Power system asset risk prediction method and system based on big data analysis

Abstract

The invention provides an electric power system asset risk prediction method and system based on big data analysis, and the method comprises the steps: carrying out the correlation analysis of all data in an electric power data set, and determining the potential correlation between an equipment operation state and environment meteorological data; analyzing the risk influence degree of the environmental meteorological factors on the power system assets according to the potential association relationship, and determining key risk factors; constructing a risk scene based on the key risk factors; power system asset operation is simulated in each risk scene, and the risk level of each risk scene is determined based on the probability of power system asset failure in each risk scene and the interference influence degree of the power system asset failure on the power grid system; and matching the environmental meteorological data of the power system assets monitored in real time with the risk scene, and determining the risk level of the power system assets having a fault. According to the invention, the comprehensiveness and accuracy of asset risk prediction are improved.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a method and system for predicting power system asset risks based on big data analysis. Background Art

[0002] In the operation and maintenance of power systems, accurate prediction of asset risks is crucial to ensure the safe and stable operation of the power grid. At present, the common power system asset risk prediction method is to conduct risk assessment based on the operating parameter thresholds of the equipment, mainly setting fixed thresholds for the key operating parameters of the equipment. When the equipment operating parameters exceed these thresholds, the equipment is judged to be at risk.

[0003] Existing methods can, to a certain extent, make a preliminary assessment of the risk of power system assets and help operation and maintenance personnel to promptly discover some obvious equipment abnormalities. However, they ignore the impact of equipment operating environment factors and historical fault records of the power grid on asset risks. The operating status of power equipment depends not only on its own operating parameters, but also on the surrounding environmental factors (such as humidity, air pressure, pollution level, etc.) and the past faults of the power grid. For example, in an environment with high humidity, the insulation performance of electrical equipment may decline, and there may be potential failure risks even if its operating parameters do not exceed the threshold. At the same time, the historical fault records of the power grid contain a lot of useful information, such as the occurrence patterns of certain faults, the weak links of related equipment, etc. The existing methods fail to make full use of this information, resulting in the prediction of asset risks that is not comprehensive and accurate enough, and it is impossible for power operation and maintenance personnel to clearly grasp the asset risk situation. Summary of the invention

[0004] The present invention provides a method and system for predicting power system asset risks based on big data analysis, which are used to improve the comprehensiveness and accuracy of asset risk prediction, so that power operation and maintenance personnel can more clearly grasp the power system asset risk situation and achieve more effective power system operation and maintenance and management.

[0005] In a first aspect, the present invention provides a method for predicting power system asset risks based on big data analysis, comprising: Conduct correlation analysis on various data in the power data set to determine the potential correlation between equipment operation status and environmental meteorological data; Analyze the risk impact of environmental meteorological factors on power system assets based on the potential correlation relationship and determine key risk factors; Construct risk scenarios based on the key risk factors; each risk scenario represents a combination of key risk factors at different levels; Simulate the operation of power system assets under each risk scenario, and determine the risk level of each risk scenario based on the probability of power system asset failure under each risk scenario and the degree of interference of power system asset failure on the power grid system; The environmental meteorological data of the power system assets monitored in real time are matched with the risk scenarios, and the risk level of the matched target risk scenarios is determined as the risk level of failure of the power system assets.

[0006] In a second aspect, the present invention further provides a power system asset risk prediction system based on big data analysis, which is applied to the power system asset risk prediction method based on big data analysis as described in the first aspect; the power system asset risk prediction system based on big data analysis includes: The correlation analysis module is used to perform correlation analysis on each data in the power data set to determine the potential correlation between the equipment operation status and the environmental meteorological data; A risk factor screening module is used to analyze the risk impact of environmental meteorological factors on power system assets based on the potential correlation relationship and determine key risk factors; A risk scenario construction module, used to construct risk scenarios based on the key risk factors; each risk scenario represents a combination of key risk factors at different levels; The risk scenario simulation module is used to simulate the operation of power system assets under each risk scenario, and determine the risk level of each risk scenario based on the probability of power system asset failure under each risk scenario and the degree of interference of power system asset failure on the power grid system; The risk prediction module is used to match the environmental meteorological data of the power system assets monitored in real time with the risk scenario, and determine the risk level of the matched target risk scenario as the risk level of the power system asset failure.

