An automatic voltage control method based on active power trend determination

Abstract

The present invention relates to the field of voltage control technology, and in particular to an automatic voltage control method based on active power trend determination. The method comprises the following steps: performing multi-scale monitoring data acquisition according to a power grid management system to generate multi-scale power matrix data; performing new energy output power prediction according to the multi-scale power matrix data to generate new energy power prediction data; performing spatiotemporal node similarity processing according to the multi-scale power matrix data, and performing node voltage change trend analysis to obtain node voltage trend data; performing power grid voltage stability margin assessment according to the new energy power prediction data and the node voltage trend data to generate voltage stability margin assessment data; performing intelligent voltage regulation instruction generation based on the voltage stability margin assessment data to obtain intelligent voltage regulation instruction data. The present invention realizes automatic control of power grid voltage through power grid voltage stability margin assessment, and effectively solves the voltage stability problem caused by new energy output fluctuations.

Description

Technical Field

[0001] The present invention relates to the technical field of voltage control, and in particular to an automatic voltage control method based on active power trend determination. Background Art

[0002] With the continuous expansion of the scale of power systems and the increasing load, the voltage control problem of power systems has become more and more complex and important. The purpose of voltage control is to maintain the voltage of each node in the power system within the allowable range, ensure the safe and stable operation of the power system, and improve the quality of power supply. The proportion of distributed renewable energy (such as solar energy, wind energy, etc.) in the power system has gradually increased, promoting the diversification of power production, but due to the uncertainty and volatility of its power generation characteristics, voltage control has become more complicated. The access of distributed renewable energy leads to significant fluctuations in the voltage level of the power grid at different times and locations, which will cause problems such as voltage over-limit and equipment overload. Therefore, how to effectively control the voltage and ensure the safe and stable operation of the power system has become a technical problem that needs to be solved urgently. However, traditional voltage control methods mainly rely on local voltage deviations for adjustment. For example, when the voltage of a certain node is lower than the set value, the voltage is increased by adjusting the reactive compensation equipment (such as shunt capacitors, reactors) or generator excitation near the node. These methods often cannot cope with the voltage fluctuations caused by transient disturbances, and when the proportion of renewable energy grid connection is high, it is difficult to meet the needs of real-time response, resulting in poor control effect. Summary of the invention

[0003] Based on this, the present invention provides an automatic voltage control method based on active power trend determination to solve at least one of the above technical problems.

[0004] To achieve the above object, an automatic voltage control method based on active power trend determination comprises the following steps:

[0005] Step S1: Processing the power grid topology structure according to the power grid management system to generate power grid topology structure data; collecting multi-scale monitoring data based on the power grid topology structure data to generate multi-scale power matrix data;

[0006] Step S2: extracting new energy node status from the multi-scale power matrix data to obtain new energy monitoring status data; predicting new energy output power based on the new energy monitoring status data to generate new energy power prediction data;

[0007] Step S3: performing line impedance analysis according to the power grid topology data to obtain power grid impedance topology data; performing spatiotemporal node similarity processing on the multi-scale power matrix data based on the power grid topology data to generate spatiotemporal similarity vector data;

[0008] Step S4: identifying voltage change sensitive nodes on the spatiotemporal similarity vector data through the grid impedance topology data to generate key voltage sensitive node data; performing node voltage change trend analysis based on the key voltage sensitive node data to obtain node voltage trend data;

[0009] Step S5: evaluating the voltage stability margin of the power grid according to the new energy power prediction data and the node voltage trend data, and generating voltage stability margin evaluation data; generating a voltage control signal based on the voltage stability margin evaluation data, and generating node voltage control signal data;

[0010] Step S6: generating an intelligent voltage regulation instruction according to the node voltage control signal data to obtain intelligent voltage regulation instruction data.

[0011] With the support of the power grid management system, the present invention can clearly depict the structure of the power grid and the connection relationship between each node by processing the power grid topology. Based on the power grid topology data, multi-scale monitoring data collection can be carried out, and power data from different levels and regions can be comprehensively collected. The collection of scale monitoring data covers different time and space dimensions of the power grid, forming a comprehensive power matrix data. Extracting the state of new energy nodes from the multi-scale power matrix data is helpful to independently analyze and understand the impact of these highly volatile energy sources on the power grid, and can accurately predict the future new energy power generation. By processing the multi-scale power matrix data with spatiotemporal node similarity, the similarity of the operating states of each node in the power grid in different time periods can be revealed, helping to identify potential risk areas and nodes. Voltage change sensitive node identification is performed on the spatiotemporal similarity vector data through the power grid impedance topology data, and the key nodes that cause voltage instability in the power grid can be effectively identified, ensuring that the power grid maintains a stable voltage level during operation. The power grid voltage stability margin is evaluated according to the new energy power prediction data and the node voltage trend data, providing a safety margin for the operation of the power grid, ensuring that the power grid can maintain normal operation under emergency conditions, and providing timely voltage regulation instructions for the power grid. According to the node voltage control signal data, intelligent voltage regulation instructions are generated, which can realize the automatic regulation of the grid voltage, improve the automation and intelligent management capabilities of the grid, and further promote the efficient utilization and consumption of renewable energy. Therefore, an automatic voltage control method based on active power trend judgment of the present invention constructs multi-scale power matrix data, integrates spatiotemporal correlation information, identifies key nodes sensitive to voltage changes and their active power change trends, and combines new energy power prediction to evaluate voltage stability margin, thereby generating a forward-looking voltage control strategy, effectively predicting and responding to voltage fluctuations caused by transient disturbances, and coordinating control of new energy output fluctuations, ultimately improving the real-time and effectiveness of voltage control, and solving the problem of poor control effect in scenarios with high penetration of new energy.

[0012] Preferably, step S1 comprises the following steps:

[0013] Step S11: performing monitoring power node analysis according to the power grid management system to obtain monitoring power node data, wherein the monitoring power node data includes new energy node data and power load node data;

[0014] Step S12: Processing the power grid topology structure based on the monitored power node data to generate power grid topology structure data;

[0015] Step S13: performing time scale correlation analysis according to the power grid topology data, and performing adaptive time scale selection to obtain multi-scale time window data;

[0016] Step S14: performing multi-scale monitoring data collection on the monitoring power node data through the multi-scale time window data to obtain multi-scale power monitoring data; performing multi-scale vector construction according to the multi-scale power monitoring data to generate multi-scale monitoring vector data;

[0017] Step S15: performing data cleaning processing on the multi-scale power monitoring data, and extracting monitoring power features to generate monitoring node power feature data;

[0018] Step S16: Reconstruct the monitoring state matrix of the monitoring node power characteristic data using the multi-scale monitoring vector data to generate multi-scale power matrix data.

[0019] The present invention monitors power node analysis according to the power grid management system, can identify the status of each monitoring node in the power grid, and is helpful to identify the relationship between renewable energy generation and power load. Based on the monitoring power node data, the power grid topology structure is processed, and the actual operation network of the power grid can be constructed, which can reflect the physical characteristics of the power grid. According to the power grid topology structure data, time scale correlation analysis can be performed to identify the relationship between different time scales in the operation of the power grid. The process of adaptive time scale selection ensures that the analysis can select the most appropriate time window according to the actual operation situation. Through multi-scale time window data, multi-scale monitoring data collection is performed on the monitoring power node data, and comprehensive monitoring of the power grid operation status can be achieved. By extracting the monitoring power features, important feature information in the power grid operation can be identified, including the power load change trend, renewable energy generation characteristics, etc. The monitoring state matrix is ​​reconstructed by using multi-scale monitoring vector data to monitor the power feature data of the monitoring node, which can effectively integrate the power feature information at different time scales.

[0020] Preferably, step S2 comprises the following steps:

[0021] Step S21: extracting new energy node status according to multi-scale power matrix data to obtain new energy monitoring status data;

[0022] Step S22: performing multi-scale time series feature fusion according to the new energy monitoring status data to generate time series monitoring feature fusion data;

[0023] Step S23: Calculate historical power generation based on the time series monitoring feature fusion data to obtain multi-scale power generation data;

[0024] Step S24: performing spatiotemporal new energy output analysis on the time series monitoring feature fusion data through multi-scale power generation data to generate new energy output related data;

[0025] Step S25: performing long-term and short-term power forecasts of new energy sources according to the new energy output correlation data and the power grid topology data, and generating short-term power forecast data and long-term power forecast data respectively;

[0026] Step S26: Perform multi-time scale integration processing on the short-term power prediction data and the long-term power prediction data to generate new energy power prediction data.

[0027] The present invention extracts the state of new energy nodes according to multi-scale power matrix data, can effectively identify and monitor the operating status of new energy nodes, and ensure that the access and operation of new energy can match the power grid demand. According to the new energy monitoring state data, multi-scale time series feature fusion is performed, and the feature information at different time scales can be effectively integrated. Based on the time series monitoring feature fusion data, the historical power generation is calculated, and the power generation capacity of new energy nodes in the past time period can be accurately evaluated. Through the multi-scale power generation data, the time series monitoring feature fusion data is analyzed for spatiotemporal new energy output, which can reveal the changing law of new energy power generation under different time and space conditions. According to the new energy output correlation data and the power grid topology data, the long-term and short-term power forecast of new energy is carried out. The short-term power forecast can provide real-time power generation capacity estimation for power grid dispatching, while the long-term power forecast provides a forward-looking reference for power grid planning and strategy formulation. The short-term power forecast data and the long-term power forecast data are integrated and processed at multiple time scales, and the prediction results of different time scales can be effectively combined to improve the accuracy and practicality of the prediction.

