A method and system for dynamic compensation of zero current secondary voltage drop

Abstract

The present invention relates to the field of dynamic voltage compensation technology, specifically to a method and system for dynamic compensation of zero-current secondary voltage drop, comprising the following steps: collecting current intensity data and voltage data in a power grid, analyzing time series data, determining the current intensity trend, segmenting the current and voltage data using a sliding window, calculating the trend change of each window, and predicting the moment when the current tends to zero. In the present invention, by performing real-time analysis and prediction on current and voltage data, the critical moment when the current tends to zero is accurately determined, and by refining the current change rate, the optimization of the load configuration of the power grid is achieved, so that the voltage regulation is more accurate, the response speed of the power grid under the condition of rapid current change is significantly improved, and the voltage fluctuation is effectively offset. By dynamically adjusting the transformer tap position and the real-time control of the distributed power supply switch, the instant response capability of the power system to unstable voltage is improved, and the flexibility and adaptability of the power grid are enhanced.

Description

Technical Field

[0001] The present invention relates to the technical field of dynamic voltage compensation, and in particular to a method and system for dynamic compensation of zero-current secondary voltage drop. Background Art

[0002] Dynamic voltage compensation is a power system technology used to regulate and stabilize network voltage and ensure the quality and reliability of power supply. The key lies in real-time monitoring of the voltage level of the power system and dynamically adjusting the compensation mechanism to respond to changes in load or power supply conditions. Dynamic voltage compensation usually includes a voltage regulator, a control algorithm, and a feedback mechanism. The components work together to quickly offset voltage fluctuations.

[0003] Among them, the zero-current secondary voltage drop dynamic compensation method is a technology specifically used to compensate for the voltage drop caused when the current suddenly drops to zero. In the power system, sudden changes in current can lead to voltage quality problems, especially when the current is close to zero, the voltage will suddenly drop, affecting the operation of the equipment and the stable transmission of electricity. This method dynamically adjusts the compensation strategy to correct this voltage drop in real time, ensuring that the power system can operate stably even under extreme conditions. It plays a key role in maintaining the reliability of the power grid and reducing power losses, and is especially suitable for environments with high requirements for power quality.

[0004] Existing technologies rely on fixed voltage regulation strategies and feedback mechanisms. Faced with rapidly changing grid conditions, such fixed strategies are inflexible and cannot accurately respond to sudden changes in current demand. This lack of real-time optimization and adjustment capabilities is particularly prominent in facilities or industrial areas where voltage quality is critical, causing serious safety consequences. In addition, when extreme grid events occur, the system cannot provide appropriate voltage compensation due to the lack of a refined instant response mechanism, further increasing the risk of power loss and equipment damage. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a zero-current secondary voltage drop dynamic compensation method and system.

[0006] In order to achieve the above object, the present invention adopts the following technical solution: a zero current secondary voltage drop dynamic compensation method, comprising the following steps:

[0007] S1: Collect current intensity data and voltage data in the power grid, analyze time series data, determine the current intensity trend, use sliding windows to segment current and voltage data, calculate the trend change of each window, and predict the moment when the current tends to zero, generating a current zero point prediction data set;

[0008] S2: Based on the current zero point prediction data set, the current data around the zero current moment is sampled and analyzed by data sliding average and differential processing, the current change rate is refined, and a zero current rate subdivision diagram is obtained;

[0009] S3: Based on the zero current rate segmentation diagram, dynamically plan the grid load, cross-analyze the current and load data, calculate the optimal load transfer measures, optimize the grid node load, and generate the grid load optimization configuration;

[0010] S4: Based on the grid load optimization configuration, a linear programming algorithm is used to adjust the transformer tap position and control the distributed power switch in real time, adjust the grid load and voltage, and record the adjustment process to obtain a compensation control log;

[0011] S5: Based on the compensation control log, a convolutional neural network is applied to monitor the real-time data of voltage and load, and the response of voltage and load is detected through data comparison and trend analysis to generate voltage-load response information;

[0012] S6: Based on the voltage load response information, using a decision tree and a rule engine, if it is determined that the voltage stability does not meet expectations, the load is redistributed and the voltage regulation parameters are calculated, and a secondary optimization is performed to obtain a secondary load adjustment plan.

[0013] As a further solution of the present invention, the current zero point prediction data set includes a current trend line, a zero point time, and a trend confidence level; the zero current rate subdivision diagram includes a rate peak, a change frequency, and a rate stability zone; the power grid load optimization configuration includes a node optimization index, a load distribution diagram, and a load adjustment plan; the compensation control log includes operation adjustment information, a timestamp record, and operation impact analysis results; the voltage load response information includes load change impact results, voltage adjustment results, and response time evaluation results; and the secondary load adjustment plan includes an adjustment target, expected impact, and an implementation schedule.