[0007] In a third aspect, the present invention also provides an electronic device, comprising: a memory for storing a computer software program; a processor for reading and executing the computer software program, thereby implementing any of the above-mentioned methods for predicting power system asset risks based on big data analysis.

[0008] In a fourth aspect, the present invention further provides a non-transitory computer-readable storage medium, wherein a computer software program is stored in the storage medium, and when the computer software program is executed by a processor, the power system asset risk prediction method based on big data analysis as described in any one of the above-mentioned methods.

[0009] In a fifth aspect, the present invention also provides a computer program product, including a computer program, which, when executed by a processor, implements any of the above-mentioned methods for predicting power system asset risks based on big data analysis.

[0010] The power system asset risk prediction method based on big data analysis provided by the embodiment of the present invention analyzes the potential correlation between the equipment operation status and environmental factors and meteorological factors through correlation, so that the potential correlation can be fully utilized to screen out the key risk factors with greater risk impact on the power system assets, avoiding the interference of irrelevant factors. Further, the operation of the power system assets is simulated in each risk scenario to accurately obtain the risk level of each risk scenario. In the subsequent risk prediction process of the power system assets, the risk level of the power system asset failure can be accurately predicted based on the matching results of the environmental meteorological data of the power system assets and the risk scenario. Therefore, the embodiment of the present invention can comprehensively consider various risk factors affecting the power system assets for risk prediction, improve the comprehensiveness and accuracy of the prediction of asset risks, and enable power operation and maintenance personnel to more clearly grasp the risk situation of the power system assets and realize more effective operation and management of the power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 It is a flow chart of a method for predicting power system asset risk based on big data analysis provided by an embodiment of the present invention; Figure 2 It is a structural schematic diagram of a power system asset risk prediction system based on big data analysis provided by an embodiment of the present invention; Figure 3 An embodiment diagram of an electronic device provided by an embodiment of the present invention; Figure 4 An embodiment diagram of a computer-readable storage medium provided for an embodiment of the present invention. DETAILED DESCRIPTION

[0012] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by those skilled in the art without making creative work are within the scope of protection of the present invention. In the description of the present invention, the terms "first" and "second" are used only for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the number of indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the said features. In the description of the present invention, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined.

[0013] In the description of the present invention, the term "for example" is used to mean "used as an example, illustration or explanation". Any embodiment described as "for example" in the present invention is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is given to enable any technician in the field to implement and use the present invention. In the following description, details are listed for the purpose of explanation. It should be understood that a person of ordinary skill in the art can recognize that the present invention can be implemented without using these specific details. In other examples, well-known structures and processes will not be elaborated in detail to avoid obscuring the description of the present invention with unnecessary details. Therefore, the present invention is not intended to be limited to the embodiments shown, but is consistent with the widest scope consistent with the principles and features disclosed in the present invention.

[0014] Optional, see Figure 1 , Figure 1 It is a flow chart of the power system asset risk prediction method based on big data analysis provided by the present invention. The execution subject of the power system asset risk prediction method based on big data analysis in the embodiment of the present invention is a risk prediction system. Therefore, the power system asset risk prediction method based on big data analysis includes: Step 10: Perform correlation analysis on each item of data in the power data set to determine the potential correlation between the equipment operation status and the environmental meteorological data.

[0015] Optionally, the power system assets in the embodiment of the present invention are various power resources in the power grid system that constitute the power system and provide support for its operation, power supply, etc. Therefore, the power system assets may include power generation assets, transmission assets, distribution assets, etc., power generation assets such as power generation equipment, transmission assets such as transmission substations, and distribution assets such as distribution equipment.

[0016] Optionally, the power data set in the embodiment of the present invention includes equipment operation status information, historical power grid fault data, environmental factor data and meteorological factor data. Equipment operation status information includes operation parameters, start and stop status, etc. of power system assets; historical power grid fault data includes the time, location, and type of fault; environmental factor data includes the geographical environment and pollution level of the area where the power system assets are located; meteorological factor data includes temperature, humidity, wind speed, etc.

[0017] Therefore, the risk prediction system obtains the power data set, performs correlation analysis on the data in the power data set, and determines the potential correlation between the equipment operation status of the power system assets and the environmental factors and meteorological factors, as described in steps 101 to 104. In one embodiment, under the conditions of humidity exceeding 80%, temperature exceeding 30°C and wind speed exceeding 15m / s, the probability of failure of a certain type of transformer increases significantly, and the potential correlation between the transformer and humidity and temperature is obtained.