[0028] Preferably, step S25 comprises the following steps:

[0029] Step S251: construct a multivariate regression model for the new energy output related data using a gradient descent algorithm to obtain a short-term new energy output prediction model;

[0030] Step S252: training the short-term renewable energy output prediction model through multi-scale power generation data to generate an optimized short-term renewable energy prediction model;

[0031] Step S253: using the optimized short-term new energy prediction model to perform short-term energy power prediction on the time series monitoring feature fusion data to generate short-term power prediction data;

[0032] Step S254: acquiring weather forecast data for new energy nodes based on the power grid topology data;

[0033] Step S255: performing an influencing factor analysis on the new energy output related data through weather forecast data to generate weather influencing factor data;

[0034] Step S256: Using a preset long-term and short-term neural network model to perform transfer learning on the weather influencing factor data and the time series monitoring feature fusion data, and construct a long-term new energy prediction model;

[0035] Step S257: Perform long-term energy power prediction on the short-term power prediction data and the weather influencing factor data through the long-term new energy prediction model to generate long-term power prediction data.

[0036] The present invention uses a gradient descent algorithm to construct a multivariate regression model for the output-related data of new energy, which can effectively capture the relationship between the output of new energy and related factors. By establishing a short-term new energy output prediction model, an accurate short-term power generation capacity estimate can be provided for power grid dispatching. By optimizing the short-term new energy prediction model to predict the short-term energy power of the time series monitoring feature fusion data, the change of the power generation capacity of new energy can be reflected in real time. The weather forecast of the new energy node is obtained based on the power grid topology data, and meteorological information closely related to the power generation of new energy, such as wind speed, temperature and radiation, can be collected. The influencing factor analysis of the output-related data of new energy through weather forecast data can reveal the influence of weather changes on the output of new energy. The prediction ability and scope of application of the model can be improved by using existing data resources. The constructed long-term new energy prediction model can not only consider short-term fluctuations, but also make reasonable predictions on long-term trends. The long-term energy power prediction of the short-term power prediction data and the weather influencing factor data can provide the power grid with a power generation capacity estimate for a long period of time in the future.

[0037] Preferably, step S3 comprises the following steps:

[0038] Step S31: extracting node geographic coordinates according to the power grid topology data to generate power grid node coordinate data;

[0039] Step S32: performing inter-node distance matrix processing according to the grid node coordinate data to generate inter-node distance matrix data;

[0040] Step S33: performing line impedance matrix processing according to the power grid topology data to generate line impedance matrix data and power grid impedance topology data;

[0041] Step S34: calculating the spatial correlation coefficient of the line impedance matrix data through the node distance matrix data to generate spatial correlation matrix data;

[0042] Step S35: Calculate the time series mutual information of the multi-scale power matrix data to obtain time correlation matrix data;

[0043] Step S36: performing spatiotemporal correlation fusion according to the spatial correlation matrix data and the temporal correlation matrix data to generate spatiotemporal correlation tensor data;

[0044] Step S37: Performing spatiotemporal weighted similarity processing on the multi-scale power matrix data through the spatiotemporal correlation tensor data to generate spatiotemporal similarity vector data.

[0045] The present invention extracts node geographic coordinates according to the grid topology data, can obtain the specific location of each node in the grid, and helps to identify the physical relationship between the nodes. According to the grid node coordinate data, the node distance matrix is ​​processed, and the distance relationship between each node of the grid can be calculated, which helps to identify the relative position and connectivity between each node in the grid, and helps the grid dispatch and optimization. According to the grid topology data, the line impedance matrix is ​​processed, and the influence of each line in the grid on the current can be revealed. The spatial correlation coefficient of the line impedance matrix data is calculated by the node distance matrix data, which helps to identify which nodes in the grid are affected by the spatial distance. The time series mutual information calculation of the multi-scale power matrix data can effectively identify the time characteristics of the power load and the power generation capacity. According to the spatial correlation matrix data and the time correlation matrix data, the time-space correlation fusion can be performed, and the information of the two dimensions of space and time can be effectively combined, and the interaction of the grid state in time and space can be comprehensively considered. The multi-scale power matrix data is processed by the time-space weighted similarity through the time-space correlation tensor data, and the similarity of each node in the grid in space and time can be quantified.

[0046] Preferably, step S33 includes the following steps:

[0047] Step S331: extracting power line parameters from the power grid topology data through the power grid management system to obtain power line parameter data;

[0048] Step S332: constructing a distributed line model according to the power line parameter data to obtain electrical line model data;

[0049] Step S333: performing impedance frequency simulation on the electrical circuit model data to generate impedance frequency response data;

[0050] Step S334: mapping the line impedance attributes of the power grid topology data through the impedance frequency response data to generate power grid impedance topology data;

[0051] Step S335: Calculate the inter-node harmonic impedance according to the grid impedance topology data to generate line harmonic impedance data;

[0052] Step S336: Use the line harmonic impedance data to perform impedance sensitivity analysis on the power line parameter data to generate line impedance matrix data.

[0053] The present invention extracts power line parameters from power grid topology data through a power grid management system, and can collect basic physical characteristics and parameter information of power lines. Distributed line models are constructed according to power line parameter data, and the physical characteristics of power lines can be converted into electrical model data. Impedance frequency simulation is performed on electrical line model data, and the impedance characteristics of power lines at different frequencies can be analyzed, and the frequency response characteristics of lines can be revealed, which is helpful to identify harmonics and instability problems that occur under different frequency conditions. Line impedance attribute mapping is performed on power grid topology data through impedance frequency response data, and a detailed description of the overall impedance characteristics of the power grid can be provided, revealing the electrical relationship between each node and line in the power grid. Inter-node harmonic impedance calculation is performed according to power grid impedance topology data, and the response ability of each node in the power grid to harmonic current can be evaluated. Impedance sensitivity analysis is performed on power line parameter data using line harmonic impedance data, and the sensitivity of power lines to different harmonic frequencies can be evaluated, ensuring the stability and reliability of the power grid in the face of distributed new energy access and active power trend changes.

[0054] Preferably, step S4 comprises the following steps:

[0055] Step S41: identifying voltage change sensitive nodes using the time-space similarity vector data through the power grid impedance topology data to generate key voltage sensitive node data;

[0056] Step S42: Calculate the active power change rate of the key voltage sensitive node data to generate active power change rate data;

[0057] Step S43: Calculating the change amplitude according to the active power change rate data to generate active power change amplitude data;

[0058] Step S44: marking the active power variation amplitude data with trend changes by using preset trend evolution label data to generate active power trend evolution label data;

[0059] Step S45: classifying the real-time active power trend mode according to the active power trend evolution label data, and evaluating the voltage impact degree to generate active power trend impact factor data;

[0060] Step S46: Analyze the node voltage change trend of key voltage sensitive node data using the active power trend influencing factor data to obtain node voltage trend data.

[0061] The present invention identifies nodes sensitive to voltage changes by using grid impedance topology data to identify spatiotemporal similarity vector data, and can effectively identify key nodes that are sensitive to voltage changes, ensuring timely response and adjustment when voltage fluctuates. The active power change rate is calculated for key voltage sensitive node data, and the power change speed of these sensitive nodes can be quantified. The change amplitude is calculated based on the active power change rate data to reflect the degree of response of the power grid when the load or power generation changes drastically. The active power change amplitude data is marked with trend changes by preset trend evolution label data, and power changes of different amplitudes can be classified. The real-time mode of active power changes in the power grid is evaluated to identify factors that have a greater impact on voltage changes. The node voltage change trend analysis is performed on the key voltage sensitive node data by using active trend influence factor data, which directly indicates the direction of voltage changes that will occur at these nodes in the future.

[0062] Preferably, step S41 includes the following steps:

[0063] Step S411: performing Z-score standardization processing on the spatiotemporal similarity vector data to generate standard spatiotemporal similarity vector data; performing similarity threshold processing on the standard spatiotemporal similarity vector data to obtain node similarity threshold data;

[0064] Step S412: performing grouping processing of similar feature nodes according to the spatiotemporal similarity vector data, and screening potential sensitive nodes according to the node similarity threshold data, to obtain potential power grid sensitive node data;

[0065] Step S413: Calculate the voltage change rate of the potential power grid sensitive node data to obtain the sensitive node voltage rate data;

[0066] Step S414: using the power grid impedance topology data to perform topology importance assessment on potential power grid sensitive node data to generate topology importance assessment data;

[0067] Step S415: performing dynamic response characteristic analysis on potential power grid sensitive node data to generate dynamic response characteristic data;

[0068] Step S416: performing a multi-criteria comprehensive evaluation process on the potential power grid sensitive node data based on the sensitive node voltage rate data, the topology importance evaluation data, and the dynamic response characteristic data to generate multi-criteria scoring data;

[0069] Step S417: Perform sensitive node verification processing on potential power grid sensitive node data through multi-criteria scoring data, and screen key voltage sensitive nodes to obtain key voltage sensitive node data.

