An intelligent dynamic monitoring system and method for power grid dispatching system

Abstract

The present invention discloses an intelligent dynamic monitoring system and method for a power grid dispatching system, wherein a filtering unit is provided to first filter the bus voltages in each line in the power grid obtained in real time, filter out the bus voltages whose absolute values ​​exceed the first threshold value and exceed the preset time length, and then make further voltage over-limit judgments on this basis, that is, adopt two-level thresholds to jointly realize the bus voltage over-limit judgment, while reducing the number of bus voltage over-limit alarm data and ensuring the stable operation of the system. In addition, a prediction model is provided to realize the prediction control of the power grid dispatching system, wherein the prediction model predicts and calculates the absolute value of the bus voltage after the set time length and the voltage change rate within the set time length for comprehensive judgment, and performs corresponding processing according to the prediction judgment results, thereby realizing the prediction dispatching of the main grid, making the main grid dispatching more accurate and reasonable, and improving the power grid operation efficiency.

Description

Technical Field

[0001] The present invention relates to an intelligent dynamic monitoring system and method for a power grid dispatching system, and in particular to an intelligent dynamic monitoring system and method for judging over-limit of a main grid bus voltage. Background Art

[0002] In the main grid dispatching process, a single threshold is usually used to determine whether there is an over-limit alarm for the bus voltage. If the threshold is set too small, the amount of bus voltage over-limit alarm information data will be large, making it difficult to find truly useful alarm data. If the threshold is set too large, important alarm information will be ignored, resulting in greater grid volatility and reduced operational stability of the power system. Therefore, how to reasonably set the threshold to improve the monitoring efficiency of the grid dispatching system and ensure safe and reliable operation of the grid is a current problem. For example, the patent application with announcement number 204008825U adopts an approach that adds an overvoltage monitoring module and an alarm module to the existing overvoltage protector, monitors the overvoltage in real time through the overvoltage monitoring module, and when the overvoltage protection action value is reached, the alarm module sends an alarm and prompts the staff to realize voltage monitoring. This judgment method is too simple and can easily cause system misoperation.

[0003] In addition, even if the corresponding threshold is set according to the current scale of grid load, as the system runs and the grid load changes, the set threshold will gradually become unsuitable for the current grid dispatching and monitoring system, which can easily cause system misjudgment and low accuracy. In the process of over-limit judgment, the judgment is only based on the collected real-time value of the bus voltage. When the grid fluctuation can be quickly restored, it will cause system misjudgment and fail to realize intelligent prediction and alarm.

[0004] Therefore, in order to solve the above problems, it is necessary to provide a new intelligent dynamic monitoring system and method for power grid dispatching system. Summary of the invention

[0005] The purpose of the present invention is to overcome the shortcomings of the prior art and propose an intelligent dynamic monitoring system and method for a power grid dispatching system, which adopts two-level thresholds to jointly realize bus voltage over-limit judgment, and also sets up a prediction model to realize predictive control of the power grid dispatching system, thereby reducing the number of bus voltage over-limit alarm data while ensuring stable operation of the system.

[0006] An intelligent dynamic monitoring method for a power grid dispatching system of the present invention adopts the following steps:

[0007] S1: Obtain and store the load and bus voltage of each line in the power grid in real time;

[0008] S2: Filter out bus voltages whose absolute values ​​exceed the first threshold and exceed a preset time length, and extract load changes in the bus line whose voltage exceeds the limit in a preset time period before the voltage exceeds the limit;

[0009] S3: query the voltage change trend similar to the load change from the historical database and input it into the prediction model. The prediction model predicts and calculates the absolute value of the bus voltage after a set time and the voltage change rate within the set time;

[0010] S4: Determine whether the bus voltage exceeds the limit based on the predicted absolute value of the bus voltage after the set time and the voltage change rate within the set time, and perform corresponding processing based on the judgment result.