[0014] As a further solution of the present invention, the current intensity data and voltage data in the power grid are collected, the time series data are analyzed, the current intensity trend is determined, the current and voltage data are segmented using a sliding window, the trend change of each window is calculated, and the moment of approaching zero current is predicted. The steps of generating a current zero point prediction data set are specifically as follows:

[0015] S101: collecting current intensity data and voltage data in the power grid, calculating the average value of the data every minute, smoothing short-term fluctuations, and regularly recording the data collected by the equipment every minute to generate a smoothed data set;

[0016] S102: Based on the smoothed data set, set the sliding window to a period of five minutes, calculate the average current and voltage of each window, update the window data minute by minute, and record the change trend after each update to obtain a window trend data set;

[0017] S103: Based on the window trend data set, filter the data windows tending to zero current, perform second-level time slice analysis on the windows, refine the predicted time point of zero current, and generate a current zero point prediction data set.

[0018] As a further solution of the present invention, based on the current zero point prediction data set, the current data around the zero current moment is sampled and analyzed by data sliding average and differential processing, and the current change rate is refined to obtain the zero current rate subdivision diagram. Specifically, the steps are as follows:

[0019] S201: Based on the current zero point prediction data set, select data ten minutes before and after the zero current, calculate the current change rate within the time period, and record the current change per minute to obtain an overview of the rate change;

[0020] S202: Based on the rate change overview, apply local difference technology to refine the current change, analyze the data within each minute, extract the current change details at the time point, and generate a current detail change graph;

[0021] S203: Based on the current detail change diagram, perform detail comparison of data points, determine the consistency of the current change trend through data verification at consecutive time points, and obtain a zero current rate subdivision diagram.

[0022] As a further solution of the present invention, based on the zero current rate segmentation diagram, the grid load is dynamically planned, the current and load data are cross-analyzed, the optimal load transfer measures are calculated, and the grid node load is optimized. The steps of generating the grid load optimization configuration are specifically as follows:

[0023] S301: Based on the zero current rate segmentation diagram, analyze the key time period of current drop, use the time series data comparison method to analyze the load fluctuation between the difference nodes, and evaluate the load fluctuation of the power grid nodes to obtain the power grid load fluctuation diagram;

[0024] S302: Based on the power grid load fluctuation diagram, cross-comparison is performed on the current and power grid node load data, a load transfer path between nodes is determined, and a load adjustment sequence between nodes is optimized to obtain a load transfer path diagram;

[0025] S303: Based on the load transfer path diagram, real-time data monitoring is performed to optimize the load of power nodes, adjust the energy distribution between nodes, and tune the power output of each node to generate an optimized configuration of the power grid load.

[0026] As a further solution of the present invention, based on the grid load optimization configuration, a linear programming algorithm is used to adjust the transformer tap position and control the distributed power switch in real time, adjust the grid load and voltage, and record the adjustment process. The steps of obtaining the compensation control log are specifically as follows:

[0027] S401: Based on the grid load optimization configuration, a linear programming algorithm is used to adjust the transformer tap position, monitor and control the distributed power switch in real time, perform load and voltage optimization adjustment, and obtain an adjustment feedback overview;

[0028] S402: Based on the regulation feedback overview, continuously optimize the control of the transformer tap position and the power switch, refine the voltage and load control, optimize the grid operation efficiency, and obtain grid regulation information;

[0029] S403: Based on the power grid control information, record the time point, tap position and switch status of each adjustment, and summarize the information to obtain a compensation control log.

[0030] As a further solution of the present invention, the linear programming algorithm is according to the formula:

[0031] ;

[0032] Calculates the optimal tap position and power switch state and adjusts the feedback overview value, where is the cost vector, is the decision variable vector, including the transformer tap position and the distributed generation switch state, For the The power output at a point in time, is the power output of the power supply at the previous time point, For the The voltage value at a time point, is the target voltage value, and are the weight coefficients of power supply stability and voltage stability respectively.

[0033] As a further solution of the present invention, based on the compensation control log, a convolutional neural network is applied to monitor the real-time data of voltage and load, and the response of voltage and load is detected through data comparison and trend analysis. The steps of generating voltage load response information are specifically as follows:

[0034] S501: Based on the compensation control log, a convolutional neural network is applied to monitor voltage and load data, analyze instant changes in voltage and load, mark key time points of fluctuations, and mark abnormal fluctuations between voltage and load, and generate voltage and load dynamic data;

[0035] S502: Based on the voltage-load dynamic data, analyze the interaction between voltage and load, extract and record the characteristics of each fluctuation event, identify key characteristics of voltage and load changes, and obtain a grid response feature set;

[0036] S503: Based on the grid response feature set, evaluate and analyze the voltage and load behavior of the grid, organize key data points and response features, and generate voltage load response information.