[0070] The present invention performs Z-score standardization on the spatiotemporal similarity vector data, which can eliminate the dimensional influence between different features and make each feature compared under the same standard. According to the spatiotemporal similarity vector data, similar feature nodes are grouped and processed. Through grouping and screening, nodes that cause grid instability under voltage changes and load fluctuations can be effectively identified, ensuring that these potential sensitive nodes can be centrally monitored and regulated in grid management, and improving the safety of the grid. The voltage change rate of potential grid sensitive node data is calculated, and the voltage fluctuation of these nodes during grid operation can be quantified. The topological importance of potential grid sensitive node data is evaluated using grid impedance topology data, and the relative importance of nodes in the grid can be evaluated, revealing which nodes have a greater impact on the overall performance of the grid. The dynamic response characteristic analysis of potential grid sensitive node data can reveal the response characteristics of these nodes under different operating conditions. The response characteristics of these nodes under different operating conditions can be revealed, ensuring that the key nodes that have a greater impact on grid stability can be accurately identified in a complex grid environment. By performing sensitive node verification processing on potential power grid sensitive node data through multi-criteria scoring data, it can ensure that the sensitive nodes finally confirmed have a higher voltage change risk and importance, providing an effective monitoring and control object for power grid management.

[0071] Preferably, step S5 comprises the following steps:

[0072] Step S51: evaluating the voltage stability margin of the power grid according to the new energy power prediction data and the node voltage trend data, and generating voltage stability margin evaluation data;

[0073] Step S52: dynamically adjusting the preset initial voltage control threshold value according to the voltage stability margin evaluation data to obtain a dynamic voltage control threshold value;

[0074] Step S53: using the voltage stability margin assessment data to perform voltage fluctuation analysis on the node voltage trend data, and calculating the voltage deviation prediction value to generate the voltage deviation prediction value;

[0075] Step S54: Calculating the voltage regulation target value according to the voltage deviation prediction value to obtain voltage regulation target value data;

[0076] Step S55: generating a voltage control signal for the voltage regulation target value data through a dynamic voltage control threshold, thereby generating node voltage control signal data.

[0077] The present invention evaluates the voltage stability margin of the power grid according to the power prediction data of new energy and the node voltage trend data, and can comprehensively consider the volatility of new energy power generation and the change of node voltage, so as to ensure that the power grid can maintain a good voltage level in the scenario of large-scale access of new energy, and prevent system failures caused by voltage instability. According to the real-time voltage stability evaluation result, the control threshold is adjusted in time to adapt to the changes in the operating conditions of the power grid. The voltage fluctuation analysis of the node voltage trend data using the voltage stability margin evaluation data can quantitatively predict the deviation of the future voltage, which is helpful to formulate corresponding control strategies in advance when the voltage fluctuation intensifies, so as to ensure that the power grid can respond quickly and maintain voltage stability when facing emergencies. The voltage regulation target value is calculated according to the voltage deviation prediction value, which can provide a clear voltage regulation target for the power grid dispatching, and ensure that a reasonable regulation scheme can be formulated when the voltage deviation occurs. The calculated voltage regulation target value is converted into an actual control instruction, so as to ensure that the control instruction not only meets the safety operation requirements, but also can quickly respond to the changes in the power grid state, thereby improving the response speed and control accuracy of the entire automatic voltage control system.

[0078] Preferably, step S6 comprises the following steps:

[0079] Step S61: Obtaining distribution data of controllable resources of the power grid;

[0080] Step S62: Decomposing the controllable resource distribution data of the power grid into regional voltage control targets by using the power grid impedance topology data to generate regional voltage control resource data;

[0081] Step S63: Utilizing the regional voltage control resource data to optimize the scheduling of the node voltage control signal data, and generating a new energy coordinated control strategy to generate an automatic voltage control strategy;

[0082] Step S64: generating intelligent voltage regulation instructions according to the automatic voltage control strategy to obtain intelligent voltage regulation instruction data.

[0083] The present invention obtains the distribution data of controllable resources in the power grid, which can fully understand the specific location and capabilities of the controllable resources in the power grid. The regional voltage control target is decomposed for the distribution data of controllable resources in the power grid through the power grid impedance topology data, and the overall voltage control target is refined into specific control targets for each region, ensuring that the voltage control strategy is more targeted and operational. The regional voltage control resource data is used to optimize the scheduling of regional controllable resources for the node voltage control signal data, which can achieve the reasonable allocation of controllable resources and ensure that the voltage level of the power grid can be effectively maintained in the case of fluctuations in new energy power generation. Intelligent voltage regulation instructions are generated according to the automatic voltage control strategy, and the formulated voltage control strategy can be converted into specific operation instructions to ensure that the power grid can respond to voltage changes in real time. BRIEF DESCRIPTION OF THE DRAWINGS

[0084] Figure 1 It is a schematic flow chart of the steps of the automatic voltage control method based on active power trend determination of the present invention;

[0085] Figure 2 for Figure 1 Detailed implementation steps of step S2 in the flowchart;

[0086] Figure 3 for Figure 1 Detailed implementation steps of step S5;

[0087] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0088] The technical method of the present invention is described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by technicians in this field without creative work are within the scope of protection of the present invention.

[0089] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0090] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0091] To achieve this, please refer to Figures 1 to 3 The present invention provides an automatic voltage control method based on active power trend determination, comprising the following steps:

[0092] Step S1: Processing the power grid topology structure according to the power grid management system to generate power grid topology structure data; collecting multi-scale monitoring data based on the power grid topology structure data to generate multi-scale power matrix data;

[0093] Step S2: extracting new energy node status from the multi-scale power matrix data to obtain new energy monitoring status data; predicting new energy output power based on the new energy monitoring status data to generate new energy power prediction data;

[0094] Step S3: performing line impedance analysis according to the power grid topology data to obtain power grid impedance topology data; performing spatiotemporal node similarity processing on the multi-scale power matrix data based on the power grid topology data to generate spatiotemporal similarity vector data;

[0095] Step S4: identifying voltage change sensitive nodes on the spatiotemporal similarity vector data through the grid impedance topology data to generate key voltage sensitive node data; performing node voltage change trend analysis based on the key voltage sensitive node data to obtain node voltage trend data;

[0096] Step S5: evaluating the voltage stability margin of the power grid according to the new energy power prediction data and the node voltage trend data, and generating voltage stability margin evaluation data; generating a voltage control signal based on the voltage stability margin evaluation data, and generating node voltage control signal data;

[0097] Step S6: generating an intelligent voltage regulation instruction according to the node voltage control signal data to obtain intelligent voltage regulation instruction data.

[0098] In the embodiment of the present invention, reference Figure 1 The above is a schematic flow chart of the steps of an automatic voltage control method based on active power trend determination of the present invention. In this embodiment, the automatic voltage control method based on active power trend determination includes the following steps:

[0099] Step S1: Processing the power grid topology structure according to the power grid management system to generate power grid topology structure data; collecting multi-scale monitoring data based on the power grid topology structure data to generate multi-scale power matrix data;

[0100] In an embodiment of the present invention, a power grid management system (such as a SCADA system) is used to obtain real-time power grid equipment status information, including circuit breaker status, disconnector status, etc. Then, based on the acquired equipment status information and pre-stored power grid basic data (such as line parameters, transformer parameters, etc.), a graph theory method is used to construct a power grid topology diagram, and it is converted into a data format for storage, such as an adjacency matrix or a node-branch association matrix, which is the power grid topology data. Next, based on the constructed power grid topology, a multi-scale power sensor network (such as a smart meter, a micro-synchronous phasor measurement unit, etc.) is used to collect power data such as voltage, current, power, etc. at different time scales (such as seconds, minutes, hours), and it is organized into a multi-dimensional matrix form according to the collection time and node information, which is the multi-scale power matrix data.

[0101] Step S2: extracting new energy node status from the multi-scale power matrix data to obtain new energy monitoring status data; predicting new energy output power based on the new energy monitoring status data to generate new energy power prediction data;

[0102] In an embodiment of the present invention, the corresponding positions of new energy nodes (such as photovoltaic power stations, wind farms, etc.) in multi-scale power matrix data are identified based on the power grid topology data, and the voltage, current, power and other information of these nodes are extracted to construct new energy monitoring status data. Then, using the extracted new energy monitoring status data, combined with meteorological data (such as irradiance, wind speed, etc.) and historical output data, a suitable new energy output prediction model is selected. For example, for photovoltaic output prediction, a neural network model based on historical data and weather forecasts can be selected. For wind power output prediction, a physical model based on numerical weather forecasts and wind turbine characteristics can be selected. By training and testing the selected prediction model, new energy power prediction data of different time scales in the future period of time is generated.