[0018] Step 20, analyzing the risk impact of environmental meteorological factors on power system assets based on potential correlation relationships, and determining key risk factors.

[0019] Furthermore, the risk prediction system analyzes the target environmental factors and target meteorological factors that affect the equipment operating status of the power system assets according to the potential correlation relationship, and combines the target environmental factors and target meteorological factors into different factor combinations. Continuing with the above embodiment, the potential correlation relationship is: when the humidity exceeds 80%, the temperature exceeds 30°C, and the wind speed is greater than 15m / s, the probability of transformer failure increases significantly. Therefore, the target environmental factors are temperature and humidity, and the target meteorological factor is wind speed. The factor combination can be {(temperature); (humidity); (wind speed); (humidity, temperature); (humidity, wind speed); (temperature, wind speed); (humidity, temperature, wind speed)}.

[0020] Furthermore, the risk prediction system analyzes the risk impact degree of power system assets under different factor combinations, that is, the risk impact degree of different factor combinations on the equipment operating status of power system assets, and determines the factors in the factor combinations with greater risk impact as key risk factors, as specifically described in steps 201 to 205.

[0021] Step 30, construct risk scenarios based on key risk factors.

[0022] Furthermore, the risk prediction system divides the key risk factors into different levels based on the stored historical experience, and combines the key risk factors at different levels to obtain risk scenarios. In one embodiment, temperature is divided into high temperature and low temperature, humidity is divided into high humidity and low humidity, wind speed is divided into high wind speed and low wind speed, and the risk scenarios are high temperature, high humidity and high wind speed, high temperature, low humidity and high wind speed, low temperature, high humidity and low wind speed, etc.

[0023] Step 40, simulating the operation of power system assets under each risk scenario, and determining the risk level of each risk scenario based on the probability of power system asset failure under each risk scenario and the degree of interference impact of power system asset failure on the power grid system.

[0024] Furthermore, the operation of power system assets is simulated under each risk scenario, and the risk prediction system analyzes the probability of power system asset failure under each risk scenario based on the historical power grid failure data of the power system assets, wherein the historical power grid failure data is summarized based on historical experience.

[0025] Furthermore, the risk prediction system determines the scope and duration of power outages in the power grid system when the power system assets fail, and determines the degree of interference of the power system asset failure on the power grid system based on the preset rules and the power outage scope and duration of the power grid system, wherein the preset rules such as the larger the power outage scope, the greater the interference impact, and the longer the duration, the greater the interference impact can be specifically calculated through a weighted algorithm and will not be elaborated here.

[0026] Furthermore, the risk prediction system determines the risk level of each risk scenario based on the probability of power system asset failure in each risk scenario and the degree of interference of the power system asset failure on the power grid system, as described in steps 401 to 404.

[0027] Step 50, matching the environmental meteorological data of the power system assets monitored in real time with the risk scenario, and determining the risk level of the matched target risk scenario as the risk level of the power system asset failure.

[0028] Furthermore, the risk prediction system monitors the environmental meteorological data of the power system assets in real time, and the environmental meteorological data are environmental factor data and meteorological factor data, and matches the environmental factor data and meteorological factor data of the power system assets monitored in real time with the risk scenario to obtain a target risk scenario in the risk scenario that matches the environmental factor data and meteorological factor data.

[0029] Furthermore, the risk prediction system determines the risk level of the matched target risk scenario as the risk level of power system asset failure.

[0030] The embodiment of the present invention analyzes the potential correlation between the equipment operation status and environmental factors and meteorological factors through correlation, so that the potential correlation can be fully utilized to screen out the key risk factors with greater risk impact on the power system assets, avoiding the interference of irrelevant factors. Further, the operation of the power system assets is simulated under each risk scenario to accurately obtain the risk level of each risk scenario. In the subsequent risk prediction process of the power system assets, the risk level of the power system asset failure can be accurately predicted based on the matching results of the environmental meteorological data of the power system assets and the risk scenario. Therefore, the embodiment of the present invention can comprehensively consider various risk factors affecting the power system assets for risk prediction, improve the comprehensiveness and accuracy of the prediction of asset risks, and enable power operation and maintenance personnel to more clearly grasp the risk situation of the power system assets and realize more effective operation and management of the power system.

