Power grid safety early warning analysis method and system

Abstract

The invention discloses a power grid safety early warning analysis method and system, and belongs to the technical field of power grids, and the method comprises the steps: obtaining real-time operation data and real-time risk assessment parameters of a power grid in a monitoring period; calculating an operation state evaluation coefficient of the power grid in the monitoring period according to the real-time operation data in the monitoring period; determining a fault probability value of the power grid in the monitoring period according to the risk assessment parameters in the monitoring period; determining a power grid operation stability evaluation coefficient in the monitoring period according to the operation state evaluation coefficient of the power grid in the monitoring period and the fault probability value of the power grid in the monitoring period, and generating power grid safety early warning information according to the power grid operation stability evaluation coefficient; through the implementation of the method and the device, the problem that the finally obtained power grid safety early warning result is inaccurate due to the fact that the analysis on the operation data and the risk assessment parameters is lacked in the prior art and the analysis on the direct factors influencing the power grid fault is not deep enough can be solved.

Description

Technical Field

[0001] The present invention relates to the technical field of power grids, and in particular to a power grid security early warning analysis method and system. Background Art

[0002] The power grid bears the heavy responsibility of reliably transmitting electric energy to end users. Once a line trips and a power outage occurs, it will have a great impact on the production and living order of enterprises, institutions and residents along the line. The main cause of power outages is power grid failure. Therefore, power grid failure analysis is necessary.

[0003] The patent application with publication number CN116937543A discloses a safety early warning analysis system for power grids. The system collects basic data from multiple data sources through a data access module to obtain multi-source data. The basic data includes exogenous meteorological data, exogenous public data, and distribution network business data. After obtaining the multi-source data, the multi-source data is fused and processed by a data fusion module, and then processed by a data analysis module to obtain an early warning result. However, in this method of analyzing faults through multi-source data, the impact of the collected multi-source data, such as exogenous meteorological data and exogenous public data, on the fault is indirect, and the distribution network business data is the factor that directly affects the fault. The main factors that affect power grid faults in the distribution network business data are operating data and risk assessment parameters. However, the patent application lacks analysis of operating data and risk assessment parameters, resulting in insufficient in-depth analysis of the direct factors affecting power grid faults, which leads to inaccurate early warning results. Summary of the Invention

[0004] The embodiments of the present invention provide a power grid security early warning analysis method and system, which can solve the problem that the existing technology lacks analysis of operating data and risk assessment parameters, resulting in insufficient analysis of direct factors affecting power grid failures, and causing inaccurate power grid security early warning results.

[0005] An embodiment of the present invention provides a power grid security early warning analysis method, comprising:

[0006] Acquire real-time operating data and real-time risk assessment parameters of the power grid during the monitoring period; wherein the operating data includes: operating voltage, operating current, operating power, and operating temperature; and the risk assessment parameters include: operating voltage value of key equipment, operating current value of key equipment, temperature value of key equipment, number of circuit breaker operations, and insulation resistance value;

[0007] Calculate the average operating voltage change rate, average operating current change rate, average operating power change rate, and average operating temperature change rate during the monitoring period based on the real-time operating data during the monitoring period, and calculate the operating status assessment coefficient of the power grid during the monitoring period based on the average operating voltage change rate, average operating current change rate, average operating power change rate, and average operating temperature change rate during the monitoring period;

[0008] Constructing a fault risk characteristic matrix of the power grid within the monitoring period according to the risk assessment parameters within the monitoring period, and determining a fault probability value of the power grid within the monitoring period according to the fault risk characteristic matrix within the monitoring period;

[0009] According to the operation status evaluation coefficient of the power grid during the monitoring period and the fault probability value of the power grid during the monitoring period, the power grid operation stability evaluation coefficient during the monitoring period is determined, and the power grid safety warning information is generated according to the power grid operation stability evaluation coefficient.

[0010] Furthermore, the calculation of the average operating voltage change rate, the average operating current change rate, the average operating power change rate, and the average operating temperature change rate within the monitoring period based on the real-time operating data within the monitoring period includes:

[0011] Divide the monitoring period into several monitoring time periods, and obtain real-time operation data corresponding to the several monitoring time periods;

[0012] For each monitoring period, the operating voltage change rate within the current monitoring period is determined based on the duration of the current monitoring period and the real-time operating voltage; the operating current change rate within the current monitoring period is determined based on the duration of the current monitoring period and the real-time operating current; the operating power change rate within the current monitoring period is determined based on the duration of the current monitoring period and the real-time operating power; and the operating temperature change rate is determined based on the duration of the current monitoring period and the real-time operating temperature.

[0013] The average operating voltage change rate is calculated based on the operating voltage change rate in each monitoring period, the average operating current change rate is calculated based on the operating current change rate in each monitoring period, the average operating power change rate is calculated based on the operating power change rate in each monitoring period, and the average operating temperature change rate is calculated based on the operating temperature change rate in each monitoring period.

[0014] Furthermore, the calculation of the operating status assessment coefficient of the power grid during the monitoring period based on the average operating voltage change rate, the average operating current change rate, the average operating power change rate, and the average operating temperature change rate during the monitoring period includes:

[0015] The operating status assessment coefficient of the power grid during the monitoring period is calculated by using the average operating voltage change rate, average operating current change rate, average operating power change rate, average operating temperature change rate and the operating status assessment coefficient calculation formula during the monitoring period;

[0016] The calculation formula of the operating status assessment coefficient is specifically:

[0017] ZPX=Z1*PDY+Z2*PDL+Z3*PGL+Z4*PWD;

[0018] Among them, ZPX represents the operating status assessment coefficient of the power grid during the monitoring period; PDY represents the average operating voltage change rate; PDL represents the average operating current change rate; PGL represents the average operating power change rate; PWD represents the average operating temperature change rate; Z1, Z2, Z3 and Z4 are the weight coefficients of PDY, PDL, PGL and PWD respectively.

[0019] Furthermore, constructing a fault risk feature matrix of the power grid within the monitoring period according to the risk assessment parameters within the monitoring period includes:

[0020] Divide the monitoring period into several monitoring periods, and obtain real-time risk assessment parameters corresponding to the monitoring periods;

[0021] Determine the abnormal operating voltage risk value based on the real-time operating voltage value of key equipment in each monitoring period;

[0022] Determine the phase current abnormality risk value based on the operating current value of key equipment in each monitoring period;

[0023] Determine the abnormal temperature risk angle of key equipment based on the temperature of key equipment in each monitoring period;

[0024] determining a circuit breaker operation number difference according to the circuit breaker operation number in each monitoring period;

[0025] Determine the insulation resistance difference based on the insulation resistance value of each monitoring period;

[0026] Based on the abnormal operating voltage risk value, phase current risk value, key equipment temperature abnormality risk angle, circuit breaker operation times difference and insulation resistance difference, a fault risk feature matrix within the monitoring period is constructed.

