primary frequency components. A Hanning window can be used to reduce spectral leakage and improve low-frequency signal resolution, while variational mode decomposition can be combined to remove harmonic interference, resulting in more accurate FFT calculation results.

[0071] The main frequency component f of voltage fluctuation is obtained by improved FFT calculation dom , that is, the dominant frequency of voltage anomaly, if f dom < the first frequency threshold, indicating that the voltage fluctuation is caused by the low-frequency output fluctuation of photovoltaic / wind power; if f dom >The second frequency threshold indicates that there is harmonic pollution or interference from power electronic equipment in the power grid; if the first frequency threshold <f dom < the second frequency threshold, it may be caused by reactive power fluctuation or switching process.

[0072] S32: Performing time-frequency analysis using wavelet transform according to the grid state parameter data set.

[0073] Since the improved FFT is suitable for periodic signal analysis, but wind power and photovoltaic output fluctuations are often short-term non-periodic changes, wavelet transform (WT) is introduced for time-frequency analysis to detect voltage mutation points and local fluctuation trends.

[0074] S33: Determine the abnormal feature vector based on the main frequency components and the time-frequency analysis results.

[0075] Based on the calculation results of the improved FFT and wavelet transform, the key voltage anomaly features are extracted and the anomaly feature vector F(Q) is constructed. The calculation formula is as follows:

[0076] F(Q)={f dom ,E w ,σ(U),Δ 2 U,PF}.

[0077] Among them, E w is the local wavelet energy, which indicates the intensity of voltage anomaly and is calculated by wavelet transform. σ(U) is the voltage standard deviation, Δ 2 U is the second-order voltage change rate, and PF is the power factor.

[0078] Second-order voltage change rate Δ 2 U is mainly used to detect sudden voltage disturbances, such as voltage fluctuations caused by short-term impact load access or sudden changes in wind speed.

[0079] The power factor PF can be calculated using the following formula:

[0080]

[0081] Where P is active power and Q is reactive power. If PF is too low, it indicates insufficient reactive power compensation in the system, which may reduce voltage stability. If PF is too high, it indicates that reactive power demand is low and no additional compensation is needed.

[0082] In one exemplary embodiment, an improved FFT algorithm is used to quickly extract voltage anomaly signals from a grid state parameter dataset, and combined with wavelet transforms to improve the accuracy of identifying short-term voltage fluctuations, making it suitable for complex disturbances caused by fluctuations in renewable energy output. This application can not only identify voltage fluctuations, but also generate abnormal feature vectors to further distinguish short-term voltage disturbances from long-term voltage deviations, providing basic data support for subsequent fault tracing and optimizing scheduling parameters. Compared with traditional static voltage monitoring methods, this application can analyze dynamic change trends in real time, improve the accuracy of voltage anomaly identification, avoid misjudgments, and enhance the safety of grid operation.

[0083] S4: According to the abnormal feature vector, the probabilistic reasoning method based on the Bayesian network is used to determine the fault source type and the corresponding fault source for tracing.

[0084] In an exemplary embodiment, the abnormal feature vector is analyzed to identify the type of voltage anomaly, and the type of fault source and the corresponding fault source for tracing are determined. Traditional methods often rely on fixed threshold judgments, which makes it difficult to accurately distinguish short-term voltage disturbances from persistent voltage deviations. This application is based on a data-driven Bayesian network, combining historical grid state parameters and current voltage anomaly types to achieve accurate identification of fault sources and intelligent optimization scheduling. Historical grid state parameters provide the operating status of the grid in different time periods, which is of great significance for tracing the source of the fault, and thus accurately distinguish between short-term voltage disturbances and persistent voltage deviations.

[0085] In an exemplary embodiment, the pre-processed abnormal feature vector is used as an input variable of a Bayesian network to determine the fault source type and the corresponding fault source for tracing.

