Electric power safety monitoring management method

Abstract

The invention discloses an electric power safety monitoring management method, and relates to the technical field of electric power monitoring, and the method comprises the steps: obtaining a basic data set, calculating the dependency weight between devices, obtaining an associated device set, and constructing an initial network diagram according to the associated device set; determining an abnormal device and a corresponding abnormal time node; obtaining a first association group and a second association group according to the association device set; overlapping the first association group and the second association group with the initial network diagram to obtain a main path and an auxiliary path; calculating the risk intensity value of each device in the main path, obtaining the disposal efficiency and evaluation indexes of different time nodes, and obtaining the optimal intervention time; calculating a risk delay growth curve of the secondary path, identifying a risk acceleration inflection point, and determining the optimal prevention time; by quantifying electrical dependence and physical distance, the limitation that a traditional topology network only pays attention to physical connection is broken through, a fault conduction key path is accurately locked, and cross-device chain reaction is quantified.

Description

Technical Field

[0001] The present invention relates to the technical field of power monitoring, and in particular to a power safety monitoring and management method. Background Art

[0002] Substations are a crucial component of power systems. Traditional methods for substation safety monitoring and management rely primarily on manual inspections and scheduled maintenance, resulting in incomplete equipment monitoring and delayed safety risk warnings. In particular, current technology fails to adequately consider the risks associated with inter-device interactions. When a device fails, it's often impossible to prevent the fault from spreading to connected devices, leading to a wider range of failures and compromising the overall safety of the substation.

[0003] Existing fault analysis and assessments often focus on a single device, failing to fully consider the cascading impact of device failures on other devices. Furthermore, static network topologies make it difficult to dynamically adapt to device damage or line reconfiguration. Furthermore, the delayed onset of faults over time is ignored, further limiting the effectiveness and timeliness of fault resolution.

[0004] Therefore, developing a power safety monitoring and management method that can comprehensively monitor equipment, accurately warn of risks, dynamically adapt to changes, and consider spatiotemporal continuity has become an urgent need to improve substation safety. Summary of the Invention

[0005] This application solves the problems of insufficient substation equipment monitoring and low safety in the existing technology by providing an electric power safety monitoring and management method, and fully analyzes the potential correlations between multiple devices to achieve comprehensive monitoring, timely warning and fault elimination.

[0006] This application provides a method for monitoring and managing power safety, the method comprising: S1: Obtain the basic data set, calculate the dependency weights between devices and obtain the associated device set, and build the initial network graph based on the associated device set; S2: Monitor device status in real time, identify abnormal devices and corresponding abnormal time nodes; obtain a first association group and a second association group based on the associated device set; overlap the first association group and the second association group with the initial network diagram to obtain a primary path and a secondary path; S3: Calculate the risk intensity value of each device in the primary path, obtain the disposal efficiency and evaluation indicators at different time nodes, and determine the optimal intervention time. Calculate the risk delay growth curve of the secondary path, identify the risk growth inflection point, and determine the optimal prevention time. S4: If the risk intensity value is not greater than the high threshold, between the optimal intervention time and the optimal prevention time, identify alternative devices based on the status and alternative indicators of the abnormal device, and construct a complementary characteristic matrix to determine the alternative transmission path.

[0007] Furthermore, the method further includes: S5: obtaining historical intervention data records, generating a high-dimensional state vector and defining a set of intervention actions, and constructing a digital twin model; based on the historical intervention data records and the digital twin model, simulating the response process under different fault scenarios, and introducing random disturbances, and obtaining the optimal intervention actions and corresponding intervention times under different abnormal states according to the reward function indicator; S6: Obtain the target area and several substations within the target area, and determine a central site; obtain real-time data of adjacent sites, determine the cross-site dependency weight based on the parameters of the adjacent station interconnection lines, and divide the collaborative defense area; when abnormal equipment is detected, identify the abnormal state, obtain the optimal intervention action and corresponding intervention time of the central site according to step S5, and determine the cross-site propagation probability and resource requirements of the central site, determine the global risk constraints based on the collaborative defense area, and generate a global management plan.

[0008] Furthermore, the basic data set includes electrical parameters, physical locations and control logic of the power equipment; The dependency weight is a parameter value used to quantify the energy transfer dependency between devices; According to the associated device set, obtain specific device nodes, establish associated edges, and mark dependency weights to form the initial network graph; The associated device set includes a direct association set and an indirect association set; according to the direct association set, devices associated with the abnormal device are obtained to form a first association group; according to the indirect association set, devices associated with the abnormal device are obtained to form a second association group.

[0009] Furthermore, the first association group and the second association group are overlapped with the initial network diagram to obtain a main path and a secondary path, respectively, including: determining the specific location of the abnormal device in the initial network diagram, extracting the device nodes in the first association group, and connecting the device nodes to form a main path; extracting the device nodes in the second association group, and connecting the device nodes to form a secondary path.

[0010] Furthermore, the risk intensity value refers to the dynamic fault risk level of the device under specific time and space coordinates; the specific time and space coordinates are the spatial coordinates and time axis coordinates of the device, the spatial coordinates are used to locate the physical location of the fault propagation, and the time axis coordinates are used to quantify the dynamic evolution process of the fault over time; Based on the current waveform data of abnormal equipment obtained based on the data set, the fault energy accumulation is calculated, and evaluation indicators are set to evaluate and verify the disposal efficiency at different time nodes to obtain the optimal intervention time; the fault energy accumulation refers to the fault energy generated over time, which is used to measure the fault destructive capacity; the evaluation indicators include the energy blocking rate and the energy consumption reduction rate per unit time.

[0011] Furthermore, the risk delay growth curve of the secondary path is calculated, the risk growth inflection point is identified, and the optimal prevention time is determined, including: obtaining the device nodes in the secondary path and the corresponding initial dependency weights, calculating the risk secondary value corresponding to the time change of each device node, and generating the risk delay growth curve based on the risk secondary value and the time node; calculating the risk change rate based on the risk secondary value, identifying the risk growth inflection point, and determining the optimal prevention time; The initial value of the risk sub-value of each device is 0, and the risk sub-value at the corresponding moment is calculated based on the initial dependency weight, environmental acceleration factor and line degradation rate.