[0103] Step S3: performing line impedance analysis according to the power grid topology data to obtain power grid impedance topology data; performing spatiotemporal node similarity processing on the multi-scale power matrix data based on the power grid topology data to generate spatiotemporal similarity vector data;

[0104] In an embodiment of the present invention, according to the grid topology data and line parameter information generated in step S1, the line impedance is calculated using a circuit analysis method (such as a node voltage method, a loop current method, etc.), and the calculation result is associated with the grid topology data to generate grid impedance topology data. For example, the resistance and reactance values ​​of each line can be stored in a matrix form and correspond to the line information in the grid topology data. Then, based on the grid topology data, the dynamic time warping (DTW) algorithm or other similarity measurement methods are used to calculate the similarity of power data such as voltage, current, and power of different nodes in the multi-scale power matrix data in different time periods, and the calculation result is constructed as spatiotemporal similarity vector data, in which each element represents the similarity between two nodes in a specific time period.

[0105] Step S4: identifying voltage change sensitive nodes on the spatiotemporal similarity vector data through the grid impedance topology data to generate key voltage sensitive node data; performing node voltage change trend analysis based on the key voltage sensitive node data to obtain node voltage trend data;

[0106] In an embodiment of the present invention, in combination with the grid impedance topology data and the spatiotemporal similarity vector data, a sensitivity analysis method based on graph theory (such as node importance index, network betweenness centrality, etc.) is used to identify key nodes that are more sensitive to voltage changes, and generate key voltage sensitive node data. For example, the electrical distance from each node to other nodes can be calculated, and the nodes that have a greater impact on the propagation of voltage fluctuations can be identified by combining the spatiotemporal similarity vector data of the nodes. Then, according to the key voltage sensitive node data, the voltage time series data of the corresponding node in the multi-scale power matrix data is extracted, and the voltage change trend is analyzed using a time series analysis method (such as a moving average method, an exponential smoothing method, an ARIMA model, etc.), and the node voltage trend data is generated. For example, the voltage change trend of the key node in the future period can be predicted, including the trend of voltage rising, falling or maintaining stability.

[0107] Step S5: evaluating the voltage stability margin of the power grid according to the new energy power prediction data and the node voltage trend data, and generating voltage stability margin evaluation data; generating a voltage control signal based on the voltage stability margin evaluation data, and generating node voltage control signal data;

[0108] In an embodiment of the present invention, based on the new energy power prediction data and the node voltage trend data, combined with the power grid operation constraints (such as voltage upper and lower limits, line transmission capacity, etc.), a static or dynamic voltage stability analysis method (such as PV curve analysis, continuous power flow calculation, etc.) is used to evaluate the power grid voltage stability margin and generate voltage stability margin evaluation data. For example, the voltage stability margin index of each node under different new energy output scenarios can be calculated, such as the difference between the voltage collapse critical point and the current voltage. Then, based on the voltage stability margin evaluation data, a voltage control strategy is designed. For example, according to the size and change trend of the voltage stability margin, fuzzy control, predictive control and other methods can be used to generate node voltage control signal data. For example, according to the reduction speed of the voltage stability margin, a voltage regulation signal for advance control can be generated to avoid voltage over-limit.

[0109] Step S6: generating an intelligent voltage regulation instruction according to the node voltage control signal data to obtain intelligent voltage regulation instruction data.

[0110] In an embodiment of the present invention, specific intelligent voltage regulation instruction data is generated according to the node voltage control signal data, combined with the actual operation status of the power grid and the regulation characteristics of the control device (such as response speed, regulation accuracy, etc.). For example, according to the size and direction of the node voltage control signal, switching instructions for reactive compensation equipment (such as capacitor banks, static var compensators, etc.) can be generated, or regulation instructions for on-load tap-changing transformers, flexible AC transmission equipment, etc. can be generated, thereby achieving precise control of the grid voltage. For example, according to the node voltage control signal, instructions for putting a specific capacitor bank into operation or removing it can be generated, or instructions for adjusting the gear of the on-load tap-changing transformer can be generated, ultimately achieving automatic control of voltage in a distributed new energy consumption scenario.

[0111] Preferably, step S1 comprises the following steps:

[0112] Step S11: performing monitoring power node analysis according to the power grid management system to obtain monitoring power node data, wherein the monitoring power node data includes new energy node data and power load node data;

[0113] Step S12: Processing the power grid topology structure based on the monitored power node data to generate power grid topology structure data;

[0114] Step S13: performing time scale correlation analysis according to the power grid topology data, and performing adaptive time scale selection to obtain multi-scale time window data;

[0115] Step S14: performing multi-scale monitoring data collection on the monitoring power node data through the multi-scale time window data to obtain multi-scale power monitoring data; performing multi-scale vector construction according to the multi-scale power monitoring data to generate multi-scale monitoring vector data;

[0116] Step S15: performing data cleaning processing on the multi-scale power monitoring data, and extracting monitoring power features to generate monitoring node power feature data;

[0117] Step S16: Reconstruct the monitoring state matrix of the monitoring node power characteristic data using the multi-scale monitoring vector data to generate multi-scale power matrix data.

[0118] In an embodiment of the present invention, data is exchanged with a power grid management system (such as a SCADA system, a power distribution management system, etc.) to obtain real-time information of each power node in the power grid. In combination with the power grid geographic information system (GIS) data and the power system network parameter information, a graph theory method is used to construct a power grid topology diagram. First, each monitored power node is mapped to the geographic space, and a node-branch association matrix is ​​constructed based on the connection relationship between the nodes. Then, in combination with the line parameter information, corresponding electrical parameters are assigned to each branch, such as resistance, reactance, line length, etc. Finally, the node-branch association matrix and the line parameter information are integrated to form complete power grid topology data. According to the operation characteristics and analysis objectives of the power system, multiple time scales are preset, such as seconds, minutes, hours, etc., and a time scale combination that can effectively reflect the dynamic characteristics of the power system is selected, and the time window size corresponding to each time scale is determined. The multi-scale monitoring data of the monitoring power node data is collected through multi-scale time window data. Each time scale contains data from the past T moments, such as the second time scale contains data from the past 10 seconds, the minute time scale contains data from the past 10 minutes, and the hour time scale contains data from the past 10 hours. The power data of each node at different time scales are arranged in chronological order to construct multi-scale monitoring vector data. Data cleaning processing is performed on the multi-scale power monitoring data, including missing value processing, outlier detection and processing, etc. Feature extraction is performed on the cleaned data, and statistical features such as the average value, standard deviation, peak-to-valley difference, and power factor of the power load can be extracted. Signal processing methods such as wavelet transform and empirical mode decomposition can also be used to extract frequency characteristics and waveform characteristics of the power load. The extracted multiple features are combined to construct the power feature data of the monitoring node. The multi-scale monitoring vector data and monitoring node power feature data at each time scale are arranged in matrix form, and then the main eigenvectors are extracted using the PCA method, and the original data are projected into the eigenvector space to obtain a low-dimensional feature matrix. Finally, the feature matrices at different time scales are arranged in chronological order to construct multi-scale power matrix data.

[0119] As an example of the present invention, refer to Figure 2 As shown, Figure 1 Detailed implementation steps of step S2 in the embodiment are shown in the flowchart. In this embodiment, step S2 includes:

[0120] Step S21: extracting new energy node status according to multi-scale power matrix data to obtain new energy monitoring status data;

[0121] In an embodiment of the present invention, the corresponding positions of new energy nodes (such as photovoltaic power stations, wind farms, etc.) in multi-scale power matrix data are identified based on the power grid topology data. Then, the power data of these nodes at different time scales, such as active power, reactive power, voltage, current, etc., and other related information, such as the irradiance of photovoltaic power stations, the wind speed of wind farms, etc., are extracted. The extracted data are integrated to form new energy monitoring status data.

[0122] Step S22: performing multi-scale time series feature fusion according to the new energy monitoring status data to generate time series monitoring feature fusion data;

[0123] In an embodiment of the present invention, a multi-scale time series feature fusion method is adopted. For example, a method based on deep learning, such as a combined model of a convolutional neural network (CNN) and a long short-term memory network (LSTM) can be used. First, the spatial features of the new energy monitoring status data at different time scales are extracted using a CNN network. For example, data at the minute level, hour level, and day level can be input into different CNN branches for feature extraction. Then, the extracted multi-scale feature map is input into the LSTM network to learn the temporal dependency between features at different time scales, and finally a feature vector that integrates multi-scale time series information is generated, that is, time series monitoring feature fusion data.

[0124] Step S23: Calculate historical power generation based on the time series monitoring feature fusion data to obtain multi-scale power generation data;

[0125] In an embodiment of the present invention, known historical meteorological data (such as irradiance, wind speed, etc.) and a new energy output model are used, combined with the time series monitoring feature fusion data generated in step S22, to calculate the new energy power generation at different time scales. For example, for photovoltaic power generation, a photovoltaic output prediction model based on a physical model can be used, and historical irradiance, temperature and other meteorological data and photovoltaic component parameters can be input to calculate historical photovoltaic output data; for wind power generation, a wind power output prediction model based on meteorological data such as wind speed, wind direction, air density and wind turbine parameters can be used to calculate historical wind power output data. The historical power generation data calculated at different time scales are integrated to generate multi-scale power generation data.