[0027] Furthermore, determining the operating voltage abnormality risk value based on the real-time operating voltage value of key equipment in each monitoring period includes:

[0028] For each monitoring period, a characteristic curve of the key equipment operating voltage value of the current monitoring period is generated based on the real-time key equipment operating voltage value of the current monitoring period;

[0029] Determine a first area based on the key equipment operating voltage value characteristic curve and the preset key equipment operating voltage value characteristic curve during the current monitoring period; wherein the first area is the area enclosed by a line segment of the key equipment operating voltage value characteristic curve located above the preset key equipment operating voltage value characteristic curve and the preset key equipment operating voltage value characteristic curve;

[0030] Determine the second area based on the area enclosed by the characteristic curve of the key equipment operating voltage value during the current monitoring period and the horizontal axis of the first rectangular coordinate system where the characteristic curve of the key equipment operating voltage value is located; wherein the horizontal axis of the first rectangular coordinate system represents the monitoring time, and the vertical axis of the second rectangular coordinate system represents the key equipment operating voltage value;

[0031] Divide the first area by the second area to obtain the operating voltage abnormality multiplier value during the current monitoring period;

[0032] A first set is generated according to the operating voltage anomaly multiplier value of each monitoring period, the maximum subset and the minimum subset of the first set are obtained, and the operating voltage anomaly risk value is determined according to the average of the difference between the maximum subset of the first set and the minimum subset of the first set.

[0033] Furthermore, determining the phase current abnormality risk value according to the operating current value of the key equipment in each monitoring period includes:

[0034] Generating a second set according to the operating current values of the key equipment in each monitoring period, obtaining a mean value of the second set, and marking the mean value in the second set as a phase current mean value;

[0035] Constructing a second rectangular coordinate system; wherein the horizontal axis of the second rectangular coordinate system is the monitoring time, and the vertical axis of the second rectangular coordinate system is the phase current average;

[0036] generating a phase current mean curve in a second rectangular coordinate system according to points marked as phase current mean values in the second set;

[0037] The third area and the first time length are determined based on the phase current mean curve and the preset phase current mean threshold curve of the current monitoring period, and the phase current abnormality risk value is determined based on the third area and the first time length; wherein, the third area is the area enclosed by the line segment of the phase current mean curve above the preset phase current mean threshold curve and the preset phase current mean threshold curve; the first time length is the time length corresponding to the line segment of the phase current mean curve above the preset phase current mean threshold curve.

[0038] Furthermore, the abnormal temperature risk angle of key equipment is determined based on the temperature of key equipment in each monitoring period, including:

[0039] For each monitoring period, the temperature change value of the key equipment in the current monitoring period is obtained according to the difference between the temperature of the key equipment in the current monitoring period and the temperature of the key equipment in the next monitoring period;

[0040] Construct a third rectangular coordinate system; wherein the horizontal axis of the third rectangular coordinate system is the monitoring time, and the vertical axis of the third rectangular coordinate system is the temperature change value of the key equipment;

[0041] A key equipment temperature change value curve is generated in the third rectangular coordinate system based on the temperature change value of the key equipment in each monitoring period; the key equipment temperature abnormality risk angle is determined based on the acute angle value formed when the key equipment temperature change value curve intersects with the preset key equipment temperature change threshold curve for the first time.

[0042] Furthermore, determining a circuit breaker operation number difference according to the circuit breaker operation number in each monitoring period, and determining an insulation resistance difference according to the insulation resistance value in each monitoring period, includes:

[0043] generating a third set according to the number of circuit breaker operations in each monitoring period, and determining a difference in the number of circuit breaker operations according to an average of differences between a maximum subset of the third set and a minimum subset of the third set;

[0044] A fourth set is generated according to the insulation resistance values of each monitoring period, and an insulation resistance difference is determined according to an average of the differences between the largest subset of the fourth set and the smallest subset of the fourth set.

[0045] Furthermore, determining the fault probability value of the power grid within the monitoring period according to the fault risk characteristic matrix within the monitoring period includes:

[0046] Inputting the fault risk feature matrix within the monitoring period into the logistic regression model so that the logistic regression model outputs a fault probability value of the power grid within the monitoring period;

[0047] The construction of the logistic regression model includes:

[0048] Obtaining several historical risk assessment parameters and failure probability value labels corresponding to each historical risk assessment parameter;

[0049] For each historical risk assessment parameter, a fault risk feature matrix sample of the current historical risk assessment parameter is generated according to the historical risk assessment parameter;

[0050] An initial logistic regression model is constructed, with the fault risk feature matrix samples as input and the fault probability value labels corresponding to the fault risk feature matrix samples as output. The initial logistic regression model is trained until the initial logistic regression model reaches a preset convergence condition, thereby generating the logistic regression model.

[0051] Based on the above method embodiment, the present invention provides a corresponding system embodiment;

[0052] An embodiment of the present invention provides a microgrid steady-state operation control system, comprising: a real-time data acquisition module, an operation status assessment coefficient determination module, a fault probability value determination module, and an early warning information generation module;

[0053] The real-time data acquisition module is used to obtain real-time operating data and real-time risk assessment parameters of the power grid during the monitoring period; wherein the operating data includes: operating voltage, operating current, operating power and operating temperature; the risk assessment parameters include: operating voltage value of key equipment, operating current value of key equipment, temperature value of key equipment, number of circuit breaker operations and insulation resistance value;

[0054] The operating status assessment coefficient determination module is used to calculate the average operating voltage change rate, the average operating current change rate, the average operating power change rate and the average operating temperature change rate within the monitoring period based on the real-time operating data within the monitoring period, and calculate the operating status assessment coefficient of the power grid within the monitoring period based on the average operating voltage change rate, the average operating current change rate, the average operating power change rate and the average operating temperature change rate within the monitoring period;

[0055] The fault probability value determination module is used to construct a fault risk characteristic matrix of the power grid within the monitoring period according to the risk assessment parameters within the monitoring period, and determine the fault probability value of the power grid within the monitoring period according to the fault risk characteristic matrix within the monitoring period;

[0056] The warning information generation module is used to determine the grid operation stability assessment coefficient within the monitoring period based on the grid operation status assessment coefficient and the grid failure probability value within the monitoring period, and generate grid security warning information based on the grid operation stability assessment coefficient.