[0037] As a further solution of the present invention, based on the voltage load response information, using a decision tree and a rule engine, if it is determined that the voltage stability does not meet expectations, the load distribution is re-performed and the voltage regulation parameters are calculated, and secondary optimization is performed to obtain the steps of a secondary load adjustment plan:

[0038] S601: Based on the voltage load response information, using a decision tree and a rule engine, quantitatively analyzing voltage stability, detecting deviations from the standard, identifying power nodes that do not achieve expected performance, and generating voltage stability analysis results;

[0039] S602: Based on the voltage stability analysis result, recalculate the load distribution in the power grid, adjust the voltage-related parameters, optimize the voltage, meet the voltage stability requirements, and form an overview of the power grid load configuration;

[0040] S603: Based on the grid load configuration overview, implement secondary optimization of the grid, adjust the load and voltage of key nodes, determine that the load of the power node reaches the optimal state, and obtain a secondary load adjustment plan.

[0041] A zero-current secondary voltage drop dynamic compensation system, the zero-current secondary voltage drop dynamic compensation system is used to execute the above-mentioned zero-current secondary voltage drop dynamic compensation method, the system comprises:

[0042] The data prediction module collects the current intensity data and voltage data in the power grid, applies the sliding window technology to segment the data, calculates the data change trend within the window, and predicts the zero current moment to obtain the zero current prediction set;

[0043] The zero current refinement module performs sliding average and differential analysis on the data around the zero current moment based on the zero current prediction set, refines the current change rate, and obtains a current rate subdivision diagram;

[0044] The load optimization configuration module dynamically plans the grid load based on the current rate segmentation diagram, cross-analyzes the current data and the load data, and optimizes the grid node load to obtain the load optimization configuration;

[0045] The real-time control execution module uses a linear programming algorithm to adjust the transformer tap position and control the distributed power switch in real time according to the load optimization configuration, adjusts the grid load and voltage, and obtains a compensation control log;

[0046] The stability analysis module applies a convolutional neural network based on the compensation control log, monitors the real-time data of voltage and load, performs data comparison and trend analysis, detects voltage and load responses, and combines a decision tree and a rule engine. If the voltage stability does not meet expectations, the load is reallocated and the voltage parameters are adjusted to obtain a secondary load adjustment plan.

[0047] Compared with the prior art, the advantages and positive effects of the present invention are:

[0048] In the present invention, by real-time analysis and prediction of current and voltage data, the critical moment when the current tends to zero is accurately determined, and by refining the current change rate, the optimization of the power grid load configuration is achieved, so that the voltage regulation is more accurate, and the response speed of the power grid under the condition of rapid current changes is significantly improved, and the voltage fluctuation is effectively offset. By dynamically adjusting the transformer tap position and real-time control of distributed power switches, the power system's immediate response capability to unstable voltage is improved, the flexibility and adaptability of the power grid are enhanced, and the strengthening of monitoring and trend analysis makes the power system operation safer and improves the compensation efficiency and reliability of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 It is a schematic diagram of the workflow of the present invention;

[0050] Figure 2 This is a detailed flow chart of S1 of the present invention;

[0051] Figure 3 This is a detailed flow chart of S2 of the present invention;

[0052] Figure 4 This is a detailed flow chart of S3 of the present invention;

[0053] Figure 5 This is a detailed flow chart of S4 of the present invention;

[0054] Figure 6 This is a detailed flow chart of S5 of the present invention;

[0055] Figure 7 This is a detailed flow chart of S6 of the present invention;

[0056] Figure 8 It is a system schematic diagram of the present invention. DETAILED DESCRIPTION

[0057] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0058] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, "multiple" means two or more, unless otherwise clearly and specifically defined.

[0059] Example

[0060] See also Figure 1 The present invention provides a technical solution: a method for dynamic compensation of zero current secondary voltage drop, comprising the following steps:

[0061] S1: Collect current intensity data and voltage data in the power grid, analyze time series data, determine the current intensity trend, use sliding windows to segment current and voltage data, calculate the trend change of each window, and predict the moment when the current tends to zero, generating a current zero point prediction data set;

[0062] S2: Based on the current zero point prediction data set, the current data around the zero current moment is sampled and analyzed through data sliding average and differential processing, the current change rate is refined, and the zero current rate subdivision diagram is obtained;

[0063] S3: Based on the zero current rate segmentation diagram, dynamically plan the grid load, cross-analyze the current and load data, calculate the optimal load transfer measures, optimize the grid node load, and generate the grid load optimization configuration;

[0064] S4: Based on the optimal configuration of the grid load, a linear programming algorithm is used to adjust the transformer tap position and control the distributed power switch in real time, adjust the grid load and voltage, and record the adjustment process to obtain the compensation control log;

[0065] S5: Based on the compensation control log, a convolutional neural network is applied to monitor the real-time data of voltage and load, and the response of voltage and load is detected through data comparison and trend analysis to generate voltage-load response information;

[0066] S6: Based on the voltage load response information, using the decision tree and rule engine, if it is determined that the voltage stability does not meet expectations, the load is redistributed and the voltage regulation parameters are calculated, and a secondary optimization is performed to obtain a secondary load adjustment plan.