[0126] Step S24: performing spatiotemporal new energy output analysis on the time series monitoring feature fusion data through multi-scale power generation data to generate new energy output related data;

[0127] In an embodiment of the present invention, the power grid is divided into multiple geographical areas according to the geographical location information of the new energy node. Then, for each area, the sliding window method is used to align the multi-scale power generation data and the time series monitoring feature fusion data in the time dimension to form data samples of multiple time windows. Each sample contains the multi-scale power generation data of the area in a specific time window and the time series monitoring feature fusion data of all nodes. The correlation coefficient between the time series monitoring feature fusion data of each node and the multi-scale power generation data of the area is calculated, such as the Pearson correlation coefficient, the Spearman correlation coefficient, etc. The larger the absolute value of the correlation coefficient, the stronger the correlation between the monitoring data of the node and the new energy output of the area. Identify the features that are significantly correlated with the output of the new energy node, for example: the output of the photovoltaic power station is significantly positively correlated with the sunshine intensity and the temperature; the output of the wind turbine is significantly positively correlated with the wind speed. The new energy output correlation data includes: the correlation between the output of the new energy node and different features; the output change law of the new energy node in different time periods and different locations.

[0128] Step S25: performing long-term and short-term power forecasts of new energy sources according to the new energy output correlation data and the power grid topology data, and generating short-term power forecast data and long-term power forecast data respectively;

[0129] In the embodiment of the present invention, for short-term prediction, a method based on time series analysis, such as ARIMA model, Prophet model, etc., or a method based on machine learning, such as support vector machine, neural network, etc., can be selected. For long-term prediction, a method based on numerical weather forecast data and physical model, or a method based on deep learning, such as recurrent neural network, long short-term memory network, etc. can be selected. According to the prediction target and data characteristics, a suitable prediction model is selected, and the model is trained and optimized to generate short-term power prediction data and long-term power prediction data respectively.

[0130] Step S26: Perform multi-time scale integration processing on the short-term power prediction data and the long-term power prediction data to generate new energy power prediction data.

[0131] In an embodiment of the present invention, for the overlapping time periods of short-term prediction results and long-term prediction results, a time weighting method can be used for data fusion. The closer the prediction result is to the current time, the greater its weight; conversely, the smaller the weight. For example, for the prediction value of the first hour in the future, a weight of 0.8 can be assigned to the short-term prediction result and a weight of 0.2 to the long-term prediction result; for the prediction value of the 24th hour in the future, a weight of 0.2 can be assigned to the short-term prediction result and a weight of 0.8 to the long-term prediction result. Through the time weighting method, the accuracy advantage of short-term prediction and the trend guidance of long-term prediction are combined to generate more accurate and smoother new energy power prediction data.

[0132] Preferably, step S25 comprises the following steps:

[0133] Step S251: construct a multivariate regression model for the new energy output related data using a gradient descent algorithm to obtain a short-term new energy output prediction model;

[0134] Step S252: training the short-term renewable energy output prediction model through multi-scale power generation data to generate an optimized short-term renewable energy prediction model;

[0135] Step S253: using the optimized short-term new energy prediction model to perform short-term energy power prediction on the time series monitoring feature fusion data to generate short-term power prediction data;

[0136] Step S254: acquiring weather forecast data for new energy nodes based on the power grid topology data;

[0137] Step S255: performing an influencing factor analysis on the new energy output related data through weather forecast data to generate weather influencing factor data;

[0138] Step S256: Using a preset long-term and short-term neural network model to perform transfer learning on the weather influencing factor data and the time series monitoring feature fusion data, and construct a long-term new energy prediction model;

[0139] Step S257: Perform long-term energy power prediction on the short-term power prediction data and the weather influencing factor data through the long-term new energy prediction model to generate long-term power prediction data.

[0140] In an embodiment of the present invention, the historical renewable energy power generation data under the corresponding time scale is used as the output variable to construct a multivariate regression model. The model can consider various forms such as linear regression, ridge regression, Lasso regression, and select a suitable model according to the characteristics of the data. The model is constructed by using a gradient descent algorithm for parameter optimization. For example, batch gradient descent, random gradient descent, or small batch gradient descent can be used to iteratively update the model parameters until the loss function of the model converges to a preset threshold range, thereby obtaining a preliminary short-term renewable energy output prediction model. The multi-scale power generation data is divided into a training set and a validation set by a sliding window method, and the model parameters are fine-tuned using the training set data, and the prediction performance of the model is evaluated using the validation set data, and finally the optimized short-term renewable energy prediction model is obtained. The time series monitoring feature fusion data is input into the optimized short-term renewable energy prediction model to obtain the renewable energy output prediction result for a period of time in the future (for example, the next 2 hours), that is, the short-term power prediction data. According to the geographical location information of the new energy nodes in the power grid topology data, call the external weather forecast interface (such as the API interface of the National Meteorological Administration, the API interface of the commercial meteorological data provider, etc.) to obtain the weather forecast data of the location of the new energy node in the future (for example, the next 5 days). Identify the weather factors that have a significant impact on the output of new energy and their degree of influence. For example, statistical indicators such as the correlation coefficient and partial correlation coefficient between different weather factors and the output of new energy can be calculated, or a model based on machine learning can be constructed to quantitatively analyze the degree of influence of different weather factors on the output of new energy. Use the weather influence factor data and the time series monitoring feature fusion data as input to perform transfer learning on the LSTM model. For example, the weather influence factor data can be used as an additional input of the LSTM model, or the weather influence factor data can be spliced ​​with the time series monitoring feature fusion data as the input of the LSTM model, and the parameters of the LSTM model can be fine-tuned according to the new energy output correlation data to adapt it to a specific new energy prediction scenario. Input the short-term power prediction data and the weather influence factor data into the long-term new energy prediction model to obtain the new energy output prediction results for a longer period of time in the future (for example, the next 5 days), that is, the long-term power prediction data.

[0141] Preferably, step S3 comprises the following steps:

[0142] Step S31: extracting node geographic coordinates according to the power grid topology data to generate power grid node coordinate data;

[0143] Step S32: performing inter-node distance matrix processing according to the grid node coordinate data to generate inter-node distance matrix data;

[0144] Step S33: performing line impedance matrix processing according to the power grid topology data to generate line impedance matrix data and power grid impedance topology data;

[0145] Step S34: calculating the spatial correlation coefficient of the line impedance matrix data through the node distance matrix data to generate spatial correlation matrix data;

[0146] Step S35: Calculate the time series mutual information of the multi-scale power matrix data to obtain time correlation matrix data;

[0147] Step S36: performing spatiotemporal correlation fusion according to the spatial correlation matrix data and the temporal correlation matrix data to generate spatiotemporal correlation tensor data;

[0148] Step S37: Performing spatiotemporal weighted similarity processing on the multi-scale power matrix data through the spatiotemporal correlation tensor data to generate spatiotemporal similarity vector data.

[0149] In an embodiment of the present invention, the power grid topology data generally includes the geographical location information of each node, such as longitude and latitude coordinates. According to the data format, the power grid topology data is parsed, the longitude and latitude coordinates of each node are extracted, and stored as power grid node coordinate data. The geographical distance between any two nodes is calculated using a geographic distance calculation formula (such as the Haversine formula or the Vincenty formula). The calculated distance value is stored in a matrix, the rows and columns of the matrix represent different nodes, and the matrix elements represent the distances between the corresponding nodes, forming node distance matrix data. According to the line connection information, a matrix corresponding to the power grid topology diagram is constructed, the rows and columns of the matrix represent different nodes, and if there is a line connection between the two nodes, the corresponding matrix element is 1, otherwise it is 0. Then, according to the line parameter information, the non-zero elements in the matrix are replaced with the impedance values ​​of the corresponding lines to form line impedance matrix data. At the same time, the line impedance information is associated with the power grid topology diagram to construct the power grid impedance topology data. Using the node distance matrix data and the line impedance matrix data, the correlation coefficient between the distance between each two nodes and the corresponding line impedance is calculated, such as the Pearson correlation coefficient, the Spearman correlation coefficient, etc. The calculated correlation coefficients are stored in a matrix, where the rows and columns of the matrix represent different nodes, and the matrix elements represent the spatial correlation between the corresponding nodes, forming spatial correlation matrix data. For example, active power, voltage, etc., calculate the mutual information of time series data between any two nodes. Mutual information can measure the linear or nonlinear statistical dependency between two time series to form time correlation matrix data. The spatial correlation matrix and the time correlation matrix are used as two dimensions of the tensor to form a three-dimensional spatiotemporal correlation tensor data. The spatial correlation matrix and the time correlation matrix can also be linearly weighted fused to obtain a two-dimensional spatiotemporal correlation matrix data. Calculate the spatiotemporal weighted similarity of any two nodes in the multi-scale power matrix data at different time scales. For example, the similarity of the power data of two nodes can be calculated using a method based on Euclidean distance or dynamic time warping (DTW), and the similarity is weighted according to the spatiotemporal correlation tensor data to obtain the spatiotemporal weighted similarity. The calculated spatiotemporal weighted similarity is stored in a vector, and each element in the vector represents the spatiotemporal similarity between two nodes at a specific time scale, forming spatiotemporal similarity vector data.