[0012] Furthermore, the high threshold is pre-set and is used to measure the abnormality degree of the abnormal device; The replacement indicators include functional compensation degree, capacity matching rate and switching delay. The functional compensation degree is a measure of the percentage of the replacement equipment that functionally covers the key capabilities of the abnormal equipment; the capacity matching rate is the adaptation ratio of the replacement equipment capacity to the abnormal equipment requirements; the switching delay is the time required to switch from the abnormal equipment to the replacement equipment, including mechanical action time, grid synchronization time and system stabilization time.

[0013] Furthermore, a complementary characteristic matrix is ​​constructed according to the state of the abnormal device, including: the complementary characteristic matrix is ​​a three-dimensional matrix including candidate replacement devices, abnormal state values, and comprehensive values; the abnormal state of the abnormal device is analyzed to determine the function loss rate, capacity gap, and risk diffusion rate, and a weighted sum is performed to obtain the abnormal state value; The comprehensive value is the difference between the substitution index value and the abnormal state value of each candidate replacement device. The larger the difference, the greater the replacement capability of the candidate replacement device. Determining an alternative transmission path includes: determining a number of candidate paths according to a complementary characteristic matrix, determining an initial alternative device for each candidate path according to a connection relationship between devices, retaining candidate paths corresponding to initial alternative devices having a dependency weight greater than 0.6 for the abnormal device, arranging the dependency weights of subsequent alternative devices in descending order, and selecting them in sequence to form an alternative transmission path.

[0014] Furthermore, the historical intervention data record includes the fault type, occurrence time, intervention action, loss result and time tag. The time tag includes the theoretical optimal intervention time and the actual optimal time. The actual optimal time is the time obtained through expert analysis after the abnormality occurs. The high-dimensional state vector includes device status, risk assessment indicators and environmental parameters; the risk assessment indicators include the risk intensity value change trend, the main path propagation speed and the delay change speed of the secondary path; The intervention action set includes early intervention, delayed intervention and action granularity. The action granularity is a fixed time step, and the specific number of delay or advance steps is intelligently selected according to the discretization of the time step; the reward function indicators include disaster reduction benefits, operating costs and system stability; the random disturbances include fluctuations in new energy output, sudden load changes and sensor noise.

[0015] Furthermore, based on the basic data set of each substation in the target area, a central site and its defense radius are determined; based on the capacity of the central site, the coordination range is determined to form a defense radius; The parameters of the adjacent station tie lines include the voltage level ratio and the power transmission ratio. The collaborative defense area is divided according to the cross-station dependency weight and the minimum defense radius of the remaining stations. A clustering method is used, and the optimization goal is to maximize the connection density within the area and minimize the connection between areas. The collaborative defense area includes a strong coupling area, a medium coupling area, and a weak coupling area. The cross-site transmission probability is the rate of change of the risk intensity value after the central site intervention action within a unit time. The unit time is dynamically set according to the range of the target area, and its value is less than the optimal prevention time of the central site.

[0016] One or more technical solutions provided in this application have at least the following technical effects or advantages: By quantifying electrical dependencies and physical distances, the system breaks through the limitations of traditional topology networks that only focus on physical connections, accurately identifies the critical paths for fault transmission, superimposes dynamic risk intensity values ​​on multi-device risks, quantifies cross-device chain reactions, and resolves the problem of misjudgment of remote device risks. It also predicts the optimal intervention time to address the defect that static risk models cannot handle delayed faults. It identifies the inflection point of the risk growth rate of secondary paths, providing a final time window for preventive maintenance and avoiding secondary faults. By calculating and determining alternative transmission paths between the optimal intervention time and the optimal prevention time, it protects the current abnormal equipment, further reduces the occurrence of faults, and prevents the secondary propagation of faults. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 The figure is a flow chart of a method for monitoring and managing power safety in an embodiment of the present invention. DETAILED DESCRIPTION

[0018] To facilitate understanding of the present invention, the present application will be described more comprehensively below with reference to the relevant drawings; the drawings show preferred embodiments of the present invention, but the present invention can be implemented in many different forms and is not limited to the embodiments described herein; on the contrary, the purpose of providing these embodiments is to enable a more thorough and comprehensive understanding of the disclosed content of the present invention.

[0019] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this invention pertains; the terms used herein in the specification of the present invention are for the purpose of describing specific embodiments only and are not intended to limit the present invention; the term "and / or" used herein includes any and all combinations of one or more of the associated listed items.

[0020] Example 1: Figure 1 As shown, a power safety monitoring and management method includes: S1: Obtain the basic data set, calculate the dependency weights between devices and obtain the associated device set, and build the initial network graph based on the associated device set.

[0021] In some embodiments, the basic data set includes the electrical parameters, physical location, and control logic of the power equipment. The electrical parameters include the equipment's standard voltage, current, resistance, and other basic electrical quantities and their related characteristic parameters. The physical location is the specific location coordinates of the equipment's actual location. The control logic refers to the linkage mechanism between devices implemented through preset rules and is the core decision-making process for preventing fault propagation. For example, when device A is abnormal, the control logic automatically triggers the preset actions of devices B / C, forming a fault isolation barrier to prevent fault propagation. When the transformer temperature exceeds the limit, the upstream circuit breaker is immediately tripped and the automatic switching of the downstream capacitor is locked (to prevent overvoltage from breaking down downstream equipment). Preferably, when acquiring the real-time status of the equipment, dual-mode sensors deployed at key nodes such as transformers and circuit breakers are used to capture microsecond-level current mutations and mechanical deformations.