[0057] The following beneficial effects are achieved by implementing the present invention:

[0058] The present invention provides a power grid security early warning analysis method and system. The method obtains real-time operation data and real-time risk assessment parameters of the power grid during a monitoring period, then calculates an operation status assessment coefficient of the power grid during the monitoring period based on the real-time operation data, determines a fault probability value of the power grid during the monitoring period based on the risk assessment parameters, and finally determines a power grid operation stability assessment coefficient during the monitoring period based on the operation status assessment coefficient and the fault probability value of the power grid during the monitoring period. Power grid security early warning information is generated based on the power grid operation stability assessment coefficient. By analyzing the operation status and fault probability of the power grid based on the real-time operation data and real-time risk assessment parameters, the method solves the problem that the prior art lacks analysis of the operation data and risk assessment parameters, resulting in insufficient analysis of the direct factors affecting power grid failures and inaccurate power grid security early warning results. By collaboratively determining the power grid operation stability assessment coefficient based on both the operation status and the fault probability, the method makes the power grid operation stability analysis more comprehensive and accurate, improves the accuracy of the power grid security early warning information generated based on the power grid operation stability assessment coefficient, and ensures the safe operation of the power grid. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 The present invention provides a flowchart of a power grid security early warning analysis method according to an embodiment of the present invention.

[0060] Figure 2 This is a structural diagram of a power grid security early warning analysis system provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0061] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0062] See also Figure 1 To address the problem that the existing technology lacks analysis of operating data and risk assessment parameters, resulting in insufficient in-depth analysis of direct factors affecting power grid failures and inaccurate power grid security early warning results, an embodiment of the present invention provides a power grid security early warning analysis method, including:

[0063] Step S1: Acquire real-time operating data and real-time risk assessment parameters of the power grid during a monitoring period; wherein the operating data includes: operating voltage, operating current, operating power, and operating temperature; and the risk assessment parameters include: operating voltage value of key equipment, operating current value of key equipment, temperature value of key equipment, number of circuit breaker operations, and insulation resistance value;

[0064] Step S2: Calculating the average operating voltage change rate, average operating current change rate, average operating power change rate, and average operating temperature change rate during the monitoring period based on the real-time operating data during the monitoring period, and calculating the operating status assessment coefficient of the power grid during the monitoring period based on the average operating voltage change rate, average operating current change rate, average operating power change rate, and average operating temperature change rate during the monitoring period;

[0065] Step S3: constructing a fault risk characteristic matrix of the power grid within the monitoring period according to the risk assessment parameters within the monitoring period, and determining a fault probability value of the power grid within the monitoring period according to the fault risk characteristic matrix within the monitoring period;

[0066] Step S4: Determine the grid operation stability evaluation coefficient within the monitoring period according to the grid operation status evaluation coefficient and the grid failure probability value within the monitoring period, and generate grid security warning information according to the grid operation stability evaluation coefficient.

[0067] For step S1, the real-time operating data and real-time risk assessment parameters of the power grid during the monitoring period are obtained, wherein the operating data include the operating voltage, operating current, operating power and operating temperature of the power grid during the monitoring period; the risk assessment parameters include the operating voltage value of key equipment, the operating current value of key equipment, the temperature value of key equipment, the number of circuit breaker operations and the insulation resistance value. The operating voltage, operating current, operating power and operating temperature in the operating data refer to the overall operating data of the power grid, and the operating voltage value, operating current value and temperature value in the risk assessment parameters specifically refer to the operating voltage value, operating current value and temperature value of key equipment. The key equipment in the risk assessment parameters specifically refers to the transformer, the number of circuit breaker operations refers to the total number of times the circuit breaker performs tripping or closing operations within a period of time, and the insulation resistance value refers to the resistance value corresponding to the leakage current flowing through the insulating material when a DC voltage is applied.

[0068] Preferably, after the real-time operation data and the real-time risk assessment parameters are obtained, the obtained real-time operation data and risk assessment parameters may be preprocessed, including data cleaning, denoising and format conversion.

[0069] Through operations such as data cleaning, denoising and format conversion, the data quality of the obtained real-time operation data and real-time risk assessment parameters can be significantly improved, and the accuracy of the processing results obtained after subsequent processing using these data can be improved.

[0070] For step S2, in a preferred embodiment, the calculation of the average operating voltage change rate, average operating current change rate, average operating power change rate and average operating temperature change rate within the monitoring period based on the real-time operating data within the monitoring period includes: dividing the monitoring period into several monitoring periods, and obtaining real-time operating data corresponding to the several monitoring periods; for each monitoring period, determining the operating voltage change rate within the current monitoring period based on the duration of the current monitoring period and the real-time operating voltage, determining the operating current change rate within the current monitoring period based on the duration of the current monitoring period and the real-time operating current, determining the operating power change rate within the current monitoring period based on the duration of the current monitoring period and the real-time operating power, and determining the operating temperature change rate based on the duration of the current monitoring period and the real-time operating temperature; calculating the average operating voltage change rate based on the operating voltage change rate of each monitoring period, calculating the average operating current change rate based on the operating current change rate of each monitoring period, calculating the average operating power change rate based on the operating power change rate of each monitoring period, and calculating the average operating temperature change rate based on the operating temperature change rate of each monitoring period.

[0071] Specifically, the monitoring cycle is divided into several monitoring time periods to obtain several equally divided monitoring time periods. It should be noted that since the monitoring time periods are divided into equal parts, the length of each monitoring time period is the same. For each monitoring time period, the operating voltage change of the current monitoring time period is determined based on the real-time operating voltage of the current monitoring time period, and the operating voltage change of the current monitoring time period is divided by the duration of the current monitoring time period (the duration is determined based on the interval start time and interval end time of the current monitoring time period) and the operating voltage change rate within the current monitoring time period is determined. Similarly, the operating current change rate, operating power change rate and operating temperature change rate of the current monitoring time period can be obtained.

[0072] The average operating voltage change rate for all monitoring periods within the monitoring cycle is calculated to obtain the average operating voltage change rate within the monitoring cycle. Similarly, the average operating current change rate, average operating power change rate, and average operating temperature change rate within the monitoring cycle can be calculated.

[0073] Through the above calculation, the average change rate of each operation data of the power grid during the monitoring period can be determined.