[0031] In one embodiment, the description of steps 101 to 104 is as follows: Step 101 , mapping the equipment operation status information, environmental factor data and meteorological factor data in the power data set into a multi-dimensional space to construct an initial three-dimensional correlation matrix.

[0032] Optionally, the risk prediction system performs multidimensional spatial mapping on the equipment operation status information, environmental factor data and meteorological factor data in the power data set. The multidimensional spatial mapping in the embodiment of the present invention is a three-dimensional spatial mapping, so the initial three-dimensional correlation matrix is ​​constructed. , where each matrix element in the initial three-dimensional correlation matrix represents the initial correlation strength of the equipment operation status information, environmental factor data, and meteorological factor data, and the initial correlation strength represents the ratio between the number of data instances that simultaneously contain the equipment operation status information, environmental factor data, and meteorological factor data and the total number of data instances in the power data set.

[0033] Therefore, for the equipment operation status information , Environmental Factor Data , meteorological factor data , each matrix element in the initial three-dimensional incidence matrix It can be expressed as: .in, represents the electric power dataset, Represents the first data instances, Represents the first data instances, Represents the first data instances, Indicates that it also contains the device operation status information , Environmental Factors Data and meteorological factors data The number of data instances, Represents the total number of data instances of the electric power dataset.

[0034] Step 102 , the initial three-dimensional correlation matrix is ​​screened based on a preset threshold value and each matrix element in the initial three-dimensional correlation matrix to obtain a screened three-dimensional correlation matrix.

[0035] Furthermore, the embodiment of the present invention pre-sets a preset threshold , through the preset threshold For each matrix element in the initial three-dimensional incidence matrix If , the risk prediction system ignores the correlation relationship corresponding to the matrix elements and obtains the three-dimensional correlation matrix after screening.

[0036] Therefore, each matrix element in the three-dimensional incidence matrix after screening can be expressed as: .

[0037] Step 103: Update each matrix element in the screened three-dimensional incidence matrix based on a preset time window to obtain an updated three-dimensional incidence matrix.

[0038] Furthermore, since the power data has the characteristics of time series, the data association of the equipment operation status information, environmental factor data and meteorological factor data is analyzed within the time window. Therefore, the risk prediction system calculates the updated matrix element according to the ratio between the number of data instances that simultaneously contain the equipment operation status information, environmental factor data and meteorological factor data within the preset time window and the total number of data instances within the preset time window. , the specific formula is as follows: .

[0039] in, Indicates the preset time window It also contains the device operation status information , Environmental Factors Data and meteorological factors data The number of data instances, Indicates the preset time window The total number of data instances in .

[0040] Furthermore, the risk prediction system is based on the updated matrix elements Each matrix element in the screened three-dimensional incidence matrix is ​​updated to obtain an updated three-dimensional incidence matrix.

[0041] Step 104, based on the screened three-dimensional correlation matrix and the updated three-dimensional correlation matrix, determine the potential correlation relationship between the equipment operation status and the environmental factors and the meteorological factors.

[0042] Furthermore, the risk prediction system determines the potential correlation between the equipment operating status and environmental factors and meteorological factors based on each matrix element in the screened three-dimensional correlation matrix and each matrix element in the updated three-dimensional correlation matrix, as described in steps 1041 to 1044.

[0043] The embodiment of the present invention analyzes the potential correlation between the equipment operating status and environmental factors and meteorological factors through correlation, so that the potential correlation can be fully utilized to screen out key risk factors with greater risk impact on power system assets, avoiding interference from irrelevant factors. Therefore, risk scenarios can be accurately constructed based on the key risk factors to predict the risk factors affecting power system assets, thereby improving the comprehensiveness and accuracy of asset risk prediction.

[0044] In one embodiment, the description of steps 1041 to 1044 is as follows: Step 1041 , screening is performed in the power data set based on each matrix element in the screened three-dimensional correlation matrix to obtain a preliminary correlation candidate set.

[0045] Optionally, the risk prediction system screens the power data set according to each matrix element in the screened three-dimensional association matrix, wherein the matrix element in the screened three-dimensional association matrix is ​​not 0, indicating that there is a correlation relationship between the data, and the matrix element in the screened three-dimensional association matrix is ​​0, indicating that there is no correlation relationship between the data or there is a negligible correlation relationship. Therefore, the data having corresponding matrix elements in the screened three-dimensional association matrix is ​​screened out in the power data set to obtain a preliminary correlation candidate set.