[0011] Preferably, in step S1, the load and bus voltage in each line in the power grid are acquired and stored in real time, specifically including: setting sensors at corresponding positions in each line to collect the load and bus voltage in each line in real time and store them in a historical database, extracting the stored load and bus voltage of each line from the historical database to draw the load change curve and bus voltage change curve of each line, and correspondingly associating the load change curve and bus voltage change curve of each line.

[0012] Preferably, in step S2, the load change in the bus line where the voltage exceeds the limit in a preset time period before the voltage exceeds the limit is extracted, specifically including: when the filtering unit determines that the absolute value of the bus voltage in the current line exceeds the first threshold and exceeds the preset time length, extracting the load change curve of the bus line in the preset time period before the voltage exceeds the limit from the historical database.

[0013] Preferably, the step S3 specifically includes: extracting the load change curve of the preset time period before the voltage in the bus line exceeds the limit from the historical database, and then querying and drawing the stored load change curves of each line from the historical database for comparison; when the coincidence degree between the two is higher than the fourth threshold value, it is determined that the two are similar, and the bus voltage change curve at this time is input into the prediction model according to the corresponding correlation relationship between the load change curves of each line and the bus voltage change curve; the prediction model predicts the absolute value of the bus voltage after the set time length according to the determined bus voltage change curve, and calculates the voltage change rate within the set time length according to the bus voltage change curve.

[0014] Preferably, the step S4 specifically includes:

[0015] If the absolute value of the bus voltage is greater than or equal to the second threshold value, and the voltage change rate is greater than or equal to the third threshold value, a first-level over-limit alarm is issued, and the load level of the line is determined. If the load level is high, power dispatch is performed according to the queried load change curve and power generation plan; if the load level is low, the line load is cut off;

[0016] If the absolute value of the bus voltage is greater than or equal to the second threshold value, and the voltage change rate is less than the third threshold value, a second-level over-limit alarm is issued, the queried load change curve is input into the load forecasting model, and then power dispatch is performed according to the power generation plan and the output result of the load forecasting model;

[0017] If the absolute value of the bus voltage is less than the second threshold value, and the voltage change rate is greater than or equal to the third threshold value, the absolute value of the bus voltage after outputting twice the set time is further predicted. If the absolute value of the bus voltage is still less than the second threshold value, the over-limit situation is ignored. If the absolute value of the bus voltage is greater than or equal to the second threshold value, a first-level over-limit alarm is issued, and the load level of the line is determined. If the load level is high, power dispatch is performed according to the queried load change curve and power generation plan. If the load level is low, the line load is cut off.

[0018] If the absolute value of the bus voltage is less than the second threshold value and the voltage change rate is less than the third threshold value, the over-limit situation is ignored.

[0019] Preferably, the step S3 also includes: when querying and drawing the stored load change curves of each line from the historical database for comparison, if no load change curve with a degree of overlap between the two higher than a fourth threshold is found, an alarm is output to prompt the operator to make a judgment, and perform corresponding operations according to expert advice, and then the load change curve is stored in the historical database.

[0020] Preferably, the expert database periodically evaluates the operational stability of the power grid dispatching system to obtain evaluation results and expert experience values, and inputs the above evaluation results, the first, second, and third thresholds of the system within the period, and the expert experience values ​​into the neural network model. The neural network model predicts and outputs new first, second, and third thresholds, and uses the new first, second, and third thresholds to perform intelligent dynamic monitoring of the power grid dispatching system.

[0021] Preferably, the first threshold is smaller than the second threshold.