[0086] A Bayesian network is a probabilistic graphical model consisting of nodes and directed edges. Nodes represent different grid state variables (such as voltage deviation, voltage change rate, and reactive power distribution), while directed edges represent the causal relationships between variables.

[0087] In the Bayesian network model, different fault sources generate different types of voltage anomalies, and these patterns are learned using training data derived from historical grid state parameters, operational event logs, and field measurement data.

[0088] Specifically, the abnormal feature vectors are classified and analyzed to distinguish two types of voltage anomalies, which include short-term voltage disturbances and persistent voltage deviations.

[0089] Short-term voltage disturbances typically last from milliseconds to seconds, and the corresponding fault sources include: photovoltaic cloud shadow effects, sudden wind speed changes, high-power load switching, etc. Fault sources caused by these types of faults include: high-rate voltage changes, rapid fluctuations in wavelet energy in a short period of time, small changes in low-frequency components, and sharp short-term voltage changes.

[0090] Sustained voltage deviations can last from minutes to hours. The corresponding fault source types include: long-term fluctuations in wind power / photovoltaic power, lack of reactive power, adjustments to grid operation modes, and the impact of power electronic equipment. Fault sources caused by these fault source types include: continuous exceeding of voltage offset amplitude, low-frequency disturbances, and high grid harmonic distortion rate components.

[0091] This application uses a probabilistic reasoning method based on Bayesian networks, combining historical grid state parameters and real-time measurement data to determine the current voltage anomaly type, that is, to calculate the probability of short-term voltage disturbances and persistent voltage deviations, and select the category corresponding to the maximum probability. For example:

[0092] If P(T1F(Q))>P(T2F(Q)), it is judged as a continuous voltage deviation; otherwise, it is judged as a short-term voltage disturbance, where T1 is the short-term voltage disturbance threshold and T2 is the continuous voltage deviation threshold.

[0093] After inputting the abnormal feature vector, the Bayesian network traces the fault source through conditional probability, P(G x |F(Q)) represents the various types of fault sources G given the abnormal feature vector. x Probability of occurrence; G x Indicates the xth type of fault source; the fault source corresponding to the maximum probability is selected as the tracing result.

[0094] There is a corresponding relationship between the voltage anomaly type and the fault source type, that is, different fault source types will lead to different voltage anomaly types.

[0095] S5: Generate corresponding optimized scheduling parameters based on the traced fault source.

[0096] In an exemplary embodiment, corresponding optimized dispatching parameters are generated according to the traced fault source to adjust photovoltaic and wind power dispatching strategies and reactive power compensation strategies, thereby improving the resilience of power grid power quality.

[0097] Specifically, if the fault source is caused by long-term fluctuations in photovoltaic power generation, the maximum power point tracking (MPPT) control strategy is adjusted to avoid drastic changes in power output; and the following scheduling parameters are optimized:

[0098] 1. Reduce the power tracking step size to avoid drastic changes in output power.

[0099] 2. Set a reasonable power ramp rate to suppress short-term power fluctuations.

[0100] 3. Adjust the active power regulation rate to improve the ability to adapt to voltage anomalies.

[0101] If the fault is caused by a sudden change in wind speed, use the wind turbine inertia adjustment to optimize the following scheduling parameters:

[0102] 1. The fan inertia adjustment coefficient uses the fan inertia to buffer the power fluctuation caused by sudden changes in wind speed and optimize the power change curve.

[0103] 2. Set wind turbine output change rate constraints to smooth wind power output and reduce grid impact.

[0104] If the fault source is caused by lack of reactive power, optimize the following dispatch parameters:

[0105] 1. Dynamically adjust the output of the static VAR compensation device to improve the voltage support capability.

[0106] 2. Optimize the distribution of reactive power compensation in different areas of the Flexible Alternative Current Transmission System (FACTS) to improve reactive power support efficiency.

[0107] 3. Adjust the reactive reserve capacity according to the load fluctuation of the power grid to improve the stability of the power grid under different operating conditions.