[0022] In some embodiments, the dependency weight is a parameter value used to quantify the energy transfer dependency between devices. Calculating the dependency weight between any two devices requires weighted summation based on the electrical dependency and the distance attenuation factor. The calculation formula is as follows:

[0023] in, is the dependency weight, Electrical dependence refers to the ratio of the input current of the downstream device to the output current of the upstream device, reflecting the dominant relationship of energy flow; It is the distance attenuation factor, which is calculated based on the actual distance between devices and is used to measure the impact of distance on the dependence between devices.

[0024] In some embodiments, the electrical dependence is between [0.1], and a larger electrical dependence indicates a closer energy transfer dependency. The calculation formula is:

[0025] in, is the operating current of the downstream equipment (such as the circuit breaker coil current), is the output current of the upstream device (such as the secondary current of the transformer). The distance attenuation factor is calculated as follows:

[0026] in, is the distance decay factor; is the actual distance between devices, measured in meters. A weighted sum of the electrical dependency and the actual distance between devices can address scenarios with strong electrical dependency but long physical distance. For example, if the electrical dependency between a transformer and a remote circuit breaker is 0.9 but the distance is 20 meters, the corresponding distance attenuation factor is 0.018, resulting in a dependency weight of 0.635, corresponding to a medium risk level.

[0027] In some embodiments, a threshold range [0.3, 0.8] is pre-set based on historical experimental data. If the dependency weight is less than 0.3, it indicates weak dependency, the downstream equipment can operate independently, and its cascading failure risk level is low risk; if the dependency weight is within the threshold range, it indicates moderate dependency, upstream failure causes partial failure of the downstream, and its cascading failure risk level is medium risk; if the dependency weight is greater than 0.8, it indicates heavy dependency, the downstream equipment is heavily dependent on upstream power supply, upstream failure will cause downstream paralysis, and its cascading failure risk level is high risk.

[0028] In this example, a device is used as the central device. Devices related to the central device are added to the associated device set, with each device corresponding to one associated device set. Specifically, devices with moderate and heavy dependencies are added to the directly associated device set, while devices with weak dependencies are added to the indirectly associated device set. Based on the associated device set, specific device nodes are obtained, associated edges are established, and dependency weights are labeled to form the initial network graph.

[0029] In some embodiments, devices in a directly associated set have close dependencies. If a device experiences an anomaly, directly associated devices are likely to be quickly affected, resulting in a higher probability of failure. Devices in an indirectly associated set, on the other hand, have relatively weaker dependencies, resulting in a certain buffer and delay in the impact of a single device anomaly. By distinguishing these two sets, the propagation boundaries of faults across devices with varying degrees of dependency can be clearly defined, allowing for quick and accurate determination of the range of devices potentially affected by a fault, and clarifying the direction for fault resolution and resource allocation. In practical applications, based on the direct and indirect associated sets, differentiated monitoring frequencies and maintenance plans can be developed for devices in different sets. For devices in the directly associated set, due to their high risk and rapid impact from failures, more frequent real-time monitoring is required to promptly identify potential issues and implement measures. For devices in the indirectly associated set, the monitoring frequency can be reduced, but continued attention is required to ensure timely detection of potential faults that threaten to spread. This allows for the rational allocation of monitoring resources, improved overall monitoring efficiency, and reduced operational and maintenance costs.

[0030] In this embodiment, the dependence weight value is obtained based on the calculation of electrical dependence and physical distance, which breaks through the limitation of traditional topology network that only focuses on physical connection. By quantifying the dominant relationship of energy flow, the key path of fault conduction is accurately locked, and chain connection is achieved, thereby further improving the fault blocking efficiency.

[0031] S2: Monitor the device status in real time to determine abnormal devices and corresponding abnormal time nodes; obtain a first association group and a second association group based on the associated device set; and overlap the first association group and the second association group with the initial network diagram to obtain a primary path and a secondary path.

[0032] The associated device set includes a direct association set and an indirect association set; according to the direct association set, devices associated with the abnormal device are obtained to form a first association group; according to the indirect association set, devices associated with the abnormal device are obtained to form a second association group.

[0033] In some embodiments, the device status includes normal and abnormal states. Abnormal refers to the discovery of an abnormal state, which is not a fault state. The abnormal device and the corresponding abnormal time node are determined. The abnormal time node here is the time node when the abnormality is discovered. In actual situations, the actual time node of the abnormality and the time node of the abnormality discovery are not the same. It depends on the frequency and monitoring accuracy of the monitoring equipment. In subsequent calculations, the actual time node of the abnormality should be reversely inferred based on the data indicators of the abnormality discovery. When the accuracy of the monitoring equipment is guaranteed, the time difference can be ignored. This application does not impose specific restrictions on this.

[0034] In some embodiments, overlaying the first association group and the second association group with the initial network diagram to obtain a primary path and a secondary path includes: determining the specific location of the abnormal device in the initial network diagram, extracting device nodes from the first association group, and connecting the device nodes to form the primary path; and extracting device nodes from the second association group, and connecting the device nodes to form the secondary path. The device nodes refer to corresponding devices, and one device node corresponds to one device.

[0035] S3: Calculate the risk intensity value of each device in the main path, obtain the disposal efficiency and evaluation indicators at different time nodes, and obtain the optimal intervention time; calculate the risk delay growth curve of the secondary path, identify the risk growth rate inflection point, and determine the optimal prevention time.

[0036] The risk intensity value refers to the dynamic failure risk level of the equipment at a specific time and space coordinate, and is calculated as follows:

[0037] in, is the risk intensity value; is the dependency weight of primary path device i; is the propagation attenuation coefficient, which is set to 0.05 for high-voltage equipment and 0.02 for low-voltage equipment; is the current time Abnormal time node with abnormal equipment time difference; is the total distance attenuation, is the distance between device i and the abnormal device; the risk intensity value uses the superposition of multiple device risk values ​​to reflect the chain conduction characteristics of the fault along the electrical path and quantify the dynamic evolution of the fault in the space-time continuum.