[0074] In a preferred embodiment, the calculation of the grid operation status assessment coefficient within the monitoring period based on the average operating voltage change rate, the average operating current change rate, the average operating power change rate, and the average operating temperature change rate within the monitoring period includes:

[0075] The operating status assessment coefficient of the power grid during the monitoring period is calculated by using the average operating voltage change rate, average operating current change rate, average operating power change rate, average operating temperature change rate and the operating status assessment coefficient calculation formula during the monitoring period;

[0076] The calculation formula of the operating status assessment coefficient is specifically:

[0077] ZPX=Z1*PDY+Z2*PDL+Z3*PGL+Z4*PWD;

[0078] Among them, ZPX represents the operating status assessment coefficient of the power grid during the monitoring period; PDY represents the average operating voltage change rate; PDL represents the average operating current change rate; PGL represents the average operating power change rate; PWD represents the average operating temperature change rate; Z1, Z2, Z3 and Z4 are the weight coefficients of PDY, PDL, PGL and PWD respectively.

[0079] It should be noted that the sum of Z1, Z2, Z3, and Z4 is 1. Z1, Z2, Z3, and Z4 are weighted based on the degree of influence of each operating data on the grid's operating status, determined by analyzing historical operating data. This weighting ensures that the calculated operating status assessment coefficient accurately reflects the grid's actual operating status. If the average operating voltage rate of change, average operating current rate of change, average operating power rate of change, and average operating temperature rate of change all have the same degree of influence on the grid's operating status, then the values of Z1, Z2, Z3, and Z4 are all 0.25.

[0080] Preferably, after the calculation of the grid operation status assessment coefficient within the monitoring period is completed, the grid operation status assessment coefficient within the monitoring period is compared with a preset operation status assessment coefficient threshold. The preset operation status assessment coefficient threshold can be determined by averaging a plurality of historical operation status assessment coefficients determined from historical operation data, or can be a value set by operation and maintenance personnel based on needs.

[0081] If the grid operation status assessment coefficient during the monitoring period is less than the preset operation status assessment coefficient threshold, it indicates that the grid operation is in a normal state, and no grid operation abnormality signal is generated at this time.

[0082] If the grid's operational status assessment coefficient during the monitoring period is greater than a preset operational status assessment coefficient threshold, or if the grid's operational status assessment coefficient during the monitoring period is equal to a preset operational status assessment coefficient threshold, it indicates that the grid is operating abnormally, and an operational abnormality signal is generated and sent to the grid management platform. The operational abnormality signal includes: a timestamp (indicating the time the abnormality occurred), location information (identifying the specific geographic location where the abnormality occurred, a specific node in the grid, or a specific line in the grid), type information (describing the abnormality type, which may include short circuit, overload, voltage sag, etc.), the severity of the abnormality, and abnormal operational data (including abnormal operating voltage, abnormal operating current, abnormal operating power, and abnormal operating temperature).

[0083] After receiving the abnormal operation signal, the power grid management platform immediately takes corresponding measures to quickly restore the operation of the power grid. The specific measures include: identifying the abnormal type according to the type information in the abnormal signal; determining the initial fault area according to the location information in the abnormal signal and the power grid topology diagram; by comparing the normal operation data with the abnormal operation data in the abnormal signal, combining the power grid topology diagram and the initial fault area, further narrowing the fault area on the basis of the initial fault area to obtain the target fault area. The fault point is located by the traveling wave method, the impedance method or the intelligent sensor. For example, the specific operation steps when locating the fault point using the impedance method include: measuring the voltage and current values after the fault occurs at both ends of the power grid in the target fault area, and calculating the apparent impedance from the measurement point to the fault point using the measured voltage and current values; adjusting the frequency, measuring the apparent impedance at different frequencies at different frequencies, and drawing the corresponding impedance trajectory on the complex plane. The resulting impedance trajectory is compared with a known grid model, which includes theoretical impedance characteristics for various grid regions under different conditions. The point closest to the theoretical model is determined through this comparison. The distance between this closest point and the measurement point is the distance between the fault point and the measurement point. The exact location of the fault point is determined based on the grid topology diagram and the distance between the fault point and the measurement point.

[0084] By monitoring the operating status of the power grid, we can promptly detect abnormalities in the operating status of the power grid, quickly respond to abnormalities to locate the fault point, and assist operation and maintenance personnel to repair the fault point, so as to reduce the time of failure and power outages and ensure the safe and stable operation of the power grid.

[0085] For step S3, in a preferred embodiment, the fault risk characteristic matrix of the power grid within the monitoring period is constructed based on the risk assessment parameters within the monitoring period, including: dividing the monitoring period into several monitoring time periods, and obtaining real-time risk assessment parameters corresponding to the several monitoring time periods; determining the operating voltage abnormality risk value based on the real-time key equipment operating voltage value of each monitoring time period; determining the phase current abnormality risk value based on the key equipment operating current value of each monitoring time period; determining the key equipment temperature abnormality risk angle based on the key equipment temperature of each monitoring time period; determining the circuit breaker operation number difference based on the circuit breaker operation number of each monitoring time period; determining the insulation resistance difference based on the insulation resistance value of each monitoring time period; constructing the fault risk characteristic matrix within the monitoring period based on the operating voltage abnormality risk value, the phase current abnormality risk value, the key equipment temperature abnormality risk angle, the circuit breaker operation number difference and the insulation resistance difference.

[0086] Specifically, the monitoring cycle is divided into several monitoring time periods, and the real-time risk assessment parameters corresponding to the monitoring time periods are obtained. It should be noted that the number of monitoring time periods and the length of monitoring time periods obtained after the equal division are consistent with the number of monitoring time periods and the length of monitoring time periods obtained by dividing the monitoring cycle corresponding to the operating data. After the monitoring cycle is divided into equal parts, the operating voltage abnormality risk value is determined according to the real-time operating voltage value of the key equipment in each monitoring period; the phase current abnormality risk value is determined according to the operating current value of the key equipment in each monitoring period; the key equipment temperature abnormality risk angle is determined according to the key equipment temperature in each monitoring period; the circuit breaker operation number difference is determined according to the circuit breaker operation number in each monitoring period; the insulation resistance difference is determined according to the insulation resistance value in each monitoring period; and the fault risk characteristic matrix within the monitoring period is constructed according to the operating voltage abnormality risk value, the phase current abnormality risk value, the key equipment temperature abnormality risk angle, the circuit breaker operation number difference and the insulation resistance difference.