[0108] If the fault source is caused by high-power load switching, optimize the following scheduling parameters:

[0109] 1. Optimize the switching strategy of capacitor banks or dynamic reactive compensation equipment (such as dynamic voltage restorers and static synchronous compensators) to improve local voltage stability.

[0110] 2. Adjust the start-up and shutdown sequence of reactive power compensation equipment in different load areas of the power grid to reduce the impact of load fluctuations on voltage.

[0111] 3. Set the dynamic adjustment threshold of reactive power compensation on the load side to improve the adaptive ability of reactive power compensation.

[0112] This application can intelligently analyze the source of the fault, distinguish different types of voltage anomalies, and trace back the specific equipment or new energy output changes that caused the voltage anomaly, thereby improving the intelligence level of power grid anomaly analysis. Based on the traceability analysis, this application can automatically adjust the reactive compensation strategy for different types of voltage fluctuation problems, realize intelligent optimization and scheduling of new energy power fluctuations, and improve the stability of power grid operation. Through precise fault location and optimized scheduling strategy, this application can significantly shorten the abnormal voltage recovery time, reduce the losses caused by power quality problems in the power grid, and improve the overall power supply reliability.

[0113] S6: Use optimized dispatch parameters to adjust photovoltaic and wind power dispatch strategies and reactive power compensation strategies.

[0114] S6 specifically includes:

[0115] S61: According to the optimized dispatch parameters, based on historical load data, real-time load data and weather data, the photovoltaic and wind power dispatch strategies are adjusted using a short-term load forecasting model.

[0116] S61 specifically includes:

[0117] S611: According to the optimized dispatch parameters, based on historical load data and real-time load data, a short-term load forecasting model is used to predict future load changes.

[0118] Historical load data includes historical active power data, historical reactive power data and load curves. The load curves include load variation patterns and load peak-valley characteristics on weekdays, holidays and different time periods. Load peak-valley characteristics include load variation trends during the day and at night. Real-time load data includes real-time active power, real-time reactive power, voltage, frequency, power factor and current.

[0119] To avoid voltage deviations or frequency drifts caused by sudden load changes, in an exemplary embodiment, a short-term load forecasting model, such as a long short-term memory network (LSTM) or an autoregressive integrated moving average (ARIMA) time series regression model, is used to predict future load change trends based on historical load data and real-time load data. The prediction results are used to make forward-looking adjustments to photovoltaic and wind power scheduling strategies to avoid problems of advanced or lagging regulation.

[0120] S612: Based on weather data, predict future changes in photovoltaic and wind power output.

[0121] Since the volatility of renewable energy is mainly affected by weather factors, meteorological data needs to be considered during the optimization scheduling process to adjust the photovoltaic and wind power scheduling strategies.

[0122] Among them, the main meteorological data include: light intensity and sunshine duration, wind speed and direction, ambient temperature and humidity, weather conditions, etc.; if the light intensity decreases or the weather is cloudy, the photovoltaic output is predicted to decrease, and it is necessary to increase energy storage charging or increase wind power output in advance to compensate for the power gap; if the wind speed is predicted to increase, it indicates that the wind power will increase, which may cause grid voltage fluctuations, and it is necessary to adjust the energy storage discharge strategy or optimize the wind turbine pitch angle to reduce power disturbances; if the weather is stormy, it indicates that the wind power fluctuations are severe, and it is necessary to reduce the wind power grid-connected power or optimize inertia control to maintain stability.

[0123] S613: Adjust the photovoltaic and wind power dispatch strategies based on the future load change prediction results and the future photovoltaic and wind power output change prediction results.