[0150] Preferably, step S33 includes the following steps:

[0151] Step S331: extracting power line parameters from the power grid topology data through the power grid management system to obtain power line parameter data;

[0152] Step S332: constructing a distributed line model according to the power line parameter data to obtain electrical line model data;

[0153] Step S333: performing impedance frequency simulation on the electrical circuit model data to generate impedance frequency response data;

[0154] Step S334: mapping the line impedance attributes of the power grid topology data through the impedance frequency response data to generate power grid impedance topology data;

[0155] Step S335: Calculate the inter-node harmonic impedance according to the grid impedance topology data to generate line harmonic impedance data;

[0156] Step S336: Use the line harmonic impedance data to perform impedance sensitivity analysis on the power line parameter data to generate line impedance matrix data.

[0157] In an embodiment of the present invention, the power grid management system usually stores detailed power line parameter information, such as line length, conductor model, phase line structure, etc. By interacting with the power grid management system, query and obtain the various parameter information of the required power line, and store it as power line parameter data. Select a suitable distributed line model, such as a π-type equivalent circuit model, a multi-segment π-type equivalent circuit model, etc., to model each section of the power line. The model parameters can be obtained by referring to relevant power system analysis books or standards, or they can be modified according to actual conditions. For example, the resistance, inductance and capacitance parameters of the line can be calculated according to parameters such as line length and conductor model, and used as parameters of the π-type equivalent circuit model. The model parameters of all lines are integrated to form electrical line model data. A simulation model corresponding to the actual power grid topology is built, and the electrical line model data is imported into the simulation software. Set simulation parameters, such as simulation time, simulation step, frequency range, etc., and run the simulation program. During the simulation process, excitation signals of different frequencies are applied to the line model, and the voltage and current responses at both ends of the line are recorded. According to the voltage and current responses, the line impedance is calculated, and the impedance-frequency curve, i.e., the impedance frequency response data, is plotted. According to the impedance frequency response data, extract the line impedance value at a specific frequency, such as the impedance value at 50Hz or a specific harmonic frequency. Add the extracted impedance value as a line attribute to the power grid topology data. For example, the impedance value of each line at different frequencies can be stored as attribute information and added to the corresponding line in the power grid topology diagram. Use the node impedance matrix method, the Thevenin equivalent circuit method, etc. Use the power grid impedance topology data to calculate the harmonic impedance value between any two nodes and generate line harmonic impedance data. Analyze the degree of influence of line parameter changes on harmonic impedance. For example, the rate of change of harmonic impedance before and after line parameter changes can be calculated, or other sensitivity analysis methods can be used. Based on the analysis results, construct the line impedance matrix data.

[0158] Preferably, step S4 comprises the following steps:

[0159] Step S41: identifying voltage change sensitive nodes using the time-space similarity vector data through the power grid impedance topology data to generate key voltage sensitive node data;

[0160] Step S42: Calculate the active power change rate of the key voltage sensitive node data to generate active power change rate data;

[0161] Step S43: Calculating the change amplitude according to the active power change rate data to generate active power change amplitude data;

[0162] Step S44: marking the active power variation amplitude data with trend changes by using preset trend evolution label data to generate active power trend evolution label data;

[0163] Step S45: classifying the real-time active power trend mode according to the active power trend evolution label data, and evaluating the voltage impact degree to generate active power trend impact factor data;

[0164] Step S46: Analyze the node voltage change trend of key voltage sensitive node data using the active power trend influencing factor data to obtain node voltage trend data.

[0165] In an embodiment of the present invention, the electrical distance of each node is calculated according to the grid impedance topology data, such as the shortest path distance, the resistance distance, etc. The closer the electrical distance is, the higher the correlation between the voltage changes is. Then, in combination with the spatiotemporal similarity vector data, the nodes with higher spatiotemporal similarity with other nodes are screened out, and the voltage changes of these nodes are more susceptible to the influence of other nodes. Finally, according to the electrical distance and spatiotemporal similarity, the nodes are grouped using a clustering algorithm (such as a K-means algorithm, a DBSCAN algorithm, etc.), and the node group with close electrical distance and high spatiotemporal similarity is identified as a key voltage sensitive node, and the key voltage sensitive node data is generated. According to the preset time window size (such as 1 minute, 5 minutes, etc.), the active power change amount of each key voltage sensitive node in the time window is calculated, and it is divided by the time window size to obtain the active power change rate. According to the preset time window size, the maximum and minimum values ​​of the active power change rate of each key voltage sensitive node in the time window are calculated, and the maximum value is subtracted from the minimum value to obtain the active power change amplitude. For example, a sliding window method can be used to calculate the range of the active power change rate in each time window. The calculated active power change amplitude is associated with the corresponding timestamp and node information to generate active power change amplitude data. A set of trend evolution labels are predefined, such as "rapid rise", "slow rise", "steady", "slow decline", "rapid decline", etc., and corresponding thresholds are set. According to the active power change amplitude data, it is compared with the preset threshold, and according to the comparison result, a corresponding trend evolution label is assigned to each key voltage sensitive node in each time window to generate active trend evolution label data. Different active power change trend patterns are identified, for example, "rapid rise" and "slow rise" can be classified as "upward trend", and "rapid decline" and "slow decline" can be classified as "downward trend". Then, the influence of different trend patterns on voltage is analyzed, and a corresponding weight coefficient is assigned to each trend pattern to indicate its influence on voltage, and finally the active trend influence factor data is generated. Combined with the historical voltage data of key voltage sensitive nodes, the change trend of node voltage in the future period is predicted. For example, the direction and magnitude of voltage changes in the future can be predicted based on the current active power change trend pattern and its corresponding weight coefficient, or the voltage change trend in the future can be predicted by combining historical voltage data and active power trend influencing factor data using time series analysis methods. The predicted voltage change trend is associated with the corresponding timestamp and node information to generate node voltage trend data.

[0166] Preferably, step S41 includes the following steps:

[0167] Step S411: performing Z-score standardization processing on the spatiotemporal similarity vector data to generate standard spatiotemporal similarity vector data; performing similarity threshold processing on the standard spatiotemporal similarity vector data to obtain node similarity threshold data;

[0168] Step S412: performing grouping processing of similar feature nodes according to the spatiotemporal similarity vector data, and screening potential sensitive nodes according to the node similarity threshold data, to obtain potential power grid sensitive node data;

[0169] Step S413: Calculate the voltage change rate of the potential power grid sensitive node data to obtain the sensitive node voltage rate data;

[0170] Step S414: using the power grid impedance topology data to perform topology importance assessment on potential power grid sensitive node data to generate topology importance assessment data;

[0171] Step S415: performing dynamic response characteristic analysis on potential power grid sensitive node data to generate dynamic response characteristic data;

[0172] Step S416: performing a multi-criteria comprehensive evaluation process on the potential power grid sensitive node data based on the sensitive node voltage rate data, the topology importance evaluation data, and the dynamic response characteristic data to generate multi-criteria scoring data;

[0173] Step S417: Perform sensitive node verification processing on potential power grid sensitive node data through multi-criteria scoring data, and screen key voltage sensitive nodes to obtain key voltage sensitive node data.

[0174] In an embodiment of the present invention, the spatiotemporal similarity vector data is standardized using the Z-score standardization method, and each element is converted into standard normal distribution data with a mean of 0 and a standard deviation of 1 to generate standard spatiotemporal similarity vector data. According to the spatiotemporal similarity vector data, the node pairs with similarity higher than the threshold are grouped to obtain a plurality of similar feature node groups. Then, for each node group, according to the node similarity threshold data, the nodes with similarity higher than the threshold to other nodes in the group are screened out, and these nodes are considered to be potential grid sensitive nodes. According to the preset time window size (e.g., 1 minute, 5 minutes, etc.), the voltage change of each potential grid sensitive node in the time window is calculated, and it is divided by the time window size to obtain the voltage change rate. According to the grid impedance topology data, the network centrality index of each potential grid sensitive node is calculated, such as degree centrality, betweenness centrality, closeness centrality, etc. Using the power system simulation software, a simulation model corresponding to the actual grid topology is built, and a disturbance is applied to each potential grid sensitive node, such as a step change in active power or reactive power. After the disturbance occurs, the change process of electrical quantities such as node voltage and current is recorded, and its dynamic response characteristics are analyzed. The dynamic response characteristic indicators obtained by the analysis are associated with the node information to generate dynamic response characteristic data. According to the voltage rate data of sensitive nodes, the topological importance evaluation data and the dynamic response characteristic data, a multi-criteria decision-making method (such as the analytic hierarchy process, the entropy weight method, the TOPSIS method, etc.) is used to comprehensively score the potential sensitive nodes of the power grid. First, the weight of each evaluation indicator needs to be determined according to the actual situation. For example, the weight can be determined according to expert experience, historical data analysis and other methods. Then, according to the determined weight and the evaluation indicator value of each node, the comprehensive score of each node is calculated. A screening threshold is determined according to a preset threshold or based on a statistical method (such as a percentile method based on the distribution of multi-criteria scoring data). The multi-criteria scoring data is compared with the screening threshold to screen out nodes with scores higher than the threshold. These nodes are considered to be the final key voltage sensitive nodes because they show high sensitivity in terms of voltage change rate, topological importance and dynamic response characteristics.