[0067] The current zero point prediction data set includes the current trend line, zero point time, and trend confidence. The zero current rate segmentation diagram includes rate peak, change frequency, and rate stability area. The power grid load optimization configuration includes node optimization indicators, load distribution diagram, and load adjustment plan. The compensation control log includes operation adjustment information, timestamp records, and operation impact analysis results. The voltage load response information includes load change impact results, voltage adjustment results, and response time evaluation results. The secondary load adjustment plan includes adjustment targets, expected impacts, and implementation schedule.

[0068] See also Figure 2 , collect current intensity data and voltage data in the power grid, analyze time series data, determine the current intensity trend, use sliding windows to segment current and voltage data, calculate the trend change of each window, and predict the moment when the current tends to zero. The specific steps of generating the current zero point prediction data set are:

[0069] S101: collecting current intensity data and voltage data in the power grid, calculating the average value of the data every minute, smoothing short-term fluctuations, and regularly recording the data collected by the equipment every minute to generate a smoothed data set;

[0070] To collect current intensity data and voltage data in the power grid, first collect the real-time current and voltage data of the power grid through sensor equipment, perform preliminary screening on the data, and remove those abnormal values ​​that deviate significantly from the normal range, such as extremely low or extremely high current intensity readings. Then, apply data smoothing technology to reduce the impact of short-term fluctuations by calculating the average value of data within three consecutive minutes, thereby generating smoothed data every minute. This processing not only improves the accuracy of the data, but also provides a more stable foundation for subsequent analysis.

[0071] S102: Based on the smoothed data set, set the sliding window to a period of five minutes, calculate the average current and voltage of each window, update the window data minute by minute, and record the change trend after each update to obtain a window trend data set;

[0072] Based on the smoothed data set, a regularly updated sliding window is set up, which is refreshed every five minutes. The window covers the data within the last five minutes. Each time the window is updated, the average current and voltage of all data points in the window are calculated, and how the average value changes over time is monitored. This can dynamically track the changing trends of current and voltage, and capture any abnormalities or important changes in a timely manner. By comparing the data changes in several consecutive windows, it can be preliminarily determined whether the current has a trend towards zero.

[0073] S103: based on the window trend data set, filter the data window tending to zero current, perform second-level time slice analysis on the window, refine the time point of zero current prediction, and generate a current zero point prediction data set;

[0074] Based on the window trend data set, further refined analysis is performed to select those time windows where the current trend tends to zero, and slice analysis is performed on the windows at the second level, with one data slice every 15 seconds. In a shorter time period, the average current value is refined and calculated, thereby more accurately predicting the specific moment when the current reaches zero. This helps to accurately grasp the changing trend of the current, which is critical to ensuring the stable operation and safe monitoring of the power grid.

[0075] See also Figure 3 Based on the current zero point prediction data set, the current data around the zero current moment is sampled and analyzed through data sliding average and differential processing, and the current change rate is refined to obtain the zero current rate subdivision map. The specific steps are as follows:

[0076] S201: Based on the current zero point prediction data set, select the data of ten minutes before and after the zero current, calculate the current change rate in the time period, and record the current change per minute to obtain an overview of the rate change;

[0077] Based on the current zero point prediction data set, the data of ten minutes before and after the zero current moment are selected as the analysis object. This time range is selected because it can provide sufficient data to support accurate estimation of the current change rate. First, the change rate of the current per minute in this time period is calculated and determined by the difference method (that is, the difference between the current value at the current moment and the current value at the previous moment). The continuous rate calculation results will be used to monitor how the current rate changes over time, thereby generating an overview of the rate change. The chart not only reflects how the current gradually approaches zero, but also reveals abnormal fluctuations.

[0078] S202: Based on the rate change overview, the local difference technology is applied to refine the current change, the data within each minute is analyzed, the current change details at the time point are extracted, and a current detail change graph is generated;

[0079] Based on the overview of rate changes, local difference technology is used to perform a more detailed analysis of current changes. The data within each minute is analyzed again, and a more detailed current change image is depicted by calculating the current difference per second. The application of local difference can capture subtle changes in current fluctuations in a short period of time, which are ignored in larger-scale analysis. Through this detailed level of analysis, a current detail change map can be generated, which shows the details of current changes every minute or even every second within the observation window.

[0080] S203: Based on the current detail change diagram, perform detail comparison of data points, determine the consistency of the current change trend through data verification at consecutive time points, and obtain a zero current rate subdivision diagram;

[0081] Based on the current detail change diagram, a detailed comparison between data points is performed. By comparing the current data at consecutive time points, the consistency and accuracy of the current change trend are verified. Through the comparison process, attention is paid to the consecutive data points that show a tendency towards zero current to ensure that the changes between them are coherent and consistent, which helps to confirm the accuracy of the final current zero point prediction and generate a zero current rate breakdown diagram, which shows the exact rate and pattern of the current tending to zero.