[0124] This application dynamically adjusts photovoltaic and wind power dispatch strategies based on factors such as weather and load fluctuations, ensuring the grid maintains high stability under various operating conditions. Through predictive load adjustment, this application improves the coordination of wind and solar output, reduces the scheduling difficulties caused by fluctuations in renewable energy output, and enhances renewable energy absorption capacity. Specifically, this application employs a layered optimization dispatch strategy that proactively adjusts equipment operating status before power fluctuations occur, by predicting future load changes and future changes in photovoltaic and wind power output, ensuring power allocation meets current load demands and avoiding sudden voltage deviations.

[0125] S62: According to the optimized dispatching parameters, the reactive power compensation strategy is adjusted using a dynamic fuzzy decision-making method.

[0126] In order to adjust the reactive power compensation amount within milliseconds, in an exemplary embodiment, the voltage fluctuation is first ensured to be within the allowable range. Then, a dynamic fuzzy decision method is used to calculate the reactive power adjustment amount ΔQ. The calculation formula is as follows:

[0127] ΔQ=k1·f dom +k2·E w +k3·σ(U)+k4·Δ 2 U+k5·PF.

[0128] Among them, k1 is the first weight, k2 is the second weight, k3 is the third weight, k4 is the fourth weight, and k5 is the fifth weight, which is optimized based on historical load data.

[0129] The target output power of each compensation device is determined according to the reactive power adjustment amount, and the final compensation instruction is generated. The compensation equipment includes: reactive compensation device, distributed energy storage device and flexible AC transmission device.

[0130] If the voltage fluctuation is mainly low-frequency disturbance, the dynamic reactive power compensation device should be adjusted first. Low-frequency disturbance is caused by long-term fluctuation of wind power / photovoltaic power, lack of reactive power and other factors, which leads to significant changes in the low-frequency component of voltage. It belongs to the type of continuous voltage deviation. If the short-term voltage mutation is severe, the distributed energy storage device should be called upon to provide reactive power support first. The short-term voltage mutation is caused by the rapid and short-term events such as photovoltaic cloud shadow effect, sudden change of wind speed, and high-power load switching. It belongs to the type of short-term voltage disturbance. If the harmonic distortion rate component of the power grid is high, fine adjustment should be carried out in combination with FACTS. The high harmonic distortion rate component of the power grid is caused by the influence of power electronic equipment (such as inverter) and it belongs to the type of continuous voltage deviation.

[0131] Furthermore, after receiving the final compensation instruction, the reactive compensation device is controlled within milliseconds.

[0132] The compensation current calculation formula is as follows:

[0133]

[0134] Where U is the voltage at the access point of the compensation device, and pulse width modulation (PWM) control is used to adjust the compensation current to ensure that the reactive power accurately matches the voltage demand.

[0135] If the voltage suddenly drops, the energy storage device is instructed to quickly release reactive power, enhancing the grid's support capacity. If the voltage is too high, the energy storage device is instructed to absorb reactive power, reducing voltage overshoot. Adjustment of series / parallel compensation devices improves the grid's dynamic regulation capabilities. Ultimately, the amplitude of voltage fluctuations is effectively reduced, and power quality is enhanced.

[0136] Compared to traditional fixed reactive power compensation strategies, this application calculates compensation requirements based on real-time data and utilizes advanced equipment such as FACTS to achieve millisecond-level responses, improving the dynamic adaptability of compensation control. Through dynamic fuzzy decision-making, this application can adaptively adjust the compensation amount to ensure optimal compensation results, avoid overcompensation or compensation lag, and improve grid voltage quality. When voltage fluctuations are caused by the integration of new energy sources, it can precisely control reactive power compensation equipment, reduce voltage drops, improve power quality, and enhance the stability and reliability of grid operation.