[0046] Step 1042, for each item set in the preliminary association candidate set, obtain a frequency score of each item set based on the number of occurrences of the item set combined with the first matrix element of the item set in the three-dimensional association matrix after screening and the second matrix element of the three-dimensional association matrix after updating.

[0047] Furthermore, for each item set in the preliminary association candidate set , the risk prediction system obtains each item set The number of occurrences in the preliminary association candidate set, and according to each item set The number of occurrences of and the size of the dataset in the preliminary associated candidate set are calculated for each item set The frequency of occurrence.

[0048] Furthermore, the risk prediction system is based on each item set The frequency of occurrence of each item set The first matrix element in the three-dimensional association matrix after screening and the second matrix element in the three-dimensional association matrix after updating are calculated to obtain each item set. The frequency score of , the specific formula is as follows: .

[0049] in, Representing Item Sets The number of occurrences in the preliminary association candidate set, represents the size of the dataset in the preliminary association candidate set, Representing Item Sets The first matrix element in the three-dimensional incidence matrix after screening, Representing Item Sets The size of Representing Item Sets The second matrix element in the three-dimensional incidence matrix after the update.

[0050] Step 1043: determine the item sets whose frequency scores are greater than or equal to the preset minimum support threshold as frequent item sets.

[0051] Furthermore, a preset minimum support threshold is set in advance , the risk prediction system converts each item set The frequency score of With the preset minimum support threshold Compare the numerical values ​​and determine the item sets whose frequency scores are greater than or equal to the preset minimum support threshold as frequent item sets.

[0052] Step 1044, for any two target frequent item sets in the frequent item sets, generate association rules according to the item set causal relationship and item set influence degree between the target frequent item sets, and determine the association rules as potential associations between the equipment operation status and the environmental factors and meteorological factors.

[0053] Furthermore, an embodiment of the present invention generates association rules by analyzing the order of appearance and the degree of mutual influence of any two target frequent item sets in the frequent item set in the data set. Therefore, for any two target frequent item sets in the frequent item set, the risk prediction system obtains the item set causal relationship and the item set influence degree between the target frequent item sets, wherein the item set causal relationship represents the conditional probability of the target frequent item sets.

[0054] Furthermore, the risk prediction system generates association rules based on the item set causal relationship and item set influence degree between the target frequent item sets, specifically: .

[0055] .

[0056] in, represents association rules, represents the strength of the association rule, Represents the target frequent itemsets and target frequent itemsets Item set causality, Represents the target frequent itemset For the target frequent itemsets The degree of influence of the item set; is the time interval, representing the target frequent itemset Observe the target frequent itemsets after they appear time delay in occurrence; Represents the target frequent itemset After each occurrence, , target frequent itemsets The sum of the conditional probabilities of occurrence.

[0057] Furthermore, the risk prediction system determines the association rules as potential associations between the equipment operating status and environmental factors and meteorological factors.

[0058] The embodiment of the present invention accurately obtains potential correlation relationships through item set causal relationships and item set influence degrees between frequent item sets, so that the potential correlation relationships can be fully utilized to screen out key risk factors that have a greater risk impact on power system assets, avoiding interference from irrelevant factors. Therefore, risk scenarios can be accurately constructed based on key risk factors to predict risk factors affecting power system assets, thereby improving the comprehensiveness and accuracy of asset risk prediction.

[0059] In one embodiment, the description of steps 201 to 205 is as follows: Step 201 : determining target environmental factors and target meteorological factors that affect the equipment operating status of power system assets based on potential association relationships.

[0060] Optionally, the risk prediction system analyzes the target environmental factors and target meteorological factors that affect the equipment operating status of the power system assets based on the potential correlation relationship. In one embodiment, the potential correlation relationship is: when the humidity exceeds 80%, the temperature exceeds 30°C, and the wind speed is greater than 15m / s, the probability of transformer failure increases significantly. Therefore, the target environmental factors are temperature and humidity, and the target meteorological factor is wind speed.

[0061] Step 202, for each factor combination of the target environmental factor and / or the target meteorological factor, when each factor combination occurs, if a disturbance occurs in the power system asset, the disturbance response intensity when the power system asset is disturbed is determined based on the magnitude of change in the operating state of the power system asset under the action of the factor combination.