[0022] The present invention also includes an intelligent dynamic monitoring system for a power grid dispatching system, the monitoring system comprising a sensor unit, a historical database, a filtering unit, a query unit and a control unit;

[0023] Among them, sensor modules are respectively set at corresponding positions in each line to collect the load and bus voltage in each line in real time and store them in the historical database;

[0024] When the filter unit determines that the absolute value of the bus voltage in the current line exceeds the first threshold and exceeds the preset time, the load change curve of the preset time period before the voltage in the bus line exceeds the limit is extracted from the historical database and input into the query unit;

[0025] After receiving the input load change curve, the query unit extracts the load change curve of the preset time period before the voltage in the bus line exceeds the limit from the historical database, then queries and draws the stored load change curves of each line from the historical database, and transmits the query result to the control unit;

[0026] The comparison module in the control unit receives the query result and compares it with the load change curve of the preset time period before the voltage in the bus line exceeds the limit, and obtains a voltage change trend similar to the load change curve and inputs it into the prediction model in the control unit;

[0027] The prediction model predicts the absolute value of the bus voltage after a set time period according to the determined bus voltage change curve, and calculates the voltage change rate within the set time period according to the bus voltage change curve, and transmits the prediction result to the processing module in the control unit. The processing module determines whether the bus voltage exceeds the limit based on the predicted calculation of the absolute value of the bus voltage after the set time period and the voltage change rate within the set time period, and performs corresponding processing according to the judgment result.

[0028] Preferably, the monitoring system further includes an alarm unit.

[0029] Preferably, the line loads and bus voltages extracted and stored in the historical data are used to draw line load change curves and bus voltage change curves respectively, and the line load change curves and bus voltage change curves are stored in a corresponding and associated manner.

[0030] Preferably, the comparison module in the control unit receives the query result and the load change curve of the preset time period before the voltage in the bus line exceeds the limit and compares them. When it is determined that the coincidence degree between the two is higher than the fourth threshold value, the two are judged to be similar, and the prediction model of the bus voltage change curve input into the control unit is determined based on the corresponding correlation relationship between the load change curve of each line and the bus voltage change curve.

[0031] Preferably, the processing module receives the absolute value of the bus voltage after the set time period and the voltage change rate within the set time period transmitted by the prediction module, and performs the following judgment processing:

[0032] If the absolute value of the bus voltage is greater than or equal to the second threshold value, and the voltage change rate is greater than or equal to the third threshold value, the processing module sends a first-level over-limit alarm signal to the alarm unit, and determines the load level of the line. If the load level is high, power dispatch is performed according to the load change curve and power generation plan obtained by the query unit; if the load level is low, the line load is cut off;

[0033] If the absolute value of the bus voltage is greater than or equal to the second threshold value, and the voltage change rate is less than the third threshold value, the processing module sends a secondary over-limit alarm signal to the alarm unit, inputs the load change curve obtained by the query unit into the load forecasting model, and then performs power dispatch according to the power generation plan obtained by the query unit and the output result of the load forecasting model in the processing module;

[0034] If the absolute value of the bus voltage is less than the second threshold value, and the voltage change rate is greater than or equal to the third threshold value, the control unit controls the prediction model to further predict and output the absolute value of the bus voltage after twice the set time. If the absolute value of the bus voltage is still less than the second threshold value, the over-limit situation is ignored; if the absolute value of the bus voltage is greater than or equal to the second threshold value, the processing module sends a first-level over-limit alarm signal to the alarm unit, and determines the load level of the line. If the load level is high, power dispatch is performed according to the load change curve and power generation plan obtained by the query unit; if the load level is low, the line load is cut off;

[0035] If the absolute value of the bus voltage is less than the second threshold value and the voltage change rate is less than the third threshold value, the over-limit situation is ignored.

[0036] Preferably, when the comparison module of the control unit receives the query result and the load change curve of the preset time period before the voltage in the bus line exceeds the limit and compares them, when the output is empty, that is, no load change curve whose overlap degree between the two is higher than the fourth threshold is found, the control unit controls the processing module to output an alarm signal to the alarm unit, and the alarm unit emits sound and light to prompt the operator to make further judgments, and perform corresponding operations according to expert suggestions, and then stores this load change curve in the historical database.

[0037] Preferably, the monitoring system also includes an expert database, which periodically obtains historical data stored in the intelligent dynamic monitoring history database of the power grid dispatching system, evaluates the operating stability of the power grid dispatching system, obtains evaluation results and expert experience values, and then inputs the above-mentioned evaluation results, the first, second, and third thresholds of the system within the period, and the expert experience values ​​into the neural network model in the control unit. The neural network model predicts and outputs new first, second, and third thresholds, and uses the new first, second, and third thresholds to perform intelligent dynamic monitoring of the power grid dispatching system.