[0038] The specific space-time coordinates are the spatial coordinates and time axis coordinates of the equipment. The spatial coordinates are used to locate the physical location of the fault propagation, calculate the total distance attenuation, determine the fault impact range based on the characteristics of exponential attenuation with distance, and then accurately define the isolation boundary; the time axis coordinates are used to quantify the dynamic evolution process of the fault over time. The risk intensity value changes non-uniformly over time. Based on the time axis coordinates, the optimal intervention time can be quantitatively predicted to obtain the highest disposal efficiency.

[0039] In some embodiments, the treatment efficiency and evaluation indicators at different time nodes are obtained to obtain the optimal intervention time. Specifically, based on the current waveform data of the abnormal device obtained based on the data set, the fault energy accumulation is calculated, and the evaluation indicators are set to evaluate and verify the treatment efficiency at different time nodes to obtain the optimal intervention time. The fault energy accumulation refers to the fault energy generated over time and is used to measure the damage capacity of the fault. In some embodiments, fault current modeling is performed based on current waveform data using the formula:

[0040] in, It is through The fault energy accumulation at each moment quantifies the fault destructive force and determines the degree of equipment damage. The energy accumulation rate decreases nonlinearly with time. It is the equivalent fault resistance, which reflects the arc characteristics of the fault point. The smaller the value, the more serious the short circuit. It is set to 0.2 ohms. It is the peak value of the initial fault current, determines the energy release intensity, and is positively correlated with the short-circuit capacity; is the attenuation coefficient, which controls the current attenuation speed and reflects the damping characteristics of the system. It is set to 0.15; It is the angular frequency that determines the oscillation frequency. It is set to 314 radians per second to accurately simulate the oscillation attenuation characteristics of the fault current. The measured data show that Every 0.1 increase in the value reduces the fault energy by 37%‌; is the integral time variable, which represents the time element of the energy accumulation process, where Indicates the time difference.

[0041] Set the processing efficiency optimization function as:

[0042] in, is the time disaster reduction efficiency, which is used to obtain the efficiency peak point corresponding to the optimal intervention window; It reflects the theoretical disaster reduction effect of the intervention measures and is the accumulated fault energy corresponding to 10ms, which is used to control the upper limit of the prediction time. It is the cost of quantifying the disposal action (such as the energy consumption of the circuit breaker mechanism), and can be set to 0.3 based on actual experimental measurements. Take the derivative and set it to zero to get the extreme point.

[0043] The evaluation indicators include energy blocking rate and energy consumption reduction rate per unit time. The efficiency is highest when the energy blocking rate is equal to the average disaster reduction efficiency. The evaluation effects at different time nodes are obtained, and the optimal intervention time is determined based on the data of the evaluation indicators, that is, the efficiency peak point, for example , indicating that 2 milliseconds (ms) after the abnormal time point is the optimal intervention time, that is, the efficiency peak point. In this application, ms refers to milliseconds.

[0044] In this embodiment, based on the obtained dependency weights, the risk values ​​of multiple devices are superimposed to obtain a risk intensity value, the chain reaction across devices is quantified, and the total risk of the fault chain is accurately calculated. This solves the problem of misjudging the risk of remote devices by using only electrical weights in traditional models. At the same time, the time axis coordinates are added to predict the optimal intervention time node, solving the problem that traditional methods use static risk values ​​and cannot predict delayed faults in circuits.

[0045] In some embodiments, calculating the risk delay growth curve of the secondary path includes: obtaining device nodes in the secondary path and the corresponding initial dependency weights, calculating the risk secondary value corresponding to the time change of each device node, and generating the risk delay growth curve based on the risk secondary value and the time node. Specifically, the initial value of the risk secondary value of each device is 0, and the risk secondary value at the corresponding moment is calculated based on the initial dependency weight, the environmental acceleration factor, and the line degradation rate. The calculation formula is:

[0046] in, It is through The risk sub-value at the moment is different from the main path in that the time interval used here is in hours; is the initial dependency weight of the secondary path device y; The environmental acceleration factor is determined based on the impact of the abnormal device on the environmental conditions of subsequent devices after the abnormal device occurs. The value of the environmental acceleration factor is determined based on a specific analysis of the abnormal situation. For example, for an abnormality after a lightning strike, the environmental acceleration factor is set to 1.5. This application does not impose specific restrictions on this. It is the line degradation rate. According to the degradation effect of abnormal equipment on the line after an abnormality occurs, it can be identified through oil chromatography analysis and special gas content. The unit is the degree of degradation corresponding to each hour. For example, it is set to 0.02. This application does not impose specific restrictions on this. For the analysis of the equipment in the secondary path, the time interval can be expanded according to the actual situation, thereby improving the data analysis effect. According to the risk secondary value at different time intervals, the third-order derivative of the risk secondary value is taken to calculate the risk change rate, identify the risk growth rate inflection point (including the critical point where the risk growth rate changes from increasing to decreasing or from decreasing to increasing), and determine the optimal prevention time. The inflection point condition is inversely proportional to the line degradation rate. If the line degradation rate is 0.02, the corresponding time is 50 hours. Less than 50 hours indicates that it is the stage of accelerated risk accumulation, and more than 50 hours indicates that it is the stage of slowing risk accumulation. Then 50 hours after the abnormality is identified is the optimal prevention time, and it is also the last time window for preventive maintenance, which effectively avoids secondary failures. After obtaining the optimal intervention time and the optimal prevention time, preventive intervention is performed on the main path. After the main path action occurs, the parameter values ​​of the associated sub-path are updated, and the optimal prevention time is calculated to achieve linkage optimization of the main and sub-paths.

[0047] S4: If the risk intensity value is not greater than the high threshold, between the optimal intervention time and the optimal prevention time, identify alternative devices based on the status of the abnormal device and alternative indicators, build a complementary characteristic matrix, and determine alternative transmission paths; the alternative indicators include functional compensation degree, capacity matching rate, and switching delay.