[0062] Further, the risk prediction system arranges and combines the target environmental factors and the target meteorological factors to obtain a corresponding factor combination. Continuing with the above embodiment, the factor combination can be {(temperature); (humidity); (wind speed); (humidity, temperature); (humidity, wind speed); (temperature, wind speed); (humidity, temperature, wind speed)}.

[0063] Further, for each factor combination, when each factor combination occurs, the risk prediction system determines whether a disturbance occurs to the power system asset, that is, whether the operating state of the power system asset changes.

[0064] Furthermore, if it is determined that there is no disturbance in the power system assets, it means that when this factor combination occurs, there is no risk to the power system assets, and this factor combination is eliminated from all factor combinations.

[0065] Furthermore, if it is determined that a disturbance occurs in the power system assets, it means that when this combination of factors occurs, a risk is generated for the power system assets. Therefore, the risk prediction system obtains the extent of change in the operating status of the power system assets under the influence of the combination of factors.

[0066] Furthermore, the risk prediction system analyzes the rate of change of the operating state variables of the power system assets according to the change range of the operating state of the power system assets under the combined effect of factors, and analyzes the impact intensity of the disturbance on the power system assets according to the change rate and change range, and obtains the disturbance response intensity when the power system assets are disturbed. The specific formula is as follows: .

[0067] in, Indicates the intensity of disturbance response when a disturbance occurs in the power system assets, It indicates the extent of change in the operating status of power system assets under the influence of a combination of factors.

[0068] Step 203, based on the network topology of the power grid system, determine the number of nodes, the number of edges, the coupling relationship between nodes, and the coupling relationship between edges in the path through which the disturbance propagates, and determine the complexity of the propagation path when a disturbance occurs in the power system assets.

[0069] Furthermore, the risk prediction system determines the number of nodes, the number of edges, the coupling relationships between nodes, and the coupling relationships between edges in the path through which the disturbance propagates based on the network topology of the power grid system when a disturbance occurs to the power system assets, wherein the network topology is pre-constructed based on experience and includes the coupling relationships between nodes and the coupling relationships between edges, and the coupling relationships represent the strength of interaction.

[0070] Furthermore, the risk prediction system evaluates the complexity of the propagation path according to the number of nodes, the number of edges, the coupling relationship between nodes, and the coupling relationship between edges in the path through which the disturbance propagates, and obtains the complexity of the propagation path when the power system assets are disturbed. The specific formula is as follows: .

[0071] in, represents the complexity of the propagation path when a disturbance occurs in the power system assets, represents the number of nodes in the path, represents the number of edges in the path, Representation Node and nodes The coupling relationship between Representation Node and nodes The coupling relationship of the edges; Represents the directional weight, which characterizes the influence of the disturbance propagation direction.

[0072] Step 204 , based on the gradient of the operating state change of the power system assets in each local area of ​​the power grid system, determine the degree of disturbance fluctuation when the power system assets are disturbed.

[0073] Furthermore, the risk prediction system obtains the operating state change gradient of the power system assets in each local area of ​​the power grid system, and performs local change differences based on the operating state change gradient of the power system assets in each local area to obtain the disturbance fluctuation degree when the power system assets are disturbed. The specific formula is: .

[0074] in, Indicates the degree of disturbance fluctuation when a disturbance occurs in the power system assets. represents the number of local regions, Represents a local area The operating status change gradient of the internal power system assets, Represents a local area the number of inner subregions, Indicates sub-area With local area The coefficient of variation of the average state, Represents the minimum value function.

[0075] Step 205 , determining key risk factors based on the disturbance response intensity, propagation path complexity and disturbance fluctuation degree of the power system assets when a disturbance occurs under each combination of factors.

[0076] Furthermore, the risk prediction system determines key risk factors based on the disturbance response intensity, propagation path complexity and disturbance fluctuation degree of the power system assets when disturbances occur under each combination of factors, as described in steps 2051 to 2055.

[0077] The embodiment of the present invention accurately determines the key risk factors according to the disturbance response intensity, propagation path complexity and disturbance fluctuation degree of the power system assets when disturbance occurs under each combination of factors, thereby avoiding the interference of irrelevant factors. Therefore, it is possible to subsequently accurately construct risk scenarios based on the key risk factors to predict the risk factors affecting the power system assets, thereby improving the comprehensiveness and accuracy of the prediction of asset risks.