[0087] In a preferred embodiment, the method for determining the abnormal operating voltage risk value according to the real-time operating voltage value of key equipment in each monitoring period includes: for each monitoring period, generating a key equipment operating voltage value characteristic curve for the current monitoring period according to the real-time operating voltage value of key equipment in the current monitoring period; determining a first area according to the key equipment operating voltage value characteristic curve for the current monitoring period and a preset key equipment operating voltage value characteristic curve; wherein the first area is the area enclosed by the line segment of the key equipment operating voltage value characteristic curve located above the preset key equipment operating voltage value characteristic curve and the preset key equipment operating voltage value characteristic curve; based on The second area is determined according to the area enclosed by the characteristic curve of the operating voltage value of the key equipment in the current monitoring period and the horizontal axis of the first rectangular coordinate system where the characteristic curve of the operating voltage value of the key equipment is located; wherein, the horizontal axis of the first rectangular coordinate system represents the monitoring time, and the vertical axis of the second rectangular coordinate system represents the operating voltage value of the key equipment; the first area is divided by the second area to obtain the operating voltage abnormality multiplier value of the current monitoring period; a first set is generated according to the operating voltage abnormality multiplier value of each monitoring period, the maximum subset and the minimum subset of the first set are obtained, and the operating voltage abnormality risk value is determined according to the average of the difference between the maximum subset of the first set and the minimum subset of the first set.

[0088] Specifically, for each monitoring period, a characteristic curve of the key equipment operating voltage value for the current monitoring period is drawn based on the real-time operating voltage value of the key equipment during the current monitoring period. The operating voltage abnormality multiplier value for the current monitoring period is obtained by dividing a first area enclosed by a line segment of the key equipment operating voltage value characteristic curve located above a preset key equipment operating voltage value characteristic curve and the preset key equipment operating voltage value characteristic curve by a second area enclosed by the key equipment operating voltage value characteristic curve for the current monitoring period and the horizontal axis of the first rectangular coordinate system in which the key equipment operating voltage value characteristic curve is located.

[0089] It should be noted that the preset key equipment operating voltage value characteristic curve is an ideal voltage change curve set according to the design specifications and safety standards of the power system. The relevant national or industry standards are consulted to understand the recommended voltage level and its allowable deviation, and a preset key equipment operating voltage value characteristic curve is drawn accordingly.

[0090] A first set is generated according to the operating voltage anomaly magnification values of all monitoring periods, the maximum subset and the minimum subset of the first set are obtained, and the average of the difference between the maximum subset of the first set and the minimum subset of the first set is used as the operating voltage anomaly risk value.

[0091] In a preferred embodiment, the method of determining the phase current abnormality risk value based on the operating current value of the key equipment in each monitoring period includes: generating a second set based on the operating current value of the key equipment in each monitoring period, obtaining the mean of the second set, and marking the mean in the second set as the phase current mean; constructing a second rectangular coordinate system; wherein the horizontal axis of the second rectangular coordinate system is the monitoring time, and the vertical axis of the second rectangular coordinate system is the phase current mean; generating a phase current mean curve in the second rectangular coordinate system according to the points marked as the phase current mean in the second set; determining a third area and a first time length based on the phase current mean curve of the current monitoring period and the preset phase current mean threshold curve, and determining the phase current abnormality risk value based on the third area and the first time length; wherein the third area is the area enclosed by the line segment of the phase current mean curve above the preset phase current mean threshold curve and the preset phase current mean threshold curve; the first time length is the time length corresponding to the line segment of the phase current mean curve above the preset phase current mean threshold curve.

[0092] Specifically, the phase current abnormality risk value is the product of the first time length normalized data represented by the time length corresponding to the line segment where the phase current mean curve is above the preset phase current mean threshold curve, and the third area normalized data enclosed by the line segment where the phase current mean curve is above the preset phase current mean threshold curve and the preset phase current mean threshold curve.

[0093] It should be noted that the preset phase current mean threshold curve is a preset phase current mean threshold curve drawn based on the power system design specifications and standards, equipment technical parameters and historical operating data. In order to enable the preset phase current mean threshold curve and the phase current mean curve to be displayed simultaneously in the second rectangular coordinate system, after the phase current mean curve is drawn, the preset phase current mean threshold curve is drawn in the second rectangular coordinate system based on the power system design specifications and standards, equipment technical parameters and historical operating data.

[0094] In a preferred embodiment, the abnormal temperature risk angle of key equipment is determined based on the temperature of key equipment in each monitoring period, including: for each monitoring period, according to the difference between the temperature of key equipment in the current monitoring period and the temperature of key equipment in the next monitoring period, the temperature change value of key equipment in the current monitoring period is obtained; a third rectangular coordinate system is constructed; wherein the horizontal axis of the third rectangular coordinate system is the monitoring time, and the vertical axis of the third rectangular coordinate system is the temperature change value of key equipment; according to the temperature change value of key equipment in each monitoring period, a key equipment temperature change value curve is generated in the third rectangular coordinate system; according to the temperature change value of key equipment in each monitoring period, the abnormal temperature risk angle of key equipment is determined according to the acute angle value formed when the key equipment temperature change value curve intersects with the preset key equipment temperature change threshold curve for the first time.

[0095] Specifically, it should be noted that the preset key equipment temperature change threshold curve is drawn based on the power system design specifications and standards, key equipment parameters and historical operating data. In order to enable the key equipment temperature change value curve and the preset key equipment temperature change threshold curve to be displayed simultaneously in the third rectangular coordinate system, after the key equipment temperature change value curve is drawn, the preset key equipment temperature change threshold curve is drawn in the third rectangular coordinate system based on the power system design specifications and standards, key equipment parameters and historical operating data.

[0096] In a preferred embodiment, the difference in the number of circuit breaker operations is determined based on the number of circuit breaker operations in each monitoring period, and the difference in insulation resistance is determined based on the insulation resistance value in each monitoring period, including: generating a third set based on the number of circuit breaker operations in each monitoring period, and determining the difference in the number of circuit breaker operations based on the average of the difference between the largest subset of the third set and the smallest subset of the third set; generating a fourth set based on the insulation resistance value in each monitoring period, and determining the difference in insulation resistance based on the average of the difference between the largest subset of the fourth set and the smallest subset of the fourth set.

[0097] After obtaining the operating voltage abnormality risk value, phase current abnormality risk value, key equipment temperature abnormality risk angle, circuit breaker operation times difference and insulation resistance difference during the monitoring period, the operating voltage abnormality risk value, phase current abnormality risk value, key equipment temperature abnormality risk angle, circuit breaker operation times difference and insulation resistance difference are combined to construct a fault risk feature matrix during the monitoring period.

[0098] In a preferred embodiment, determining the fault probability value of the power grid within the monitoring period according to the fault risk characteristic matrix within the monitoring period includes: inputting the fault risk characteristic matrix within the monitoring period into a logistic regression model, so that the logistic regression model outputs the fault probability value of the power grid within the monitoring period;

[0099] Among them, the construction of the logistic regression model includes: obtaining several historical risk assessment parameters and the failure probability value labels corresponding to each historical risk assessment parameter; for each historical risk assessment parameter, generating a failure risk feature matrix sample of the current historical risk assessment parameter based on the historical risk assessment parameter; constructing an initial logistic regression model, taking the failure risk feature matrix sample as input and the failure probability value label corresponding to the failure risk feature matrix sample as output, training the initial logistic regression model until the initial logistic regression model reaches a preset convergence condition, and then generating the logistic regression model.