[0048] The high threshold is pre-set and used to measure the severity of abnormal equipment. It needs to be dynamically set based on historical experimental data. For example, if the high threshold is set to 0.6, if the risk intensity value is no greater than 0.6, the current abnormality is not critical. If the risk intensity value is greater than 0.6, it indicates that the current abnormality is critical and has a high probability of causing serious losses, requiring immediate measures such as equipment replacement or line replacement. For situations where the risk intensity value is no greater than the high threshold, further adjustments are required after the initial intervention (intervention measures performed at the optimal intervention time). The timing of the adjustment is between the optimal intervention time and the optimal prevention time, ensuring that the probability of abnormality occurrence is further eliminated while ensuring rapid loss reduction.

[0049] In some embodiments, the functional compensation degree is a measure of the percentage of the replacement equipment that functionally covers the key capabilities of the abnormal equipment, ensuring the functional integrity of the system after replacement (such as whether the spare transformer can assume the same voltage conversion function when the transformer fails), extracting the core function list (such as voltage conversion, reactive power compensation) through the equipment technical manual and real-time monitoring data to verify the functional availability (such as the no-load test results of the spare equipment); the capacity matching rate is the adaptation ratio of the replacement equipment capacity to the abnormal equipment demand to prevent overload risks (such as when a 500kV transformer fails, verifying whether the spare equipment can carry a 480MVA load); the switching delay is the time required to switch from the abnormal equipment to the replacement equipment, including mechanical action time (circuit breaker opening and closing), grid synchronization time, and system stabilization time, quantifying the impact of the switching process on system continuity.

[0050] In some embodiments, a complementary characteristic matrix is ​​constructed based on the status of abnormal devices, including: the complementary characteristic matrix is ​​a three-dimensional matrix comprising candidate replacement devices, abnormal status values, and a comprehensive value; the abnormal status of the abnormal devices is analyzed to determine the function loss rate, capacity gap, and risk diffusion rate, and a weighted sum is performed (for example, weights are set to 0.4 for the function loss rate, 0.4 for the capacity gap, and 0.2 for the risk diffusion rate) to obtain the abnormal status value. The comprehensive value is the difference between the replacement index value and the abnormal status value of each candidate replacement device. A larger difference indicates a greater replacement capability of the candidate replacement device. A difference less than 0 indicates that the candidate replacement device cannot be replaced, and all candidate replacement devices with a value greater than 0 in the complementary characteristic matrix are considered to be replacement devices.

[0051] The function loss rate refers to the ratio of the number of key functions lost by abnormal equipment to its original total number of functions, which measures the degree of system function degradation caused by the fault. For example, the original five functions of the transformer are: voltage conversion, reactive power compensation, harmonic suppression, insulation isolation and overload protection. After the abnormality is discovered, only the voltage conversion and harmonic suppression functions can be used normally. In this case, the function loss rate is 60%. At this time, the backup equipment must be activated to compensate for the missing function; the capacity gap is the difference ratio between the current load demand of the abnormal equipment and the maximum load capacity of the alternative equipment, which is used for overload risk warning and safety warning. For example, if the load demand of a faulty transformer is 480MVA (megavolt-ampere, a unit of power) and the maximum capacity of the backup transformer is 420MVA, then 60MVA of non-critical load needs to be cut off; the risk diffusion rate is the rate of change of the main path risk intensity value over time.

[0052] In some embodiments, determining an alternative transmission path includes: determining a number of candidate paths based on a complementary characteristic matrix; determining an initial alternative device for each candidate path based on inter-device connectivity; retaining candidate paths corresponding to initial alternative devices with dependency weights greater than 0.6 for the abnormal device; sorting the dependency weights of subsequent alternative devices in descending order, and selecting them sequentially to form an alternative transmission path. Specifically, based on the dependency weights of the initial alternative devices, a portion of qualified candidate paths is selected, and the abnormal device (the initial alternative device closest to the abnormal device) is determined as the first alternative. Subsequent alternative devices are sorted in descending order based on their dependency weights and selected sequentially, with only one device corresponding to the same line node being selected, ultimately forming an alternative transmission path. Based on the resulting alternative transmission path, the transmission process of the current power line is adjusted to further ensure power safety. For example, the oil temperature monitoring point data for the main transformer (device number T1) of a 500kV substation is abnormal: real-time temperature: 95°C, temperature rise rate: 8°C / min. Sustained high temperatures may cause insulating oil decomposition, leading to explosions. Construct a topological network, calculate the dependency weights of associated devices, and form direct and indirect association sets; locate abnormal time nodes, including the alarm time of the monitoring system and reverse inference of the actual abnormality start, calculate the risk intensity value of each device in the main path, obtain the optimal intervention time, and solve the problem of fire; calculate the risk delay growth curve of the secondary path, identify the risk growth rate inflection point, determine the optimal prevention time, and then determine the alternative transmission path to eliminate the risk of fire in the future.

[0053] In this embodiment, by quantifying electrical dependence and physical distance, the limitation of traditional topology networks that only focus on physical connections is broken through, the key paths of fault conduction are accurately locked, the dynamic risk intensity value is superimposed on the risks of multiple devices, the cross-device chain reaction is quantified, and the problem of misjudgment of remote device risks is solved. The optimal intervention time is predicted to solve the defect that the static risk model cannot handle delayed faults; the inflection point of the secondary path risk growth rate is identified to provide the last time window for preventive maintenance to avoid secondary faults. By calculating and determining an alternative transmission path between the optimal intervention time and the optimal prevention time, the current abnormal equipment is protected, the occurrence of faults is further reduced, the secondary propagation of faults is prevented, and greater power losses are eliminated. For the abnormal detection of transformers in power systems, especially substations, it has the effect of accurate monitoring, timely warning and abnormal handling.