[0175] As an example of the present invention, refer to Figure 3 As shown, Figure 1 Detailed implementation steps of step S5 in the flowchart, in this example, step S5 includes:

[0176] Step S51: evaluating the voltage stability margin of the power grid according to the new energy power prediction data and the node voltage trend data, and generating voltage stability margin evaluation data;

[0177] In an embodiment of the present invention, the new energy power forecast data and node voltage trend data are input into the power system analysis software, such as PSS / E, PSAT, etc., to construct a power system flow calculation model. Then, according to the preset fault or disturbance scenarios, such as new energy output fluctuations, load mutations, etc., the steady-state operation state of the power system under different operation scenarios is simulated. Finally, according to the flow calculation results, the voltage stability margin index of each node is calculated, such as the difference between the voltage collapse critical point and the current voltage, the voltage stability index, etc., to generate voltage stability margin evaluation data.

[0178] Step S52: dynamically adjusting the preset initial voltage control threshold value according to the voltage stability margin evaluation data to obtain a dynamic voltage control threshold value;

[0179] In an embodiment of the present invention, the triggering condition of voltage control is determined according to a preset voltage stability margin target, such as the difference between the voltage collapse critical point and the current voltage is not less than a certain safety value. Then, according to the voltage stability margin evaluation data, it is judged whether the current voltage stability margin meets the preset target. If the voltage stability margin does not meet the target, the preset initial voltage control threshold is dynamically adjusted according to the degree of deviation between the voltage stability margin and the target value. For example, the voltage control threshold can be adjusted linearly or nonlinearly according to the size of the deviation. The larger the deviation, the larger the adjustment range. The dynamic voltage control threshold finally obtained can be a personalized threshold for different nodes and different time periods, which is used for more refined voltage control. For example: using a fuzzy logic control algorithm, the control threshold is dynamically adjusted according to the evaluation result of the voltage stability margin. When the voltage stability margin evaluation result is lower than the preset threshold, the control threshold is lowered to improve the sensitivity of voltage control; when the voltage stability margin evaluation result is higher than the preset threshold, the control threshold is increased to reduce unnecessary voltage control operations.

[0180] Step S53: using the voltage stability margin assessment data to perform voltage fluctuation analysis on the node voltage trend data, and calculating the voltage deviation prediction value to generate the voltage deviation prediction value;

[0181] In the embodiment of the present invention, the maximum and minimum voltage values ​​of each node in the future period of time are calculated according to the predicted voltage change trend, and the maximum and minimum values ​​are compared with the preset voltage allowable fluctuation range to determine whether the node voltage will exceed the allowable range. If the predicted maximum or minimum voltage value exceeds the allowable range, the value of the exceeding part is calculated as the voltage deviation prediction value.

[0182] Step S54: Calculating the voltage regulation target value according to the voltage deviation prediction value to obtain voltage regulation target value data;

[0183] In an embodiment of the present invention, in combination with a preset voltage control strategy, for example, the voltage deviation is controlled within a certain specific value within the allowable range, and the voltage regulation target value of each node in the future period of time is calculated. For example, if the voltage deviation prediction value of a node in the next 15 minutes is 0.05pu, and the preset voltage control strategy is to control the voltage deviation within 0.02pu, then the voltage regulation target value of the node in the next 15 minutes is the current voltage value plus 0.03pu. The calculated voltage regulation target value is associated with the corresponding timestamp and node information to generate voltage regulation target value data.

[0184] Step S55: generating a voltage control signal for the voltage regulation target value data through a dynamic voltage control threshold, thereby generating node voltage control signal data.

[0185] In an embodiment of the present invention, the voltage regulation target value is compared with the dynamic voltage control threshold. If the voltage regulation target value exceeds the dynamic voltage control threshold, a corresponding control signal is generated, such as issuing a boost or buck instruction to control related equipment (such as transformer taps, reactive power compensation equipment, etc.) to perform an adjustment operation so that the node voltage approaches the target value. The control signal can be a continuous adjustment signal, such as a percentage of the adjustment amount, or a discrete switch signal, such as a switching instruction of the equipment. The generated control signal is associated with the corresponding timestamp, node information, and control device information to generate node voltage control signal data. For example, it can be defined that: when the "voltage deviation prediction value" exceeds the positive threshold, it indicates that a boost measure needs to be taken; when the "voltage deviation prediction value" is lower than the negative threshold, it indicates that a voltage reduction measure needs to be taken; when the "voltage deviation prediction value" is within the threshold range, it indicates that no voltage control operation is required.

[0186] Preferably, step S6 comprises the following steps:

[0187] Step S61: Obtaining distribution data of controllable resources of the power grid;

[0188] Step S62: Decomposing the controllable resource distribution data of the power grid into regional voltage control targets by using the power grid impedance topology data to generate regional voltage control resource data;

[0189] Step S63: Utilizing the regional voltage control resource data to optimize the scheduling of the node voltage control signal data, and generating a new energy coordinated control strategy to generate an automatic voltage control strategy;

[0190] Step S64: generating intelligent voltage regulation instructions according to the automatic voltage control strategy to obtain intelligent voltage regulation instruction data.

[0191] In an embodiment of the present invention, data is exchanged with a power grid management system to obtain the distribution of controllable resources in the power grid, including information such as type, location, capacity, and response characteristics. Controllable resources include traditional controllable resources, such as transformer taps, capacitor banks, static VAR generators (SVG), etc., and new energy controllable resources, such as photovoltaic inverters and wind turbines with active / reactive power regulation capabilities. The acquired data is sorted to form structured power grid controllable resource distribution data. According to the power grid impedance topology data, the power grid is divided into several voltage control areas using a clustering algorithm (such as spectral clustering, K-medoids clustering, etc.), and the node voltage changes in each area have a high correlation. When dividing the area, factors such as electrical distance, line impedance, and geographical location can be considered. Then, according to the node voltage control signal data in each area, the overall voltage control demand of the area, such as voltage regulation, reactive power compensation capacity, etc., is calculated. For each voltage control area, an optimization scheduling model, such as a linear programming model, a mixed integer programming model, etc., is constructed based on the regional voltage control resource data and the node voltage control signal data. The objective function of the model can be minimizing voltage deviation, minimizing control cost, maximizing new energy consumption, etc. The constraints of the model include output limits of controllable resources, voltage stability constraints, line transmission capacity constraints, etc. For example, for transformer taps, tap position adjustment instructions can be generated; for capacitor banks, capacitor switching instructions can be generated; for photovoltaic inverters with active / reactive power adjustment capabilities, active power and reactive power adjustment instructions can be generated. The instruction data needs to contain detailed control information, such as control device ID, adjustment amount, adjustment time, execution order, etc., to ensure that the control instructions can be correctly parsed and executed.

[0192] The beneficial effect of the present application is that obtaining the distribution data of controllable resources of the power grid provides comprehensive basic information for the operation of the power grid. Accurately grasping the location, capacity and characteristics of controllable resources enables the power grid to quickly identify the available resources when adjustment is required, ensuring the effective management of controllable resources, and maximizing the control ability of the power grid, especially in the face of fluctuations in new energy power generation, it can effectively respond to voltage changes and ensure the stable operation of the power grid. By analyzing the impedance topology data of the power grid, the controllable resource distribution data is decomposed into regional voltage control targets, so that the voltage control strategy can be more refined and accurate. The voltage control target of each region is clarified, which improves the flexibility and pertinence of power grid management, ensures that the corresponding voltage regulation measures can be implemented in each region according to the actual situation, and reduces the potential risks caused by voltage instability. This regionalized management mode not only improves the response speed of the power grid, but also enhances the adaptability of each region to changes in power demand. Through the voltage stability margin assessment and the adjustment of the dynamic voltage control threshold, the system can flexibly adjust the control strategy according to the real-time status and future trends of the power grid to ensure that the voltage level is always within a safe and stable range. The prediction of voltage deviation and the calculation of adjustment target values ​​make voltage control more proactive and accurate, reduce unnecessary adjustment actions, and improve the efficiency of power grid operation. Through the refined management of controllable resources and the coordinated control of new energy, the effective decomposition and optimized scheduling of regional voltage control targets are achieved, ensuring the efficient execution of control measures. The generation of intelligent voltage adjustment instructions can not only quickly respond to changes in power grid status, but also adapt to active power trends, significantly enhancing the grid's adaptive ability and fault recovery speed. It not only improves the grid's acceptance and utilization efficiency of distributed new energy, but also strengthens the grid's dynamic balance adjustment capability, effectively solving the voltage stability problem caused by the access of new energy.