[0100] Specifically, the fault risk feature matrix within the monitoring period is input into the logistic regression model. The model processes the abnormal operating voltage risk value, phase current risk value, critical equipment temperature abnormality risk angle, circuit breaker operation count difference, and insulation resistance difference within the fault risk feature matrix within the monitoring period. The model then outputs a fault probability value for the power grid within the monitoring period. This probability value ranges from 0 to 1, with 0 indicating no fault risk and 1 indicating a fault risk. Once the fault probability value for the power grid within the monitoring period is obtained, it is sent to the power grid management platform.

[0101] The construction of the logistic regression model includes the following steps:

[0102] 1. Obtain several historical risk assessment parameters and the corresponding failure probability value labels for each historical risk assessment parameter. The failure probability label is a value between 0 and 1, where 0 indicates no failure risk and 1 indicates a failure risk.

[0103] 2. For each historical risk assessment parameter, generate the operating voltage abnormality risk value sample, phase current abnormality risk value sample, key equipment temperature abnormality risk angle sample, circuit breaker operation number difference sample, and insulation resistance difference sample for the current historical risk assessment parameter based on the historical risk assessment parameter. Based on the operating voltage abnormality risk value sample, phase current abnormality risk value sample, key equipment temperature abnormality risk angle sample, circuit breaker operation number difference sample, and insulation resistance difference sample for the current historical risk assessment parameter, construct the fault risk feature matrix sample for the current historical risk assessment parameter. This step yields several fault risk feature matrix samples corresponding to the historical risk assessment parameters. Combine the fault risk feature matrix sample corresponding to each historical risk assessment parameter with the corresponding fault probability value label to obtain several training samples.

[0104] 3. Construct an initial logistic regression model, take the fault risk feature matrix samples as input, take the fault probability value labels corresponding to the fault risk feature matrix samples as output, train the initial logistic regression model until the initial logistic regression model reaches the preset convergence condition (when the prediction error is minimized or the maximum number of iterations is reached), and generate the logistic regression model.

[0105] Preferably, the expression formula of the logistic regression model is:

[0106]

[0107] Among them, P(w=1) represents the failure probability value of the power grid; DYY represents the risk value of abnormal operating voltage; XLY represents the risk value of abnormal phase current; WYJ represents the risk angle of abnormal temperature of key equipment; DCC represents the difference in the number of circuit breaker operations; JDC represents the difference in insulation resistance; v0 represents the intercept term, which represents the baseline probability of power grid failure; v1, v2, v3, v4, and v5 are the regression coefficients of DYY, XLY, WYJ, DCC, and JDC, respectively.

[0108] By constructing a fault risk feature matrix and combining it with a logistic regression model to efficiently predict the fault probability value, manual intervention can be reduced and the accuracy of the fault probability value prediction can be improved.

[0109] In step S4, after obtaining the operation status evaluation coefficient of the power grid during the monitoring period and the fault probability value of the power grid during the monitoring period, the power grid operation stability evaluation coefficient during the monitoring period is calculated using a preset power grid operation stability evaluation coefficient formula; wherein the preset power grid operation stability evaluation coefficient formula is:

[0110]

[0111] Among them, WPX represents the grid operation stability assessment coefficient during the monitoring period; P real time represents the fault probability value of the grid during the monitoring period; ZPX represents the grid operation status assessment coefficient during the monitoring period; g1 and g2 are both preset proportional factor coefficients, g2>g1>0.

[0112] The grid operation stability assessment coefficient during the monitoring period is compared with a preset first operation stability assessment coefficient threshold and a preset second operation stability assessment coefficient threshold; wherein the preset first operation stability assessment coefficient threshold is less than the preset second operation stability assessment coefficient threshold.

[0113] If the grid operation stability assessment coefficient is less than the preset first operation stability assessment coefficient threshold value during the monitoring period, it indicates that the grid operation stability is good, and a good stability signal is generated and sent to the grid management platform;

[0114] If the grid operation stability assessment coefficient during the monitoring period is greater than the preset first operation stability assessment coefficient threshold, and the grid operation stability assessment coefficient during the monitoring period is less than the preset second operation stability assessment coefficient threshold, it indicates that there is a risk of grid operation becoming unstable, and a general stability signal is generated and sent to the grid management platform;

[0115] If the grid operation stability assessment coefficient during the monitoring period is equal to the preset first operation stability assessment coefficient threshold, and the grid operation stability assessment coefficient during the monitoring period is less than the preset second operation stability assessment coefficient threshold, it indicates that there is a risk of grid operation becoming unstable, and a general stability signal is generated and sent to the grid management platform;

[0116] If the grid operation stability assessment coefficient is greater than the preset second operation stability assessment coefficient threshold value during the monitoring period, it indicates that the grid operation is in an unstable state, and a poor stability signal is generated and sent to the grid management platform;

[0117] If the grid operation stability assessment coefficient is equal to the preset second operation stability assessment coefficient threshold value during the monitoring period, it indicates that the grid operation is in an unstable state, and a poor stability signal is generated and sent to the grid management platform;

[0118] The power grid management platform generates power grid security warning information based on the received good stability signal, general stability signal or poor stability signal, and sends the power grid security warning information to the operation and maintenance personnel to assist the power grid security operation and maintenance personnel in performing corresponding maintenance processing.

[0119] Preferably, when the power grid security warning information corresponds to a good stability signal, a first-level warning notification is sent to the operation and maintenance team, and corresponding processing measures are taken to respond to the current first-level warning notification; when the power grid security warning information corresponds to a general stability signal, a second-level warning notification is sent to the operation and maintenance team and relevant responsible personnel, and corresponding processing measures are taken to respond to the current warning notification; when the power grid security warning information corresponds to a poor stability signal, a third-level warning notification is sent to the operation and maintenance team, relevant responsible personnel and emergency response team, and corresponding processing measures are taken to respond to the current warning notification.