[0038] The beneficial effects of the present invention are as follows: 1. A filtering unit is provided to filter the collected bus voltage line, and only the bus voltage whose absolute value exceeds the first threshold value and exceeds the preset time length is processed and monitored. The setting of the filtering unit can filter out the small-range fluctuation of the bus voltage, that is, the judgment and detection is started only when the bus voltage has an obvious over-limit trend, which reduces the amount of data of the system alarm, improves the monitoring efficiency, and reduces the workload of the dispatching operator;

[0039] The monitoring system makes over-limit judgments based on multi-dimensional data such as bus voltage, voltage change rate, and load change, and obtains different judgment results based on the set second and third thresholds, that is, obtains different over-limit types and severity levels, and selects corresponding dispatching operations based on different judgment results, making the main grid dispatch more accurate and reasonable, and improving the grid operation efficiency;

[0040] The monitoring system regularly evaluates and updates the first, second, and third thresholds based on expert advice and the historical operating stability of the power grid dispatching system, making the dynamic thresholds more in line with the actual operating conditions of the ever-changing main grid system, thereby improving the operating accuracy of the main grid system and reducing the occurrence of misjudgments and misoperations. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 It is a step diagram of the intelligent dynamic monitoring method of the power grid dispatching system;

[0042] Figure 2 This is a step diagram of the control unit processing model judgment processing method. DETAILED DESCRIPTION

[0043] The technical solution of the present invention is further described in detail below in conjunction with the accompanying drawings, but the protection scope of the present invention is not limited to the following.

[0044] The present invention provides an intelligent dynamic monitoring method for a power grid dispatching system, such as Figure 1 As shown, the following steps are taken:

[0045] S1: Real-time acquisition and storage of loads and bus voltages in each line of the power grid;

[0046] S2: Filter out bus voltages whose absolute values ​​exceed the first threshold and exceed a preset time length, and extract load changes in the bus line whose voltage exceeds the limit in a preset time period before the voltage exceeds the limit;

[0047] S3: query the voltage change trend similar to the load change from the historical database and input it into the prediction model. The prediction model predicts and calculates the absolute value of the bus voltage after a set time and the voltage change rate within the set time;

[0048] S4: Determine whether the bus voltage exceeds the limit based on the predicted absolute value of the bus voltage after the set time and the voltage change rate within the set time, and perform corresponding processing based on the judgment result.

[0049] Preferably, in step S1, the load and bus voltage in each line in the power grid are acquired and stored in real time, specifically including: setting sensors at corresponding positions in each line to collect the load and bus voltage in each line in real time and store them in a historical database, extracting the stored load and bus voltage of each line from the historical database to draw the load change curve and bus voltage change curve of each line, and correspondingly associating the load change curve and bus voltage change curve of each line.

[0050] Preferably, in step S2, the load change situation of the bus line with voltage over-limit in the preset time period before the voltage over-limit is extracted, specifically including: when the filtering unit determines that the absolute value of the bus voltage in the current line exceeds the first threshold value and exceeds the preset time length, the load change curve of the bus line in the preset time period before the voltage over-limit is extracted from the historical database. Since the small-range fluctuation amplitude of the bus voltage of the main grid line is small and the duration is short, it will not affect the main grid. The setting of the filtering unit can filter out such small-range fluctuations, that is, the judgment and detection will only be started when the bus voltage has an obvious over-limit trend, which reduces the amount of data of the system alarm, improves the monitoring efficiency, and reduces the workload of the dispatching operator.