[0082] See also Figure 4 ,Based on the zero current rate segmentation graph, the grid load is dynamically planned, the current and load data are cross-analyzed, the optimal load transfer measures are calculated, and the grid node load is optimized. The specific steps for generating the grid load optimization configuration are as follows:

[0083] S301: Based on the zero current rate segmentation diagram, analyze the key time period of current drop, use the time series data comparison method to analyze the load fluctuation between the difference nodes, and evaluate the load fluctuation of the power grid nodes to obtain the power grid load fluctuation diagram;

[0084] Based on the zero current rate segmentation diagram, the critical time period of current drop is identified. This stage indicates that there is abnormal load or uneven current distribution in the power grid. The time series data comparison method is used to analyze the load data of power grid nodes at different time points to identify nodes with large load fluctuations. This analysis helps to understand the load response of each node during the period of rapid current change. In this way, a power grid load fluctuation diagram can be drawn, which shows the load changes of each power grid node at critical moments.

[0085] S302: Based on the grid load fluctuation diagram, cross-comparison is performed on the current and grid node load data, a load transfer path between nodes is determined, and a load adjustment sequence between nodes is optimized to obtain a load transfer path diagram;

[0086] Based on the grid load fluctuation diagram, the current data and the load data of the grid nodes are cross-compared and analyzed. Through this analysis, the load transfer path between nodes can be accurately determined, and the load adjustment sequence between nodes can be optimized according to the actual load conditions, including identifying nodes with higher loads and nodes that need enhanced support, and arranging the load transfer between them to ensure the stable operation and efficiency of the grid. Through operation, a load transfer path diagram can be formulated, which explains the optimal path and sequence of load adjustment.

[0087] S303: Based on the load transfer path diagram, real-time data monitoring is performed to optimize the load of the power nodes, adjust the energy distribution between the nodes, and tune the power output of each node to generate an optimized configuration of the power grid load;

[0088] Based on the load transfer path diagram, the entire power grid is monitored and optimized in real time. According to the guidance of the load transfer path diagram, the energy distribution and power output between nodes are dynamically adjusted. This process involves detailed tuning of the load of each power node to ensure that each node can operate at optimal efficiency while reducing energy consumption and losses caused by uneven load. Through this fine adjustment, an optimized grid load configuration is generated, which not only improves the overall performance of the grid, but also improves its adaptability to future current fluctuations.

[0089] See also Figure 5 , based on the optimal configuration of the power grid load, a linear programming algorithm is used to adjust the transformer tap position and control the distributed power switch in real time, adjust the power grid load and voltage, and record the adjustment process. The specific steps to obtain the compensation control log are:

[0090] S401: Based on the optimal configuration of the grid load, a linear programming algorithm is used to adjust the transformer tap position, monitor and control the distributed power switches in real time, perform optimal adjustment of the load and voltage, and obtain an overview of the regulation feedback;

[0091] Based on the optimal configuration of the grid load, the application of linear programming technology to optimize the tap position of the transformer is a key means to adjust the voltage level of the grid. At the same time, real-time monitoring of the switching status of each distributed power source is carried out to ensure the coordination and consistency of the load and voltage. Synchronous control helps maintain the load balance and voltage stability of the entire grid, thereby reducing energy consumption and improving efficiency. In addition, a regulation feedback overview is generated after each adjustment. The overview records the results and system response of each control activity, providing instant performance feedback.

[0092] Linear programming algorithm, according to the formula:

[0093] ;

[0094] Calculates the optimal tap position and power switch state and adjusts the feedback overview value, where is the cost vector, such as the transformer tap adjustment and the power switch cost, is the decision variable vector, including the transformer tap position and the distributed generation switch state, For the The power output at a point in time, is the power output of the power supply at the previous time point, For the The voltage value at a time point, is the target voltage value, and are the weight coefficients of power supply stability and voltage stability, Refers to finding an optimal set of decision variables while satisfying all operations and constraints. , making the total cost of the entire power grid Minimum.

[0095] Execution process:

[0096] The original linear programming part is calculated to determine the basic transformer tap position and power switch state. At the same time, a penalty term is introduced to consider the smoothness of the power output and the stability of the voltage, where and The value of is determined through historical data analysis to balance cost and system stability. The power and voltage of the power supply at each time point are penalized to ensure that the output and voltage are close to the previous time point or the target voltage. The optimal solution including power supply and voltage adjustment is obtained by solving the enhanced linear programming problem, and the total operating cost is calculated. , thus obtaining an overview of the regulation feedback of the entire system.

[0097] S402: Based on the regulation feedback overview, continuously optimize the transformer tap position and the control of the power switch, refine the voltage and load control, optimize the grid operation efficiency, and obtain the grid regulation information;

[0098] Based on the regulation feedback overview, continue to optimize the control of transformer tap positions and distributed power switches, and further fine-tune the settings of transformer taps and power switches to cope with subtle changes in grid load and voltage. Adjustments are based on real-time data and previous regulation results to ensure that the grid can operate at the highest efficiency at all times. The implementation of optimization measures not only improves the response speed of the grid, but also increases its adaptability to future changes. Through this process, grid regulation information is generated to record all key adjustments and their impact on grid performance.