[0054] Example 2: The above example calculates the optimal intervention time through the fault energy accumulation model and the derivative extreme point. However, the power system fault has a strong nonlinear characteristic, which leads to a deviation between the theoretical optimal time and the actual effect. This example makes further improvements based on the above content.

[0055] Continue to refer Figure 1 The method also includes S5: obtaining historical intervention data records, generating a high-dimensional state vector and defining a set of intervention actions, and constructing a digital twin model; based on the historical intervention data records and the digital twin model, simulating the response process under different fault scenarios, and introducing random disturbances, and obtaining the optimal intervention actions and corresponding intervention times under different abnormal states according to the reward function indicators; the random disturbances include new energy output fluctuations, load mutations and sensor noise, etc., which are used to enhance the generalization ability of the intelligent agent.

[0056] The historical intervention data record includes fault type, occurrence time, intervention action, loss result and time label. The time label includes theoretical optimal intervention time and actual optimal time. The actual optimal time is the time obtained through expert analysis after the abnormality occurs.

[0057] The high-dimensional state vector includes equipment status, risk assessment indicators and environmental parameters; the equipment status can be obtained based on the basic data set, including at least basic parameters such as transformer temperature, line current and voltage deviation; the risk assessment indicators include the risk intensity value change trend, the main path propagation speed and the delay change speed of the secondary path; the environmental parameters include the proportion of renewable energy output (wind power / photovoltaic), system frequency fluctuations and basic environmental parameters (temperature, humidity, light, etc.).

[0058] The intervention action set includes early intervention, delayed intervention (extending observation before performing intervention action) and action granularity. The action granularity is a fixed time step, and the specific number of delay or advance steps is intelligently selected according to the discretization of the time step (such as 1ms).

[0059] The reward function indicators include disaster reduction benefits (reduction ratio of fault losses), operation costs (number of circuit breaker trips, equipment downtime) and system stability (voltage fluctuation, frequency deviation).

[0060] In some implementations, for example, three years of substation fault records are collected, including fault type, occurrence time, intervention action, and loss results. Two types of time tags are annotated: theoretical optimal intervention time and actual optimal time. A digital twin model is constructed by replicating the actual substation topology (transformer / circuit breaker / transmission line), configuring the topology and parameters (such as impedance and capacity). Preset fault scenarios cover 80% of common fault types (such as single-phase ground fault and phase-to-phase short circuit), with three severity levels (mild, moderate, and severe) for each scenario. For example, a combined "lightning strike + short circuit" fault is set, with an initial risk value of 0.5. A deep Q-network (DQN) is selected as the reinforcement learning algorithm framework, and the neural network weights are initialized (randomly or pre-trained). The capacity of the experience replay pool is set to store historical interaction data. At each time step, the agent observes the current state vector (e.g., transformer temperature 85°C, line current 1.2 times rated value, risk intensity 0.4). Actions are selected using the ε-greedy strategy, a classic exploration-exploitation balance method in reinforcement learning. The ε-greedy strategy randomly explores the action space with probability ε and selects the currently known optimal action (i.e., the action with the highest evaluation value estimated based on prior experience) with probability 1-ε to consolidate learning gains. The selected intervention action (e.g., tripping 20ms early) is executed, and the simulation environment simulates the system response (e.g., fault current is cut off and load is transferred to the backup line). The immediate reward and next state returned by the environment are observed. Positive rewards are given for a number of pre-successes (fault loss reduction > 30%), while negative rewards (i.e., penalties) are given for a number of pre-failures (fault propagation). A negative reward is given if maintaining the status quo worsens the fault. The system state is updated to the next time step (e.g., temperature drops to 80°C, current returns to normal), and the state-action-reward-new-state tuple is recorded in the experience replay pool. A batch of samples (e.g., 64) are randomly sampled from the experience replay pool to calculate the target evaluation value. The target network parameters are synchronized with the main network periodically (e.g., every 1000 steps) to stabilize the training process. Run the trained agent on an independent test set (fault scenarios that were not involved in the training) and calculate the intervention success rate (the proportion of successfully preventing the spread of the fault) and the average disaster reduction benefit. Comparison benchmark method: compare the performance with the derivative extreme point method and fixed time intervention method (such as unified tripping 100ms after the fault) of Example 1. If the test results do not meet the standards (such as the intervention success rate <90%), adjust the reward function weight (such as increasing the proportion of disaster reduction benefits to 0.7) or increase the number of training rounds. Prioritize sampling of high-reward samples to accelerate learning of key scenarios. Embed the trained agent into the substation monitoring system, receive equipment status data in real time, and output intervention suggestions. Continuously update network parameters based on the actual intervention effect to adapt to equipment aging or topology changes. The above is just an example of setting the reward function and the training process. Using reinforcement learning for training is a well-known technical means, and this application will not go into details here.

[0061] The technical solutions in the above embodiments of the present application have at least the following technical effects or advantages: This application establishes a digital twin model to simulate random disturbances, accurately capture the dynamic characteristics of the fault process, and avoid misjudgments caused by static assumptions of traditional models; comprehensively characterize the operating status of the power system, balance disaster reduction benefits and operating costs, and dynamically determine the optimal intervention actions and intervention times for different abnormal conditions caused by abnormal equipment; dynamically adjust intervention actions according to real-time state vectors to adapt to complex scenarios such as load fluctuations and changes in new energy output; through the experience replay pool and target network synchronization mechanism, fully tap the expert experience in historical intervention records to improve training efficiency and strategy stability.

[0062] Example 3: The above example optimizes the dynamic intervention time of equipment within a single substation through reinforcement learning. However, modern power systems are characterized by multi-station interconnection and regional collaboration. The defense strategy of a single site may fail to effectively implement global defense due to ignoring cross-site correlation risks (such as power flow transfer and voltage collapse caused by failures in adjacent substations).