[0051] Preferably, the step S3 specifically includes: extracting the load change curve of the preset time period before the voltage in the bus line exceeds the limit from the historical database, and then querying and drawing the stored load change curves of each line from the historical database for comparison; when the coincidence degree between the two is higher than the fourth threshold value, it is determined that the two are similar, and the bus voltage change curve at this time is input into the prediction model according to the corresponding correlation relationship between the load change curves of each line and the bus voltage change curve; the prediction model predicts the absolute value of the bus voltage after the set time length according to the determined bus voltage change curve, and calculates the voltage change rate within the set time length according to the bus voltage change curve.

[0052] Preferably, the step S4 is as follows Figure 2 As shown, specifically including:

[0053] If the absolute value of the bus voltage is greater than or equal to the second threshold value, and the voltage change rate is greater than or equal to the third threshold value, a first-level over-limit alarm is issued, and the load level of the line is determined. If the load level is high, power dispatch is performed according to the queried load change curve and power generation plan; if the load level is low, the line load is cut off;

[0054] If the absolute value of the bus voltage is greater than or equal to the second threshold value, and the voltage change rate is less than the third threshold value, a second-level over-limit alarm is issued, the queried load change curve is input into the load forecasting model, and then power dispatch is performed according to the power generation plan and the output result of the load forecasting model;

[0055] If the absolute value of the bus voltage is less than the second threshold value, and the voltage change rate is greater than or equal to the third threshold value, the absolute value of the bus voltage after outputting twice the set time is further predicted. If the absolute value of the bus voltage is still less than the second threshold value, the over-limit situation is ignored. If the absolute value of the bus voltage is greater than or equal to the second threshold value, a first-level over-limit alarm is issued, and the load level of the line is determined. If the load level is high, power dispatch is performed according to the queried load change curve and power generation plan. If the load level is low, the line load is cut off.

[0056] If the absolute value of the bus voltage is less than the second threshold value and the voltage change rate is less than the third threshold value, the over-limit situation is ignored.

[0057] Preferably, the step S3 also includes: when querying and drawing the stored load change curves of each line from the historical database for comparison, if no load change curve with a degree of overlap between the two higher than a fourth threshold is found, an alarm is output to prompt the operator to make a judgment, and perform corresponding operations according to expert advice, and then the load change curve is stored in the historical database.

[0058] Preferably, the expert database periodically evaluates the operational stability of the power grid dispatching system to obtain evaluation results and expert experience values, and inputs the above evaluation results, the first, second, and third thresholds of the system within the period, and the expert experience values ​​into the neural network model. The neural network model predicts and outputs new first, second, and third thresholds, and uses the new first, second, and third thresholds to perform intelligent dynamic monitoring of the power grid dispatching system.

[0059] Preferably, the first threshold is smaller than the second threshold.

[0060] The present invention also includes an intelligent dynamic monitoring system for a power grid dispatching system, the monitoring system comprising a sensor unit, a historical database, a filtering unit, a query unit and a control unit;

[0061] Among them, sensor modules are respectively set at corresponding positions in each line to collect the load and bus voltage in each line in real time and store them in the historical database;

[0062] When the filter unit determines that the absolute value of the bus voltage in the current line exceeds the first threshold and exceeds the preset time, the load change curve of the preset time period before the voltage in the bus line exceeds the limit is extracted from the historical database and input into the query unit;

[0063] After receiving the input load change curve, the query unit extracts the load change curve of the preset time period before the voltage in the bus line exceeds the limit from the historical database, then queries and draws the stored load change curves of each line from the historical database, and transmits the query result to the control unit;

[0064] The comparison module in the control unit receives the query result and compares it with the load change curve of the preset time period before the voltage in the bus line exceeds the limit, and obtains a voltage change trend similar to the load change curve and inputs it into the prediction model in the control unit;

[0065] The prediction model predicts the absolute value of the bus voltage after a set time period according to the determined bus voltage change curve, and calculates the voltage change rate within the set time period according to the bus voltage change curve, and transmits the prediction result to the processing module in the control unit. The processing module determines whether the bus voltage exceeds the limit based on the predicted calculation of the absolute value of the bus voltage after the set time period and the voltage change rate within the set time period, and performs corresponding processing according to the judgment result.