[0099] S403: Based on the grid control information, record the time point, tap position and switch status of each adjustment, and summarize the information to obtain a compensation control log;

[0100] Based on the grid control information, the specific information of all adjustment operations is summarized and recorded, including each adjustment of the transformer tap position, the state change of the distributed power switch, and the specific time of the operation. This record forms a compensation control log, which not only provides data support for daily operations, but also provides important information for future system evaluation and improvement, and can provide transparency and traceability of operations.

[0101] See also Figure 6 ,Based on the compensation control log, the convolutional neural network is applied to monitor the real-time data of voltage and load.,Through data comparison and trend analysis, the response of voltage and load is detected, and the steps of generating voltage load response information are as follows:

[0102] S501: Based on the compensation control log, a convolutional neural network is applied to monitor voltage and load data, analyze the instant changes of voltage and load, mark the key time points of fluctuations, and mark abnormal fluctuations between voltage and load to generate voltage and load dynamic data;

[0103] Based on the compensation control log, convolutional neural network (CNN) is used to monitor the voltage and load data in the power grid. CNN is particularly suitable for processing spatial and temporal pattern recognition in time series data, so it can efficiently analyze the instant changes in voltage and load. By training the network to identify voltage and load fluctuations within the normal operating range, the network can automatically mark those critical time points that exceed expectations. In addition, the network can also mark abnormal fluctuations between voltage and load, which is very critical for early fault detection and prevention, and generate data containing voltage and load dynamics.

[0104] S502: Based on the voltage-load dynamic data, analyze the interaction between voltage and load, extract and record the characteristics of each fluctuation event, identify the key characteristics of voltage and load changes, and obtain a grid response feature set;

[0105] Based on the voltage and load dynamic data, the interaction between voltage and load is deeply analyzed, and the convolutional neural network is used to further extract the characteristics of each marked fluctuation event, including the amplitude, duration and impact range of the fluctuation. Through data analysis, the key voltage and load change characteristics that affect the stability of the power grid are identified. This process not only helps to understand the performance of the power grid under different load conditions, but also promotes the prediction of potential problems in the future and generates a set of power grid response characteristics.

[0106] S503: Based on the grid response feature set, evaluate and analyze the voltage and load behavior of the grid, organize key data points and response features, and generate voltage and load response information;

[0107] Based on the grid response feature set, a comprehensive assessment of the voltage and load behavior of the grid is conducted. By analyzing the data in the response feature set, key data points and behavioral characteristics are sorted out to reflect the performance and response of the grid under different conditions, which helps to optimize the grid management strategy and ensure that the grid can remain efficient and stable under various operating conditions. Finally, the analysis results will be summarized and generate voltage load response information.

[0108] See also Figure 7Based on the voltage load response information, using the decision tree and rule engine, if the voltage stability is judged to be not as expected, the load distribution is redistributed and the voltage regulation parameters are calculated, and the secondary optimization is performed. The specific steps of obtaining the secondary load adjustment plan are as follows:

[0109] S601: Based on the voltage load response information, using a decision tree and a rule engine, quantitatively analyze the voltage stability, detect deviations from the standard, identify power nodes that do not meet expected performance, and generate voltage stability analysis results;

[0110] Based on the voltage load response information, the decision tree and rule engine are used to quantitatively analyze the voltage stability of the power grid. The voltage load response information obtained from the convolutional neural network is used to detect the deviation between each power node and the predetermined voltage stability standard through the decision tree model. The rule engine will apply a series of preset logical rules to evaluate the voltage data and identify those power nodes that do not meet the expected performance, so as to accurately generate voltage stability analysis results.

[0111] S602: Based on the voltage stability analysis result, recalculate the load distribution in the power grid, adjust the voltage-related parameters, optimize the voltage, meet the voltage stability requirements, and form an overview of the power grid load configuration;

[0112] Based on the results of voltage stability analysis, it involves recalculating the load distribution in the power grid and adjusting the parameters related to voltage stability. During the process, special attention is paid to those power nodes that exhibit unstable performance according to the analysis results. The voltage performance of the entire power grid is optimized by adjusting the voltage settings and load distribution, including fine-tuning the parameters of the voltage regulation equipment to ensure that each node can operate at the optimal voltage level, thereby improving the stability and efficiency of the entire system. After adjustment, a new overview of the power grid load configuration is formed.

[0113] S603: Based on the overview of the power grid load configuration, implement secondary optimization of the power grid, adjust the load and voltage of key nodes, determine that the load of the power node reaches the optimal state, and obtain a secondary load adjustment plan;

[0114] Based on the overview of the grid load configuration, the secondary optimization of the grid is implemented. According to the latest configuration overview, the load and voltage of key nodes are adjusted in detail to ensure that the power load of each node reaches the optimal state. The optimization at this stage aims to improve the overall performance of the grid and ensure voltage stability and load balance. Through this systematic secondary optimization, it can be ensured that the grid operates in the most efficient state, and ultimately a comprehensive secondary load adjustment plan is obtained.