[0120] The specific measures for responding to a Level 1 warning notification are as follows: Maintain regular monitoring frequency and data accuracy; implement established maintenance plans; reasonably distribute loads to prevent overloads; and regularly update emergency response plans. The specific measures for responding to a Level 2 warning notification are as follows: Increase monitoring of key parameters; immediately inspect critical equipment and repair potential problems; perform preventive maintenance as needed and replace aging components; adjust loads to avoid overloads and reduce circuit breaker operations; and ensure the stability of the SCADA system and communication links. The specific measures for responding to a Level 3 warning notification are as follows: Organize an emergency response team to prevent the fault from expanding; urgently repair affected equipment or lines; comprehensively inspect the system and resolve all potential problems; promptly notify users and provide temporary or alternative solutions; activate backup power supplies to protect important loads; analyze the cause of the fault and develop long-term improvement measures.

[0121] Through a graded early warning mechanism, the response scope is gradually expanded to ensure that there are sufficient resources and personnel to respond at different risk levels, send early warning notifications in a timely manner and take corresponding measures to effectively reduce the impact of abnormal situations on the power grid and improve the reliability and safety of power grid operation.

[0122] Based on the above method embodiments, the present invention provides corresponding system embodiments.

[0123] like Figure 2 As shown, an embodiment of the present invention provides a power grid security early warning analysis system, comprising: a real-time data acquisition module, an operation status assessment coefficient determination module, a fault probability value determination module and an early warning information generation module;

[0124] The real-time data acquisition module is used to obtain real-time operating data and real-time risk assessment parameters of the power grid during the monitoring period; wherein the operating data includes: operating voltage, operating current, operating power and operating temperature; the risk assessment parameters include: operating voltage value of key equipment, operating current value of key equipment, temperature value of key equipment, number of circuit breaker operations and insulation resistance value;

[0125] The operating status assessment coefficient determination module is used to calculate the average operating voltage change rate, the average operating current change rate, the average operating power change rate and the average operating temperature change rate within the monitoring period based on the real-time operating data within the monitoring period, and calculate the operating status assessment coefficient of the power grid within the monitoring period based on the average operating voltage change rate, the average operating current change rate, the average operating power change rate and the average operating temperature change rate within the monitoring period;

[0126] The fault probability value determination module is used to construct a fault risk characteristic matrix of the power grid within the monitoring period according to the risk assessment parameters within the monitoring period, and determine the fault probability value of the power grid within the monitoring period according to the fault risk characteristic matrix within the monitoring period;

[0127] The warning information generation module is used to determine the grid operation stability assessment coefficient within the monitoring period based on the grid operation status assessment coefficient and the grid failure probability value within the monitoring period, and generate grid security warning information based on the grid operation stability assessment coefficient.

[0128] It should be noted that the system embodiment described above is merely illustrative, wherein the units described as separate components may or may not be physically separate, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed across multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the present embodiment. In addition, in the drawings of the system embodiment provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, which can be specifically implemented as one or more communication buses or signal lines. A person of ordinary skill in the art can understand and implement the present invention without inventive work.

[0129] Those skilled in the art will clearly understand that for the sake of convenience and brevity, the specific working process of the system described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0130] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A power grid security early warning analysis method, characterized in that: include: Acquire real-time operating data and real-time risk assessment parameters of the power grid during the monitoring period; wherein the operating data includes: operating voltage, operating current, operating power, and operating temperature; and the risk assessment parameters include: operating voltage value of key equipment, operating current value of key equipment, temperature value of key equipment, number of circuit breaker operations, and insulation resistance value; Calculate the average operating voltage change rate, average operating current change rate, average operating power change rate, and average operating temperature change rate during the monitoring period based on the real-time operating data during the monitoring period, and calculate the operating status assessment coefficient of the power grid during the monitoring period based on the average operating voltage change rate, average operating current change rate, average operating power change rate, and average operating temperature change rate during the monitoring period; Constructing a fault risk characteristic matrix of the power grid within the monitoring period according to the risk assessment parameters within the monitoring period, and determining a fault probability value of the power grid within the monitoring period according to the fault risk characteristic matrix within the monitoring period; According to the operation status evaluation coefficient of the power grid during the monitoring period and the fault probability value of the power grid during the monitoring period, the power grid operation stability evaluation coefficient during the monitoring period is determined, and the power grid safety warning information is generated according to the power grid operation stability evaluation coefficient.

2. A power grid security early warning analysis method according to claim 1, characterized in that: The calculating, based on the real-time operating data within the monitoring period, the average operating voltage change rate, the average operating current change rate, the average operating power change rate, and the average operating temperature change rate within the monitoring period, includes: Divide the monitoring period into several monitoring time periods, and obtain real-time operation data corresponding to the several monitoring time periods; For each monitoring period, the operating voltage change rate within the current monitoring period is determined based on the duration of the current monitoring period and the real-time operating voltage; the operating current change rate within the current monitoring period is determined based on the duration of the current monitoring period and the real-time operating current; the operating power change rate within the current monitoring period is determined based on the duration of the current monitoring period and the real-time operating power; and the operating temperature change rate is determined based on the duration of the current monitoring period and the real-time operating temperature. The average operating voltage change rate is calculated based on the operating voltage change rate in each monitoring period, the average operating current change rate is calculated based on the operating current change rate in each monitoring period, the average operating power change rate is calculated based on the operating power change rate in each monitoring period, and the average operating temperature change rate is calculated based on the operating temperature change rate in each monitoring period.

3. A power grid security early warning analysis method according to claim 2, characterized in that: The calculation of the operating status assessment coefficient of the power grid during the monitoring period based on the average operating voltage change rate, the average operating current change rate, the average operating power change rate, and the average operating temperature change rate during the monitoring period includes: The operating status assessment coefficient of the power grid during the monitoring period is calculated by using the average operating voltage change rate, average operating current change rate, average operating power change rate, average operating temperature change rate and the operating status assessment coefficient calculation formula during the monitoring period; The calculation formula of the operating status assessment coefficient is specifically: ZPX=Z1*PDY+Z2*PDL+Z3*PGL+Z4*PWD; Among them, ZPX represents the operating status assessment coefficient of the power grid during the monitoring period; PDY represents the average operating voltage change rate; PDL represents the average operating current change rate; PGL represents the average operating power change rate; PWD represents the average operating temperature change rate; Z1, Z2, Z3 and Z4 are the weight coefficients of PDY, PDL, PGL and PWD respectively.