[0137] In an exemplary embodiment, a power quality management system is provided, the power quality management system comprising:

[0138] A grid state parameter acquisition module is used to obtain grid state parameters of photovoltaic and wind power access points; the grid state parameters include: voltage, current, frequency and phase angle;

[0139] A power grid state parameter data set determination module is used to determine a power grid state parameter data set based on the power grid state parameter and historical power grid state parameters; the historical power grid state parameters include: historical voltage, historical current, historical frequency and historical phase angle;

[0140] an abnormal feature vector determination module, configured to determine an abnormal feature vector based on the power grid state parameter data set by utilizing an improved fast Fourier transform algorithm and a wavelet transform;

[0141] A fault source tracing module is used to determine the fault source type and the corresponding fault source based on the abnormal feature vector and the probabilistic reasoning method of the Bayesian network;

[0142] The optimization scheduling parameter determination module is used to generate the corresponding optimization scheduling parameters according to the traced fault source;

[0143] The strategy adjustment module is used to adjust the photovoltaic and wind power scheduling strategies and reactive power compensation strategies by optimizing scheduling parameters.

[0144] This application achieves power quality assurance and improved grid operation stability when new energy is connected to the grid. Compared with traditional power quality management methods, this application has higher real-time, accuracy and intelligence levels. This application uses PMU equipment to achieve millisecond-level power quality monitoring, ensuring real-time perception of the grid status and improving the accuracy of anomaly detection; secondly, the improved FFT and wavelet transform methods can efficiently extract voltage fluctuation characteristics and improve the accuracy of voltage anomaly identification; at the same time, the use of Bayesian networks for source tracing analysis can accurately locate the source of grid faults and optimize the dispatch of new energy output, effectively improving the power quality recovery capability. In addition, this application combines FACTS technology to achieve rapid reactive power compensation, reduce voltage fluctuations and improve power supply quality; combined with weather and load forecast data, it optimizes wind and solar output, improves the new energy absorption capacity, and enhances the stability and economy of the grid.

[0145] In summary, this application achieves the optimization of power quality after renewable energy is connected to the power grid, improves the security, stability and economic benefits of the power grid, and provides reliable technical support for high-proportion renewable energy power grids.

[0146] In an exemplary embodiment, a computer device is provided. The computer device may be a server or a terminal. The internal structure diagram thereof may be as follows: Figure 2As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, abbreviated as I / O) and a communication interface. The processor, memory and input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The database of the computer device is used to store power quality management data. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a power quality management method is implemented.

[0147] Those skilled in the art will understand that Figure 2 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0148] In an exemplary embodiment, a computer device is provided, including a memory and a processor. The memory stores a computer program, and the processor implements the steps in the above method embodiments when executing the computer program.

[0149] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0150] In an exemplary embodiment, a computer program product is provided, including a computer program. When the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0151] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, database or other media used in the embodiments provided in this application may include at least one of non-volatile and volatile memory. Non-volatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory may include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM may be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0152] The databases involved in the various embodiments provided herein may include at least one of a relational database and a non-relational database. Non-relational databases may include, but are not limited to, distributed databases based on blockchains. The processors involved in the various embodiments provided herein may include, but are not limited to, general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic units, data processing logic units based on quantum computing, and the like.

[0153] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0154] This document uses specific examples to illustrate the principles and implementation methods of this application. The description of the above examples is only intended to help understand the method and core concept of this application. At the same time, for those skilled in the art, based on the concept of this application, there may be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A power quality management method, characterized in that: The power quality management method comprises: Obtaining grid status parameters of photovoltaic and wind power access points; the grid status parameters include: voltage, current, frequency and phase angle; Determining a grid state parameter data set based on the grid state parameters and historical grid state parameters; the historical grid state parameters include: historical voltage, historical current, historical frequency and historical phase angle; Determining an abnormal feature vector using an improved fast Fourier transform algorithm and wavelet transform according to the power grid state parameter data set; According to the abnormal feature vector, the fault source type and the corresponding fault source are determined based on the probabilistic reasoning method of the Bayesian network; Generate corresponding optimized scheduling parameters based on the traced fault source; The photovoltaic and wind power power dispatch strategies and reactive power compensation strategies are adjusted using optimized dispatch parameters.