[0193] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0194] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. An automatic voltage control method based on active power trend determination, characterized in that: The following steps are involved: Step S1: Processing the power grid topology structure according to the power grid management system to generate power grid topology structure data; Collect multi-scale monitoring data based on power grid topology data to generate multi-scale power matrix data; Step S2: extracting new energy node status from the multi-scale power matrix data to obtain new energy monitoring status data; predicting new energy output power based on the new energy monitoring status data to generate new energy power prediction data; Step S3: performing line impedance analysis according to the power grid topology data to obtain power grid impedance topology data; performing spatiotemporal node similarity processing on the multi-scale power matrix data based on the power grid topology data to generate spatiotemporal similarity vector data; Step S4: identifying voltage change sensitive nodes using the time-space similarity vector data through the grid impedance topology data to generate key voltage sensitive node data; Perform node voltage change trend analysis based on key voltage sensitive node data to obtain node voltage trend data; Step S5: evaluating the voltage stability margin of the power grid according to the new energy power prediction data and the node voltage trend data, and generating voltage stability margin evaluation data; generating a voltage control signal based on the voltage stability margin evaluation data, and generating node voltage control signal data; Step S6: generating an intelligent voltage regulation instruction according to the node voltage control signal data to obtain intelligent voltage regulation instruction data.

2. The automatic voltage control method based on active power trend determination according to claim 1, characterized in that: Step S1 includes the following steps: Step S11: performing monitoring power node analysis according to the power grid management system to obtain monitoring power node data, wherein the monitoring power node data includes new energy node data and power load node data; Step S12: Processing the power grid topology structure based on the monitored power node data to generate power grid topology structure data; Step S13: performing time scale correlation analysis according to the power grid topology data, and performing adaptive time scale selection to obtain multi-scale time window data; Step S14: performing multi-scale monitoring data collection on the monitoring power node data through the multi-scale time window data to obtain multi-scale power monitoring data; performing multi-scale vector construction according to the multi-scale power monitoring data to generate multi-scale monitoring vector data; Step S15: performing data cleaning processing on the multi-scale power monitoring data, and extracting monitoring power features to generate monitoring node power feature data; Step S16: Reconstruct the monitoring state matrix of the monitoring node power characteristic data using the multi-scale monitoring vector data to generate multi-scale power matrix data.

3. The automatic voltage control method based on active power trend determination according to claim 1, characterized in that: Step S2 includes the following steps: Step S21: extracting new energy node status according to multi-scale power matrix data to obtain new energy monitoring status data; Step S22: performing multi-scale time series feature fusion according to the new energy monitoring status data to generate time series monitoring feature fusion data; Step S23: Calculate historical power generation based on the time series monitoring feature fusion data to obtain multi-scale power generation data; Step S24: performing spatiotemporal new energy output analysis on the time series monitoring feature fusion data through multi-scale power generation data to generate new energy output related data; Step S25: performing long-term and short-term power forecasts of new energy sources according to the new energy output correlation data and the power grid topology data, and generating short-term power forecast data and long-term power forecast data respectively; Step S26: Perform multi-time scale integration processing on the short-term power prediction data and the long-term power prediction data to generate new energy power prediction data.

4. The automatic voltage control method based on active power trend determination according to claim 3 is characterized in that: Step S25 includes the following steps: Step S251: construct a multivariate regression model for the new energy output related data using a gradient descent algorithm to obtain a short-term new energy output prediction model; Step S252: training the short-term renewable energy output prediction model through multi-scale power generation data to generate an optimized short-term renewable energy prediction model; Step S253: using the optimized short-term new energy prediction model to perform short-term energy power prediction on the time series monitoring feature fusion data to generate short-term power prediction data; Step S254: acquiring weather forecast data for new energy nodes based on the power grid topology data; Step S255: performing an influencing factor analysis on the new energy output related data through weather forecast data to generate weather influencing factor data; Step S256: Using a preset long-term and short-term neural network model to perform transfer learning on the weather influencing factor data and the time series monitoring feature fusion data, and construct a long-term new energy prediction model; Step S257: Perform long-term energy power prediction on the short-term power prediction data and the weather influencing factor data through the long-term new energy prediction model to generate long-term power prediction data.

5. The automatic voltage control method based on active power trend determination according to claim 1, characterized in that: Step S3 includes the following steps: Step S31: extracting node geographic coordinates according to the power grid topology data to generate power grid node coordinate data; Step S32: performing inter-node distance matrix processing according to the grid node coordinate data to generate inter-node distance matrix data; Step S33: performing line impedance matrix processing according to the power grid topology data to generate line impedance matrix data and power grid impedance topology data; Step S34: calculating the spatial correlation coefficient of the line impedance matrix data through the node distance matrix data to generate spatial correlation matrix data; Step S35: Calculate the time series mutual information of the multi-scale power matrix data to obtain time correlation matrix data; Step S36: performing spatiotemporal correlation fusion according to the spatial correlation matrix data and the temporal correlation matrix data to generate spatiotemporal correlation tensor data; Step S37: Performing spatiotemporal weighted similarity processing on the multi-scale power matrix data through the spatiotemporal correlation tensor data to generate spatiotemporal similarity vector data.

6. The automatic voltage control method based on active power trend determination according to claim 5, characterized in that: Step S33 includes the following steps: Step S331: extracting power line parameters from the power grid topology data through the power grid management system to obtain power line parameter data; Step S332: constructing a distributed line model according to the power line parameter data to obtain electrical line model data; Step S333: performing impedance frequency simulation on the electrical circuit model data to generate impedance frequency response data; Step S334: mapping the line impedance attributes of the power grid topology data through the impedance frequency response data to generate power grid impedance topology data; Step S335: Calculate the inter-node harmonic impedance according to the grid impedance topology data to generate line harmonic impedance data; Step S336: Use the line harmonic impedance data to perform impedance sensitivity analysis on the power line parameter data to generate line impedance matrix data.

7. The automatic voltage control method based on active power trend determination according to claim 1, characterized in that: Step S4 includes the following steps: Step S41: identifying voltage change sensitive nodes using the time-space similarity vector data through the power grid impedance topology data to generate key voltage sensitive node data; Step S42: Calculate the active power change rate of the key voltage sensitive node data to generate active power change rate data; Step S43: Calculating the change amplitude according to the active power change rate data to generate active power change amplitude data; Step S44: marking the active power variation amplitude data with trend changes by using preset trend evolution label data to generate active power trend evolution label data; Step S45: classifying the real-time active power trend mode according to the active power trend evolution label data, and evaluating the voltage impact degree to generate active power trend impact factor data; Step S46: Analyze the node voltage change trend of key voltage sensitive node data using the active power trend influencing factor data to obtain node voltage trend data.

8. The automatic voltage control method based on active power trend determination according to claim 7, characterized in that: Step S41 includes the following steps: Step S411: performing Z-score standardization processing on the spatiotemporal similarity vector data to generate standard spatiotemporal similarity vector data; performing similarity threshold processing on the standard spatiotemporal similarity vector data to obtain node similarity threshold data; Step S412: performing grouping processing of similar feature nodes according to the spatiotemporal similarity vector data, and screening potential sensitive nodes according to the node similarity threshold data, to obtain potential power grid sensitive node data; Step S413: Calculate the voltage change rate of the potential power grid sensitive node data to obtain the sensitive node voltage rate data; Step S414: using the power grid impedance topology data to perform topology importance assessment on potential power grid sensitive node data to generate topology importance assessment data; Step S415: performing dynamic response characteristic analysis on potential power grid sensitive node data to generate dynamic response characteristic data; Step S416: performing a multi-criteria comprehensive evaluation process on the potential power grid sensitive node data based on the sensitive node voltage rate data, the topology importance evaluation data, and the dynamic response characteristic data to generate multi-criteria scoring data; Step S417: Perform sensitive node verification processing on potential power grid sensitive node data through multi-criteria scoring data, and screen key voltage sensitive nodes to obtain key voltage sensitive node data.

9. The automatic voltage control method based on active power trend determination according to claim 1, characterized in that: Step S5 includes the following steps: Step S51: evaluating the voltage stability margin of the power grid according to the new energy power prediction data and the node voltage trend data, and generating voltage stability margin evaluation data; Step S52: dynamically adjusting the preset initial voltage control threshold value according to the voltage stability margin evaluation data to obtain a dynamic voltage control threshold value; Step S53: using the voltage stability margin assessment data to perform voltage fluctuation analysis on the node voltage trend data, and calculating the voltage deviation prediction value to generate the voltage deviation prediction value; Step S54: Calculating the voltage regulation target value according to the voltage deviation prediction value to obtain voltage regulation target value data; Step S55: generating a voltage control signal for the voltage regulation target value data through a dynamic voltage control threshold, thereby generating node voltage control signal data.

10. The automatic voltage control method based on active power trend determination according to claim 1, characterized in that: Step S6 includes the following steps: Step S61: Obtaining distribution data of controllable resources of the power grid; Step S62: Decomposing the controllable resource distribution data of the power grid into regional voltage control targets by using the power grid impedance topology data to generate regional voltage control resource data; Step S63: Utilizing the regional voltage control resource data to optimize the scheduling of the node voltage control signal data, and generating a new energy coordinated control strategy to generate an automatic voltage control strategy; Step S64: generating intelligent voltage regulation instructions according to the automatic voltage control strategy to obtain intelligent voltage regulation instruction data.

Images