[0115] See also Figure 8 , a zero-current secondary voltage drop dynamic compensation system, comprising:

[0116] The data prediction module collects the current intensity data and voltage data in the power grid, applies the sliding window technology to segment the data, calculates the data change trend within the window, and predicts the zero current moment to obtain the zero current prediction set;

[0117] The zero current refinement module performs sliding average and differential analysis on the data around the zero current moment based on the zero current prediction set, refines the current change rate, and obtains a current rate subdivision diagram;

[0118] The load optimization configuration module dynamically plans the grid load based on the current rate segmentation diagram, cross-analyzes the current data and load data, and optimizes the grid node load to obtain the load optimization configuration;

[0119] The real-time control execution module uses a linear programming algorithm to adjust the transformer tap position and control the distributed power switch in real time according to the load optimization configuration, adjust the grid load and voltage, and obtain the compensation control log;

[0120] The stability analysis module applies convolutional neural networks based on the compensation control log, monitors the real-time data of voltage and load, performs data comparison and trend analysis, detects voltage and load responses, and combines decision trees and rule engines. If the voltage stability does not meet expectations, the load is reallocated and the voltage parameters are adjusted to obtain a secondary load adjustment plan.

[0121] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A method for dynamic compensation of zero current secondary voltage drop, characterized in that: The following steps are involved: Collect current intensity data and voltage data in the power grid, analyze time series data, determine the current intensity trend, use sliding windows to segment current and voltage data, calculate the trend change of each window, and predict the moment when the current tends to zero, and generate a current zero point prediction data set; Based on the current zero point prediction data set, the current data around the zero current moment is sampled and analyzed through data sliding average and differential processing, the current change rate is refined, and a zero current rate subdivision diagram is obtained; The steps of obtaining the zero current rate subdivision diagram are specifically as follows: Based on the current zero point prediction data set, data of ten minutes before and after the zero current is selected, the current change rate within the time period is calculated, and the current change per minute is recorded to obtain an overview of the rate change; Based on the rate change overview, a local difference technique is applied to refine the current change, the data within each minute is analyzed, the current change details at the time point are extracted, and a current detail change graph is generated; Based on the current detail change diagram, detailed comparison of data points is performed, and consistency of current change trend is determined by data verification at consecutive time points to obtain a zero current rate subdivision diagram; Based on the zero current rate segmentation diagram, dynamically plan the grid load, cross-analyze the current and load data, calculate the optimal load transfer measures, optimize the grid node load, and generate the grid load optimization configuration; The steps of generating the optimal configuration of the power grid load are specifically as follows: Based on the zero current rate segmentation diagram, the key time period of current drop is analyzed, the load fluctuation between different nodes is analyzed by using the time series data comparison method, and the load fluctuation of the power grid nodes is evaluated to obtain the power grid load fluctuation diagram; Based on the power grid load fluctuation diagram, the current and power grid node load data are cross-compared to determine the load transfer path between nodes, and the load adjustment sequence between nodes is optimized to obtain a load transfer path diagram; Based on the load transfer path diagram, real-time data monitoring is performed to optimize the load of power nodes, adjust the energy distribution between nodes, and tune the power output of each node to generate an optimized configuration of the power grid load; Based on the grid load optimization configuration, a linear programming algorithm is used to adjust the transformer tap position and control the distributed power switch in real time, adjust the grid load and voltage, and record the adjustment process to obtain a compensation control log; Based on the compensation control log, a convolutional neural network is applied to monitor the real-time data of voltage and load, and the response of voltage and load is detected through data comparison and trend analysis to generate voltage-load response information; Based on the voltage load response information, using a decision tree and a rule engine, if it is determined that the voltage stability does not meet expectations, the load distribution is redistributed and the voltage regulation parameters are calculated, and a secondary optimization is performed to obtain a secondary load adjustment plan.

2. The zero current secondary voltage drop dynamic compensation method according to claim 1, characterized in that: The current zero point prediction data set includes a current trend line, a zero point time, and a trend confidence level; the zero current rate segmentation diagram includes a rate peak, a change frequency, and a rate stability zone; the power grid load optimization configuration includes a node optimization index, a load distribution diagram, and a load adjustment plan; the compensation control log includes operation adjustment information, a timestamp record, and operation impact analysis results; the voltage load response information includes load change impact results, voltage adjustment results, and response time evaluation results; and the secondary load adjustment plan includes an adjustment target, an expected impact, and an implementation schedule.