[0063] Continue to refer Figure 1 The method further includes S6: obtaining a target area and several substations within the target area, and determining a central site; obtaining real-time data of adjacent sites, determining cross-site dependency weights based on adjacent station tie line parameters, and dividing collaborative defense areas; when abnormal equipment is detected, identifying the abnormal state, obtaining the optimal intervention action and corresponding intervention time of the central site according to step S5, and determining the cross-site propagation probability and resource requirements of the central site, and generating a global management plan based on global risk constraints determined in the collaborative defense area.

[0064] In some embodiments, a central site and its defense radius are determined based on the basic data set of each substation in the target area. Specifically, the structural importance coefficient of each substation is calculated based on the power grid topology data, the real-time load data of the substation, and the historical fault records (the three data are normalized and weighted summed to obtain the structural importance coefficient), and the optimal central site is determined. The real-time data is the operating status parameters of each substation.

[0065] The defense radius is formed by determining the coordination range based on the capacity of the central site. The defense radius refers to the dynamic range of action defined in the power collaborative defense system based on the power supply capacity, grid topology and equipment relevance of the central site, with the central site as the core. This radius determines which adjacent substations need to be included in the same collaborative defense area to achieve cross-site resource scheduling, risk joint defense and unified decision-making. Based on the geographical distance (such as 30km, 20km) or electrical distance (such as equivalent impedance value), combined with the dynamic adjustment of the central site capacity, this application does not impose specific restrictions. Based on the power grid geographic information system (GIS), identify substation clusters with a common power supply area. For example, a certain area contains 5 220kV substations A / B / C / D / E. Through GIS analysis, it is determined that its power supply range covers urban areas and suburbs.

[0066] The parameters of the adjacent station interconnection lines include the voltage level ratio and the power transmission ratio. The cross-station dependency weight is calculated based on the adjacent station interconnection line parameters and the dependency weight calculation method in Example 1. The collaborative defense area is divided based on the cross-station dependency weight and the minimum defense radius of the remaining sites. A clustering method is used, and the optimization goal is to maximize the connection density within the area and minimize the connection between areas. The collaborative defense area includes a strong coupling area, a medium coupling area, and a weak coupling area. For example, five substations are divided into two defense areas, {A, B, C} and {D, E}. The cross-area flow at the junction of the areas is monitored in real time, and the cross-area transmission path is automatically isolated in the event of an anomaly. The risk resistance index of each area is calculated as: the proportion of redundant lines in the area × 0.4 + the backup rate of key equipment × 0.3 + the resource self-sufficiency rate (resources available in the area / total demand) × 0.3.

[0067] In some embodiments, the cross-site propagation probability and resource requirements of the central site are determined, and a global management plan is generated based on the global risk constraints determined in the collaborative defense area, including: the cross-site propagation probability is the rate of change of the risk intensity value after the central site intervention action in a unit time, and the unit time is dynamically set according to the range of the target area, and its value is less than the optimal prevention time of the central site. The resource requirements include but are not limited to emergency repair resources, spare equipment and human resources. The global risk constraints are risk propagation constraints, fairness constraints and timeliness constraints. Specifically, the risk propagation constraint means that any resource allocation shall not cause the risk value of the adjacent area to increase by more than 15%; the fairness constraint refers to the fairness of resource allocation to avoid resource accumulation; the timeliness constraint refers to the resource allocation time, which should be minimized.

[0068] In some embodiments, a global management plan is generated based on global risk constraints determined by collaborative defense zones. The power system is monitored, managed, and exceptions handled according to the global management plan, with real-time results tracked. Key indicators include: fault propagation range (whether it breaches the defense zone), system recovery time (from anomaly occurrence to stable operation), and resource utilization (actual use / planned use ratio). This layered collaborative architecture of dynamic centers and adaptive defense zones enables spatiotemporal coordinated optimization of defense strategies, building a cross-site defense system that supports a high proportion of energy access and adapts to the needs of new power systems.

[0069] The technical solutions in the above embodiments of the present application have at least the following technical effects or advantages: This application uses dynamic dependency weights and propagation probability calculations to clarify fault transmission paths and avoid defense blind spots caused by traditional methods that ignore cross-site associations; based on quantified resource requirements and global constraints, it implements on-demand resource scheduling to avoid local resource accumulation or insufficiency; through regional division and risk constraint mechanisms, it ensures that single-site defense actions do not cause increased risks at other sites, achieving a shift from local optimality to global optimality; constructs adaptive collaborative defense areas, breaks through the limitations of traditional fixed topology models, proposes three-dimensional constraint optimization, achieves a balance between fairness and efficiency in cross-site resource allocation, and further improves the accuracy of power safety monitoring and the efficiency of abnormality handling.

[0070] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Various modifications and variations are readily apparent to those skilled in the art. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention shall be included within the scope of protection of the present invention.

Claims

1. A power safety monitoring and management method, characterized in that: include: S1: Obtain the basic data set, calculate the dependency weights between devices and obtain the associated device set, and build the initial network graph based on the associated device set; S2: Monitor the device status in real time, identify abnormal devices and corresponding abnormal time nodes; Obtain a first association group and a second association group according to the association device set; overlap the first association group and the second association group with the initial network diagram to obtain a primary path and a secondary path; S3: Calculate the risk intensity value of each device in the primary path, obtain the disposal efficiency and evaluation indicators at different time nodes, and determine the optimal intervention time. Calculate the risk delay growth curve of the secondary path, identify the risk growth inflection point, and determine the optimal prevention time. S4: If the risk intensity value is not greater than the high threshold, between the optimal intervention time and the optimal prevention time, identify alternative devices based on the status and alternative indicators of the abnormal device, and construct a complementary characteristic matrix to determine the alternative transmission path.