4. A power grid security early warning analysis method according to claim 3, characterized in that: The method of constructing a fault risk characteristic matrix of the power grid within the monitoring period according to the risk assessment parameters within the monitoring period includes: Divide the monitoring period into several monitoring periods, and obtain real-time risk assessment parameters corresponding to the monitoring periods; Determine the abnormal operating voltage risk value based on the real-time operating voltage value of key equipment in each monitoring period; Determine the phase current abnormality risk value based on the operating current value of key equipment in each monitoring period; Determine the abnormal temperature risk angle of key equipment based on the temperature of key equipment in each monitoring period; determining a circuit breaker operation number difference according to the circuit breaker operation number in each monitoring period; Determine the insulation resistance difference based on the insulation resistance value of each monitoring period; Based on the abnormal operating voltage risk value, phase current risk value, key equipment temperature abnormality risk angle, circuit breaker operation times difference and insulation resistance difference, a fault risk feature matrix within the monitoring period is constructed.

5. A power grid security early warning analysis method according to claim 4, characterized in that: Determining the abnormal operating voltage risk value based on the real-time operating voltage value of key equipment in each monitoring period includes: For each monitoring period, a characteristic curve of the key equipment operating voltage value of the current monitoring period is generated based on the real-time key equipment operating voltage value of the current monitoring period; Determine a first area based on the key equipment operating voltage value characteristic curve and the preset key equipment operating voltage value characteristic curve during the current monitoring period; wherein the first area is the area enclosed by a line segment of the key equipment operating voltage value characteristic curve located above the preset key equipment operating voltage value characteristic curve and the preset key equipment operating voltage value characteristic curve; Determine the second area based on the area enclosed by the characteristic curve of the key equipment operating voltage value during the current monitoring period and the horizontal axis of the first rectangular coordinate system where the characteristic curve of the key equipment operating voltage value is located; wherein the horizontal axis of the first rectangular coordinate system represents the monitoring time, and the vertical axis of the second rectangular coordinate system represents the key equipment operating voltage value; Divide the first area by the second area to obtain the operating voltage abnormality multiplier value during the current monitoring period; A first set is generated according to the operating voltage anomaly multiplier value of each monitoring period, the maximum subset and the minimum subset of the first set are obtained, and the operating voltage anomaly risk value is determined according to the average of the difference between the maximum subset of the first set and the minimum subset of the first set.

6. A power grid security early warning analysis method according to claim 5, characterized in that: Determining the phase current abnormality risk value based on the operating current value of the key equipment in each monitoring period includes: Generating a second set according to the operating current values of the key equipment in each monitoring period, obtaining a mean value of the second set, and marking the mean value in the second set as a phase current mean value; Constructing a second rectangular coordinate system; wherein the horizontal axis of the second rectangular coordinate system is the monitoring time, and the vertical axis of the second rectangular coordinate system is the phase current average; generating a phase current mean curve in a second rectangular coordinate system according to points marked as phase current mean values in the second set; The third area and the first time length are determined based on the phase current mean curve and the preset phase current mean threshold curve of the current monitoring period, and the phase current abnormality risk value is determined based on the third area and the first time length; wherein, the third area is the area enclosed by the line segment of the phase current mean curve above the preset phase current mean threshold curve and the preset phase current mean threshold curve; the first time length is the time length corresponding to the line segment of the phase current mean curve above the preset phase current mean threshold curve.

7. A power grid security early warning analysis method according to claim 6, characterized in that: Determine the abnormal temperature risk angle of key equipment based on the temperature of key equipment in each monitoring period, including: For each monitoring period, the temperature change value of the key equipment in the current monitoring period is obtained according to the difference between the temperature of the key equipment in the current monitoring period and the temperature of the key equipment in the next monitoring period; Construct a third rectangular coordinate system; wherein the horizontal axis of the third rectangular coordinate system is the monitoring time, and the vertical axis of the third rectangular coordinate system is the temperature change value of the key equipment; A key equipment temperature change value curve is generated in the third rectangular coordinate system based on the temperature change value of the key equipment in each monitoring period; the key equipment temperature abnormality risk angle is determined based on the acute angle value formed when the key equipment temperature change value curve intersects with the preset key equipment temperature change threshold curve for the first time.

8. A power grid security early warning analysis method according to claim 7, characterized in that: Determining a circuit breaker operation count difference based on the circuit breaker operation count in each monitoring period, and determining an insulation resistance difference based on the insulation resistance value in each monitoring period, including: generating a third set according to the number of circuit breaker operations in each monitoring period, and determining a difference in the number of circuit breaker operations according to an average of differences between a maximum subset of the third set and a minimum subset of the third set; A fourth set is generated according to the insulation resistance values of each monitoring period, and an insulation resistance difference is determined according to an average of the differences between the largest subset of the fourth set and the smallest subset of the fourth set.

9. A power grid security early warning analysis method according to claim 8, characterized in that: Determining the fault probability value of the power grid within the monitoring period according to the fault risk characteristic matrix within the monitoring period includes: Inputting the fault risk feature matrix within the monitoring period into the logistic regression model so that the logistic regression model outputs a fault probability value of the power grid within the monitoring period; The construction of the logistic regression model includes: Obtaining several historical risk assessment parameters and failure probability value labels corresponding to each historical risk assessment parameter; For each historical risk assessment parameter, a fault risk feature matrix sample of the current historical risk assessment parameter is generated according to the historical risk assessment parameter; An initial logistic regression model is constructed, with the fault risk feature matrix samples as input and the fault probability value labels corresponding to the fault risk feature matrix samples as output. The initial logistic regression model is trained until the initial logistic regression model reaches a preset convergence condition, thereby generating the logistic regression model.

10. A power grid security early warning analysis system, characterized in that: include: Real-time data acquisition module, operation status assessment coefficient determination module, fault probability value determination module and warning information generation module; The real-time data acquisition module is used to obtain real-time operating data and real-time risk assessment parameters of the power grid during the monitoring period; wherein the operating data includes: operating voltage, operating current, operating power and operating temperature; the risk assessment parameters include: operating voltage value of key equipment, operating current value of key equipment, temperature value of key equipment, number of circuit breaker operations and insulation resistance value; The operating status assessment coefficient determination module is used to calculate the average operating voltage change rate, the average operating current change rate, the average operating power change rate and the average operating temperature change rate within the monitoring period based on the real-time operating data within the monitoring period, and calculate the operating status assessment coefficient of the power grid within the monitoring period based on the average operating voltage change rate, the average operating current change rate, the average operating power change rate and the average operating temperature change rate within the monitoring period; The fault probability value determination module is used to construct a fault risk characteristic matrix of the power grid within the monitoring period according to the risk assessment parameters within the monitoring period, and determine the fault probability value of the power grid within the monitoring period according to the fault risk characteristic matrix within the monitoring period; The warning information generation module is used to determine the grid operation stability assessment coefficient within the monitoring period based on the grid operation status assessment coefficient and the grid failure probability value within the monitoring period, and generate grid security warning information based on the grid operation stability assessment coefficient.

Images