[0078] In one embodiment, the description of steps 2051 to 2055 is as follows: Step 2051 , for each factor combination, simulate the recovery of the disturbance to the power system assets, and obtain the delayed recovery time for the power system assets to recover to the original operating state.

[0079] Optionally, for each factor combination, the risk prediction system simulates the recovery of the disturbance to the power system assets, and obtains the first lag time when the power system assets recover to a level of one-quarter of the original operating state, that is, the performance of the power system assets as a whole reaches 25% of the original normal operation. Continue to detect, and obtain the second lag time when the power system assets recover to a level of one-half of the original operating state, that is, the performance of the power system assets as a whole reaches 50% of the original normal operation. Continue to detect, and obtain the third lag time when the power system assets recover to a level of three-quarters of the original operating state, that is, the performance of the power system assets as a whole reaches 75% of the original normal operation. Continue to detect, and obtain the fourth lag time when the power system assets recover to the original operating state.

[0080] Furthermore, the risk prediction system calculates the average value according to the first lag time, the second lag time, the third lag time and the fourth lag time to obtain the lag recovery time for the power system assets to recover to the original operating state.

[0081] Step 2052, based on the disturbance response intensity, propagation path complexity and disturbance fluctuation degree of the power system assets, determine the risk vulnerability degree of the power system assets to each factor combination.

[0082] Furthermore, the risk prediction system determines the risk vulnerability of power system assets to each combination of factors based on the disturbance response intensity, propagation path complexity and disturbance fluctuation degree of the power system assets. , among which, the risk vulnerability The specific formula is as follows: .

[0083] Step 2053, based on the risk vulnerability degree and delayed recovery time of the power system assets when each factor combination occurs, determine the risk sensitivity degree of the power system assets when each factor combination occurs.

[0084] Furthermore, the risk prediction system determines the risk sensitivity of power system assets to each factor combination according to the risk vulnerability and delayed recovery time of power system assets to each factor combination. , where risk sensitivity The specific formula is as follows: .

[0085] in, Indicates the risk vulnerability Combination of factors The partial derivative of , by taking the partial derivative, we can get the sensitivity of risk vulnerability to the change of factor combination; Indicates the hysteresis recovery time.

[0086] Step 2054 determines the comprehensive risk index of each factor combination based on the risk vulnerability and risk sensitivity of each factor combination.

[0087] Furthermore, the risk prediction system determines the comprehensive risk index of each factor combination based on the risk vulnerability and risk sensitivity of each factor combination. , the specific formula is as follows: .

[0088] Step 2055: determine a factor combination whose comprehensive risk index is greater than or equal to a preset index threshold as a risk factor combination, and determine the factors in the risk factor combination as key risk factors.

[0089] Furthermore, the risk prediction system determines a factor combination whose comprehensive risk index is greater than or equal to a preset index threshold as a risk factor combination, and determines the factors in the risk factor combination as key risk factors, wherein the preset index threshold is set according to actual conditions.

[0090] The embodiment of the present invention accurately determines the key risk factors and avoids the interference of irrelevant factors. Therefore, it is possible to accurately construct risk scenarios based on the key risk factors to predict the risk factors affecting the power system assets, thereby improving the comprehensiveness and accuracy of the prediction of asset risks.

[0091] In one embodiment, the description of steps 401 to 404 is as follows: Step 401 : for each risk scenario, determine the repair and regulation capability of the power grid system based on the historical power grid fault data of the power system assets.

[0092] Optionally, the risk prediction system collects historical grid fault data of power system assets, which includes fault type, fault repair time, time and location of fault occurrence, type of power system assets involved (such as transformers, lines, switches, etc.), fault cause, and impact of the fault on grid operation (such as power outage scope, voltage fluctuation, etc.).

[0093] Furthermore, for each risk scenario, the risk prediction system analyzes the historical fault data of the power grid to determine the repair and adjustment capabilities of the power grid system when facing faults in various risk scenarios.

[0094] In one embodiment, the repair and adjustment capability can be the average time from the occurrence of a fault to its repair, or can be an indicator of the ability of the power grid system to maintain a certain power supply level during the repair process, etc. Therefore, the risk prediction system analyzes the historical fault data of the power grid, and calculates the average time from the occurrence of different types of faults to their repair, or an indicator of the ability of the system to maintain a certain power supply level during the repair process.