2. The power quality management method according to claim 1, characterized in that: Determining the abnormal feature vector based on the power grid state parameter data set by using an improved fast Fourier transform algorithm and wavelet transform specifically includes: determining the main frequency components of the voltage fluctuations using an improved fast Fourier transform algorithm based on the grid state parameter data set; Performing time-frequency analysis using wavelet transform according to the power grid state parameter data set; An abnormal feature vector is determined according to the main frequency components and the time-frequency analysis results.

3. The power quality management method according to claim 2, characterized in that: Determining the abnormal feature vector according to the main frequency components and the time-frequency analysis results specifically includes: Using the formula F(Q)={f dom , E w ,σ(U),Δ 2 U, PF} determine the abnormal feature vector F(Q); where f dom is the main frequency component of voltage fluctuation, E w is the local wavelet energy, σ(U) is the voltage standard deviation, Δ 2 U is the second-order voltage change rate, and PF is the power factor.

4. The power quality management method according to claim 1, characterized in that: The use of optimized scheduling parameters to adjust photovoltaic and wind power scheduling strategies and reactive power compensation strategies specifically includes: Based on the optimized dispatch parameters, a short-term load forecasting model is used to adjust the photovoltaic and wind power dispatch strategies based on historical load data, real-time load data, and weather data, including sunlight intensity, sunshine duration, wind direction, and wind speed. According to the optimized dispatching parameters, the dynamic fuzzy decision-making method is used to adjust the reactive power compensation strategy.

5. The power quality management method according to claim 4, characterized in that: The optimization of scheduling parameters and the use of a short-term load forecasting model to adjust the photovoltaic and wind power scheduling strategies based on historical load data, real-time load data, and weather data specifically include: According to the optimized dispatch parameters, based on historical load data and real-time load data, a short-term load forecasting model is used to predict future load changes; Predict future changes in photovoltaic and wind power output based on weather data; The photovoltaic and wind power power dispatch strategies are adjusted based on the forecast results of future load changes and the forecast results of future photovoltaic and wind power output changes.

6. The power quality management method according to claim 4, characterized in that: The reactive power compensation strategy is adjusted by using a dynamic fuzzy decision-making method based on the optimized dispatch parameters, specifically including: Using the formula ΔQ=k1·f dom +k2·E w +k3·σ(U)+k4·Δ 2 U+k5·PF determines the reactive power adjustment ΔQ; Using the formula Determine the compensation current I comp ; Among them, k1 is the first weight, k2 is the second weight, k3 is the third weight, k4 is the fourth weight, k5 is the fifth weight, f dom is the main frequency component of voltage fluctuation, E w is the local wavelet energy, σ(U) is the voltage standard deviation, Δ 2 U is the second-order voltage change rate, PF is the power factor, and U is the voltage at the compensation device access point.

7. A power quality management system, characterized in that: The power quality management system includes: A grid state parameter acquisition module is used to obtain grid state parameters of photovoltaic and wind power access points; the grid state parameters include: voltage, current, frequency and phase angle; A power grid state parameter data set determination module is used to determine a power grid state parameter data set based on the power grid state parameter and historical power grid state parameters; the historical power grid state parameters include: historical voltage, historical current, historical frequency and historical phase angle; an abnormal feature vector determination module, configured to determine an abnormal feature vector based on the power grid state parameter data set by utilizing an improved fast Fourier transform algorithm and a wavelet transform; A fault source tracing module is used to determine the fault source type and the corresponding fault source based on the abnormal feature vector and the probabilistic reasoning method of the Bayesian network; The optimization scheduling parameter determination module is used to generate the corresponding optimization scheduling parameters according to the traced fault source; The strategy adjustment module is used to adjust the photovoltaic and wind power scheduling strategies and reactive power compensation strategies by optimizing scheduling parameters.

8. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the power quality management method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the power quality management method according to any one of claims 1 to 6 is implemented.

10. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the power quality management method according to any one of claims 1 to 6 is implemented.

Images