3. The zero current secondary voltage drop dynamic compensation method according to claim 1, characterized in that: Collect the current intensity data and voltage data in the power grid, analyze the time series data, determine the current intensity trend, use a sliding window to segment the current and voltage data, calculate the trend change of each window, and predict the moment when the current tends to zero. The specific steps for generating the current zero point prediction data set are as follows: Collect current intensity data and voltage data in the power grid, calculate the average value of the data every minute, smooth out short-term fluctuations, and regularly record the data collected by the equipment every minute to generate a smoothed data set; Based on the smoothed data set, a sliding window is set to a period of five minutes, the average current and voltage of each window are calculated, the window data is updated minute by minute, and the change trend after each update is recorded to obtain a window trend data set; Based on the window trend data set, the data windows tending to zero current are screened, the windows are subjected to second-level time slice analysis, the time point of zero current is refined and predicted, and a current zero point prediction data set is generated.

4. The zero current secondary voltage drop dynamic compensation method according to claim 1, characterized in that: Based on the grid load optimization configuration, a linear programming algorithm is used to adjust the transformer tap position and control the distributed power switch in real time, adjust the grid load and voltage, and record the adjustment process. The specific steps of obtaining the compensation control log are as follows: Based on the optimized configuration of the power grid load, a linear programming algorithm is used to adjust the transformer tap position, monitor and control the distributed power switch in real time, perform load and voltage optimization adjustment, and obtain an overview of the regulation feedback; Based on the regulation feedback overview, continuously optimize the transformer tap position and the control of the power switch, refine the voltage and load control, optimize the grid operation efficiency, and obtain the grid regulation information; Based on the power grid regulation information, the time point, tap position and switch status of each adjustment are recorded, and the information is summarized to obtain a compensation control log.

5. The method for dynamic compensation of zero current secondary voltage drop according to claim 4, characterized in that: The linear programming algorithm is based on the formula: ； Calculates the optimal tap position and power switch state and adjusts the feedback overview value, where is the cost vector, is the decision variable vector, including the transformer tap position and the distributed generation switch state, For the The power output at a point in time, is the power output of the power supply at the previous time point, For the The voltage value at a time point, is the target voltage value, and are the weight coefficients of power supply stability and voltage stability respectively.

6. The zero current secondary voltage drop dynamic compensation method according to claim 1, characterized in that: Based on the compensation control log, a convolutional neural network is applied to monitor the real-time data of voltage and load, and the response of voltage and load is detected through data comparison and trend analysis. The steps of generating voltage and load response information are specifically as follows: Based on the compensation control log, a convolutional neural network is applied to monitor voltage and load data, analyze instant changes in voltage and load, mark key time points of fluctuations, and mark abnormal fluctuations between voltage and load to generate voltage and load dynamic data; Based on the voltage-load dynamic data, the interaction between voltage and load is analyzed, the characteristics of each fluctuation event are extracted and recorded, the key characteristics of voltage and load changes are identified, and a grid response feature set is obtained; Based on the grid response feature set, the voltage and load behavior of the grid are evaluated and analyzed, key data points and response features are sorted out, and voltage load response information is generated.

7. The zero current secondary voltage drop dynamic compensation method according to claim 1, characterized in that: Based on the voltage load response information, using the decision tree and rule engine, if it is determined that the voltage stability does not meet expectations, the load distribution is re-performed and the voltage regulation parameters are calculated, and secondary optimization is performed to obtain the steps of the secondary load adjustment plan: Based on the voltage load response information, using a decision tree and a rule engine, quantitatively analyzing voltage stability, detecting deviations from standards, identifying power nodes that do not meet expected performance, and generating voltage stability analysis results; Based on the voltage stability analysis results, recalculate the load distribution in the power grid, adjust the voltage-related parameters, optimize the voltage, meet the voltage stability requirements, and form an overview of the power grid load configuration; Based on the grid load configuration overview, secondary optimization of the grid is implemented, load and voltage are adjusted for key nodes, the load of the power nodes is determined to reach an optimal state, and a secondary load adjustment plan is obtained.

8. A zero current secondary voltage drop dynamic compensation system, characterized in that: According to the method for dynamic compensation of zero current secondary voltage drop according to any one of claims 1 to 7, the system comprises: The data prediction module collects the current intensity data and voltage data in the power grid, applies the sliding window technology to segment the data, calculates the data change trend within the window, and predicts the zero current moment to obtain the zero current prediction set; The zero current refinement module performs sliding average and differential analysis on the data around the zero current moment based on the zero current prediction set, refines the current change rate, and obtains a current rate subdivision diagram; The load optimization configuration module dynamically plans the grid load based on the current rate segmentation diagram, cross-analyzes the current data and the load data, and optimizes the grid node load to obtain the load optimization configuration; The real-time control execution module uses a linear programming algorithm to adjust the transformer tap position and control the distributed power switch in real time according to the load optimization configuration, adjusts the grid load and voltage, and obtains a compensation control log; The stability analysis module applies a convolutional neural network based on the compensation control log, monitors the real-time data of voltage and load, performs data comparison and trend analysis, detects voltage and load responses, and combines a decision tree and a rule engine. If the voltage stability does not meet expectations, the load is reallocated and the voltage parameters are adjusted to obtain a secondary load adjustment plan.

Images