[0066] Preferably, the monitoring system further includes an alarm unit.

[0067] Preferably, the line loads and bus voltages extracted and stored in the historical data are used to draw line load change curves and bus voltage change curves respectively, and the line load change curves and bus voltage change curves are stored in a corresponding and associated manner.

[0068] Preferably, the comparison module in the control unit receives the query result and the load change curve of the preset time period before the voltage in the bus line exceeds the limit and compares them. When it is determined that the coincidence degree between the two is higher than the fourth threshold value, the two are judged to be similar, and the prediction model of the bus voltage change curve input into the control unit is determined based on the corresponding correlation relationship between the load change curve of each line and the bus voltage change curve.

[0069] Preferably, the prediction model predicts and outputs the absolute value of the bus voltage after the set time according to the determined bus voltage change curve, including: the prediction model extracts the bus voltage V after the set time from the bus voltage change curve according to the bus voltage change curve determined by the query unit. 设 , and then predict the absolute value of the bus voltage after the set time by comparing the output of the comparison module with the comparison similarity a Wherein a is the similarity value between the load change curve extracted in the preset time period before the voltage in the bus line exceeds the limit and the line load change curve determined by the query.

[0070] The prediction model calculates the voltage change rate within the set time period according to the bus voltage change curve, including: the prediction model extracts the number of half-waves n of voltage fluctuation within the set time period according to the bus voltage change curve determined by the query unit, and if the number of half-waves is 1, then ; where V 当前 is the current bus voltage value, t is the set time;

[0071] If the number of half-waves is greater than 1, the sum of the voltage change rates of all half-waves within the set time is calculated.和 ,but ; The voltage change rate of one half-wave is equal to the peak-to-valley voltage difference of the half-wave / the length of the half-wave, n>1.

[0072] Preferably, the processing module receives the absolute value of the bus voltage after the set time period and the voltage change rate within the set time period transmitted by the prediction module, and performs the following judgment processing:

[0073] If the absolute value of the bus voltage is greater than or equal to the second threshold value, and the voltage change rate is greater than or equal to the third threshold value, the processing module sends a first-level over-limit alarm signal to the alarm unit, and determines the load level of the line. If the load level is high, power dispatch is performed according to the load change curve and power generation plan obtained by the query unit; if the load level is low, the line load is cut off;

[0074] If the absolute value of the bus voltage is greater than or equal to the second threshold value, and the voltage change rate is less than the third threshold value, the processing module sends a secondary over-limit alarm signal to the alarm unit, inputs the load change curve obtained by the query unit into the load forecasting model, and then performs power dispatch according to the power generation plan obtained by the query unit and the output result of the load forecasting model in the processing module;

[0075] If the absolute value of the bus voltage is less than the second threshold value, and the voltage change rate is greater than or equal to the third threshold value, the control unit controls the prediction model to further predict and output the absolute value of the bus voltage after twice the set time. If the absolute value of the bus voltage is still less than the second threshold value, the over-limit situation is ignored; if the absolute value of the bus voltage is greater than or equal to the second threshold value, the processing module sends a first-level over-limit alarm signal to the alarm unit, and determines the load level of the line. If the load level is high, power dispatch is performed according to the load change curve and power generation plan obtained by the query unit; if the load level is low, the line load is cut off;

[0076] If the absolute value of the bus voltage is less than the second threshold value and the voltage change rate is less than the third threshold value, the over-limit situation is ignored.

[0077] Preferably, when the comparison module of the control unit receives the query result and the load change curve of the preset time period before the voltage in the bus line exceeds the limit and compares them, when the output is empty, that is, no load change curve whose overlap degree between the two is higher than the fourth threshold is found, the control unit controls the processing module to output an alarm signal to the alarm unit, and the alarm unit emits sound and light to prompt the operator to make further judgments, and perform corresponding operations according to expert suggestions, and then stores this load change curve in the historical database.