2. A power safety monitoring and management method according to claim 1, characterized in that: The method further includes: S5: obtaining historical intervention data records, generating a high-dimensional state vector and defining a set of intervention actions, and constructing a digital twin model; based on the historical intervention data records and the digital twin model, simulating response processes under different fault scenarios, introducing random disturbances, and obtaining optimal intervention actions and corresponding intervention times under different abnormal states according to reward function indicators; S6: Obtain the target area and several substations within the target area, and determine a central site; obtain real-time data of adjacent sites, determine the cross-site dependency weight based on the parameters of the adjacent station interconnection lines, and divide the collaborative defense area; when abnormal equipment is detected, identify the abnormal state, obtain the optimal intervention action and corresponding intervention time of the central site according to step S5, and determine the cross-site propagation probability and resource requirements of the central site, determine the global risk constraints based on the collaborative defense area, and generate a global management plan.

3. The power safety monitoring and management method according to claim 1, characterized in that: The basic data set includes electrical parameters, physical locations and control logic of the power equipment; The dependency weight is a parameter value used to quantify the energy transfer dependency between devices; According to the associated device set, obtain specific device nodes, establish associated edges, and mark dependency weights to form the initial network graph; The associated device set includes a direct association set and an indirect association set; according to the direct association set, devices associated with the abnormal device are obtained to form a first association group; according to the indirect association set, devices associated with the abnormal device are obtained to form a second association group.

4. The power safety monitoring and management method according to claim 1, wherein: Overlapping the first association group and the second association group with the initial network diagram respectively to obtain a main path and a secondary path, including: determining the specific location of the abnormal device in the initial network diagram, extracting the device nodes in the first association group, and connecting the device nodes to form a main path; extracting the device nodes in the second association group, and connecting the device nodes to form a secondary path.

5. The power safety monitoring and management method according to claim 1, characterized in that: The risk intensity value refers to the dynamic fault risk level of the device under specific time and space coordinates; the specific time and space coordinates are the spatial coordinates and time axis coordinates of the device. The spatial coordinates are used to locate the physical location of the fault propagation, and the time axis coordinates are used to quantify the dynamic evolution of the fault over time. Based on the current waveform data of abnormal equipment obtained from the data set, the accumulated fault energy is calculated, and evaluation indicators are set to evaluate and verify the treatment efficiency at different time nodes to obtain the optimal intervention time; The fault energy accumulation refers to the fault energy generated over time, which is used to measure the fault damage capability; the evaluation indicators include energy blocking rate and energy consumption reduction rate per unit time.

6. The power safety monitoring and management method according to claim 1, characterized in that: Calculate the risk delay growth curve of the secondary path, identify the risk growth inflection point, and determine the optimal prevention time, including: obtaining the device nodes in the secondary path and the corresponding initial dependency weights, calculating the risk secondary value corresponding to the time change of each device node, and generating a risk delay growth curve based on the risk secondary value and time node; calculate the risk change rate based on the risk secondary value, identify the risk growth inflection point, and determine the optimal prevention time; The initial value of the risk sub-value of each device is 0, and the risk sub-value at the corresponding moment is calculated based on the initial dependency weight, environmental acceleration factor and line degradation rate.

7. The power safety monitoring and management method according to claim 1, characterized in that: The high threshold is preset and is used to measure the abnormality degree of the abnormal device; The replacement indicators include functional compensation degree, capacity matching rate and switching delay. The functional compensation degree is a measure of the percentage of the replacement equipment that functionally covers the key capabilities of the abnormal equipment; the capacity matching rate is the adaptation ratio of the replacement equipment capacity to the abnormal equipment requirements; the switching delay is the time required to switch from the abnormal equipment to the replacement equipment, including mechanical action time, grid synchronization time and system stabilization time.

8. The power safety monitoring and management method according to claim 1, characterized in that: Constructing a complementary characteristic matrix based on the state of the abnormal device, including: the complementary characteristic matrix is ​​a three-dimensional matrix including candidate replacement devices, abnormal state values, and a comprehensive value; analyzing the abnormal state of the abnormal device to determine the function loss rate, capacity gap, and risk diffusion rate, and performing weighted summation to obtain the abnormal state value; The comprehensive value is the difference between the substitution index value and the abnormal state value of each candidate replacement device. The larger the difference, the greater the replacement capability of the candidate replacement device. Determining an alternative transmission path includes: determining a number of candidate paths according to a complementary characteristic matrix, determining an initial alternative device for each candidate path according to a connection relationship between devices, retaining candidate paths corresponding to initial alternative devices having a dependency weight greater than 0.6 for the abnormal device, arranging the dependency weights of subsequent alternative devices in descending order, and selecting them in sequence to form an alternative transmission path.

9. The power safety monitoring and management method according to claim 2, wherein: The historical intervention data record includes the fault type, occurrence time, intervention action, loss result and time tag. The time tag includes the theoretical optimal intervention time and the actual optimal time. The actual optimal time is the time obtained through expert analysis after the anomaly occurs. The high-dimensional state vector includes device status, risk assessment indicators and environmental parameters; the risk assessment indicators include the risk intensity value change trend, the main path propagation speed and the delay change speed of the secondary path; The intervention action set includes early intervention, delayed intervention and action granularity. The action granularity is a fixed time step, and the specific number of delay or advance steps is intelligently selected according to the discretization of the time step; the reward function indicators include disaster reduction benefits, operating costs and system stability; the random disturbances include fluctuations in new energy output, sudden load changes and sensor noise.

10. The power safety monitoring and management method according to claim 2, characterized in that: Based on the basic data set of each substation in the target area, a central site and its defense radius are determined; based on the capacity of the central site, the coordination range is determined to form a defense radius; The parameters of the adjacent station tie lines include the voltage level ratio and the power transmission ratio. The collaborative defense area is divided according to the cross-station dependency weight and the minimum defense radius of the remaining stations. A clustering method is used, and the optimization goal is to maximize the connection density within the area and minimize the connection between areas. The collaborative defense area includes a strong coupling area, a medium coupling area, and a weak coupling area. The cross-site transmission probability is the rate of change of the risk intensity value after the central site intervention action within a unit time. The unit time is dynamically set according to the range of the target area, and its value is less than the optimal prevention time of the central site.

Images