[0078] Preferably, the monitoring system also includes an expert database, which periodically obtains historical data stored in the intelligent dynamic monitoring history database of the power grid dispatching system, evaluates the operational stability of the power grid dispatching system, obtains evaluation results and expert experience values, and then inputs the above evaluation results, the first, second, and third thresholds of the system within the period, and the expert experience values ​​into the neural network model in the control unit, and the neural network model predicts and outputs new first, second, and third thresholds, and uses the new first, second, and third thresholds to perform intelligent dynamic monitoring of the power grid dispatching system. Specifically, in the neural network model ,in To evaluate the results, is the expert experience value, and the new second threshold and third threshold are obtained in the same way.

[0079] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing related hardware through a computer program, and the program can be stored in a computer-readable storage medium, and when the program is executed, it can include the processes of the embodiments of the above-mentioned methods. The storage medium can be a disk, an optical disk, a ROM, a RAM, etc.

[0080] The above disclosure is only the preferred embodiment of the present invention, which certainly cannot be used to limit the scope of the present invention. Therefore, equivalent changes made according to the claims of the present invention are still within the scope of the present invention.

Claims

1. An intelligent dynamic monitoring method for a power grid dispatching system, characterized in that: Use the following steps: S1: Real-time acquisition and storage of loads and bus voltages in each line of the power grid; In the step S1, the load and bus voltage of each line in the power grid are acquired in real time and stored, specifically comprising: setting sensors at corresponding positions in each line to acquire the load and bus voltage of each line in real time and store them in a historical database, extracting the stored load and bus voltage of each line from the historical database to draw the load change curve and bus voltage change curve of each line, and correspondingly associating the load change curve of each line with the bus voltage change curve; S2: Filter out bus voltages whose absolute values ​​exceed the first threshold and exceed a preset time length, and extract load changes in the bus line whose voltage exceeds the limit in a preset time period before the voltage exceeds the limit; In the step S2, the load change situation of the bus line with voltage exceeding the limit in the preset time period before the voltage exceeds the limit is extracted, which specifically includes: when the filtering unit determines that the absolute value of the bus voltage in the current line exceeds the first threshold and exceeds the preset time, the load change curve of the bus line in the preset time period before the voltage exceeds the limit is extracted from the historical database; S3: query the voltage change trend similar to the load change from the historical database and input it into the prediction model. The prediction model predicts and calculates the absolute value of the bus voltage after a set time and the voltage change rate within the set time; The step S3 specifically includes: extracting the load change curve of the preset time period before the voltage in the bus line exceeds the limit from the historical database, and then querying and drawing the stored load change curves of each line from the historical database for comparison, and when the coincidence degree between the two is higher than the fourth threshold, it is determined that the two are similar, and the bus voltage change curve at this time is determined according to the corresponding correlation relationship between the load change curves of each line and the bus voltage change curve to input the prediction model; the prediction model predicts the absolute value of the bus voltage after the set time length according to the determined bus voltage change curve, and calculates the voltage change rate within the set time length according to the bus voltage change curve; S4: judging whether the bus voltage exceeds the limit according to the absolute value of the bus voltage after the predicted calculation of the set time and the voltage change rate within the set time, and performing corresponding processing according to the judgment result; The step S4 specifically includes: If the absolute value of the bus voltage is greater than or equal to the second threshold value, and the voltage change rate is greater than or equal to the third threshold value, a first-level over-limit alarm is issued, and the load level of the line is determined. If the load level is high, power dispatch is performed according to the queried load change curve and power generation plan; if the load level is low, the line load is cut off; If the absolute value of the bus voltage is greater than or equal to the second threshold value, and the voltage change rate is less than the third threshold value, a second-level over-limit alarm is issued, the queried load change curve is input into the load forecasting model, and then power dispatch is performed according to the power generation plan and the output result of the load forecasting model; If the absolute value of the bus voltage is less than the second threshold value, and the voltage change rate is greater than or equal to the third threshold value, the absolute value of the bus voltage after outputting twice the set time is further predicted. If the absolute value of the bus voltage is still less than the second threshold value, the over-limit situation is ignored. If the absolute value of the bus voltage is greater than or equal to the second threshold value, a first-level over-limit alarm is issued, and the load level of the line is determined. If the load level is high, power dispatch is performed according to the queried load change curve and power generation plan. If the load level is low, the line load is cut off. If the absolute value of the bus voltage is less than the second threshold value and the voltage change rate is less than the third threshold value, the over-limit situation is ignored.

2. An intelligent dynamic monitoring method for a power grid dispatching system according to claim 1, characterized in that: The step S3 also includes: when querying and drawing the stored load change curves of each line from the historical database for comparison, if no load change curve with a degree of overlap higher than a fourth threshold is found, an alarm is output to prompt the operator to make a judgment, and perform corresponding operations according to expert advice, and then the load change curve is stored in the historical database.

3. An intelligent dynamic monitoring method for a power grid dispatching system according to claim 1, characterized in that: The expert database regularly evaluates the operational stability of the power grid dispatching system to obtain evaluation results and expert experience values, and inputs the above evaluation results, the first, second, and third thresholds of the system within the period, and the expert experience values ​​into the neural network model. The neural network model predicts and outputs new first, second, and third thresholds, and uses the new first, second, and third thresholds to perform intelligent dynamic monitoring of the power grid dispatching system.

4. An intelligent dynamic monitoring system for a power grid dispatching system, using the intelligent dynamic monitoring method for a power grid dispatching system as claimed in claim 1, characterized in that: The monitoring system includes a sensor unit, a historical database, a filtering unit, a query unit and a control unit; Among them, sensor modules are respectively set at corresponding positions in each line to collect the load and bus voltage in each line in real time and store them in the historical database; When the filter unit determines that the absolute value of the bus voltage in the current line exceeds the first threshold and exceeds the preset time, the load change curve of the preset time period before the voltage in the bus line exceeds the limit is extracted from the historical database and input into the query unit; After receiving the input load change curve, the query unit extracts the load change curve of the preset time period before the voltage in the bus line exceeds the limit from the historical database, then queries and draws the stored load change curves of each line from the historical database, and transmits the query result to the control unit; The comparison module in the control unit receives the query result and compares it with the load change curve of the preset time period before the voltage in the bus line exceeds the limit, and obtains a voltage change trend similar to the load change curve and inputs it into the prediction model in the control unit; The prediction model predicts the absolute value of the bus voltage after a set time period according to the determined bus voltage change curve, and calculates the voltage change rate within the set time period according to the bus voltage change curve, and transmits the prediction result to the processing module in the control unit. The processing module determines whether the bus voltage exceeds the limit based on the predicted calculation of the absolute value of the bus voltage after the set time period and the voltage change rate within the set time period, and performs corresponding processing according to the judgment result.

5. An intelligent dynamic monitoring system for a power grid dispatching system as claimed in claim 4, characterized in that: The comparison module in the control unit receives the query result and the load change curve of the preset time period before the voltage in the bus line exceeds the limit and compares them. When it is determined that the coincidence degree between the two is higher than the fourth threshold value, the two are judged to be similar, and the prediction model of the bus voltage change curve input into the control unit is determined based on the corresponding correlation relationship between the load change curve of each line and the bus voltage change curve.

6. An intelligent dynamic monitoring system for a power grid dispatching system as claimed in claim 4, characterized in that: The monitoring system also includes an expert database, which periodically obtains historical data stored in an intelligent dynamic monitoring history database of the power grid dispatching system, evaluates the operating stability of the power grid dispatching system, obtains evaluation results and expert experience values, and then inputs the above evaluation results, the first, second, and third thresholds of the system within the period, and the expert experience values ​​into a neural network model in the control unit. The neural network model predicts and outputs new first, second, and third thresholds, and uses the new first, second, and third thresholds to perform intelligent dynamic monitoring of the power grid dispatching system.

Images