[0095] Furthermore, for each risk scenario, the risk prediction system analyzes the historical fault data of the power grid to obtain the number of times the power system assets fail in the historical data under each risk scenario, as well as the total operating time of the power system assets under each risk scenario.

[0096] Furthermore, the risk prediction system calculates the probability of failure of power system assets under each risk scenario based on the number of failures and the total operating time of power system assets under each risk scenario, that is, the probability of failure = number of failures / total operating time.

[0097] Furthermore, for each risk scenario, the risk prediction system analyzes the historical fault data of the power grid to obtain interference indicators such as power outage loss, number of power outage users, voltage over-limit amplitude and power flow overload degree of the power grid system when the power system assets fail. According to each interference indicator and its corresponding impact weight, the interference impact of the power system asset failure on the power grid system is calculated, which will not be repeated here.

[0098] Step 402: correct the interference impact degree of the power system asset failure on the power grid system based on the repair and regulation capability to obtain a corrected interference degree.

[0099] Furthermore, the repair and adjustment capability in the embodiment of the present invention is the average time from the occurrence of a fault to the repair. Therefore, the risk prediction system corrects the interference impact of the power system asset fault on the power grid system by the average time from the occurrence of the fault to the repair, and obtains the corrected interference degree. The specific formula is: .

[0100] in, Indicates the interference degree after correction, Indicates the degree of interference impact, Indicates the average time from the occurrence of a fault to its repair, in milliseconds; represents the hyperbolic tangent function, when When it is close to 0.1, the function value is close to 0. The attenuation effect is small; when When the deviation from 0.1 is large, the function value approaches ,right The attenuation effect is enhanced, achieving reasonable correction of different interference impact levels.

[0101] Step 403, determining the initial risk intensity under each risk scenario based on the corrected interference level and the probability of failure of power system assets.

[0102] Furthermore, the risk prediction system determines the initial risk intensity under each risk scenario based on the corrected interference level and the probability of failure of power system assets. The specific formula is as follows: .

[0103] in, represents the initial risk intensity, Represents the probability of a power system asset failing.

[0104] Step 404 , each risk scenario is evaluated based on the change trend index of each risk scenario over time combined with the initial risk intensity to obtain the risk level of each risk scenario.

[0105] Furthermore, for each risk scenario, the risk prediction system monitors the changes in the operating parameters of the power grid system over time, and determines the change trend index of each risk scenario over time based on the changes in the operating parameters of the power grid system over time. The trend indicator of change is , the value range is .

[0106] Furthermore, the risk prediction system evaluates each risk scenario based on the change trend index of each risk scenario over time combined with the initial risk intensity to obtain the risk level of each risk scenario, as shown in steps 4041 to 4043.

[0107] The embodiment of the present invention simulates the operation of power system assets under each risk scenario and accurately obtains the risk level of each risk scenario. Therefore, it is able to accurately predict the risk level of power system asset failure based on the matching results of the environmental meteorological data of the power system assets and the risk scenario, thereby improving the comprehensiveness and accuracy of asset risk prediction, enabling power operation and maintenance personnel to more clearly grasp the risk situation of power system assets and achieve more effective operation, maintenance and management of the power system.

[0108] In one embodiment, the description of steps 4041 to 4043 is as follows: Step 4041, determine the risk adjustment factor based on the change trend indicator of each risk scenario over time.

[0109] Optionally, considering that the risk scenario changes over time and has a certain impact on the risk intensity, the risk prediction system determines the risk adjustment factor according to the change trend indicator of each risk scenario over time, where the specific formula of the risk adjustment factor is: .

[0110] in, represents the risk adjustment factor, Indicates that the value of the cosine function is adjusted according to the changing trend indicator. When the value is -1, it means the risk scenario is deteriorating. = -1, the value is 1, indicating that the risk scenario has improved; Over time As time goes by, the dynamic adjustment range of risk intensity is gradually adjusted to reflect the cumulative impact of time on changes in risk scenarios.

[0111] Step 4042, perform a comprehensive risk intensity calculation based on the risk intensity adjustment factor and initial risk intensity of each risk scenario to obtain a target risk intensity for each risk scenario.

[0112] Furthermore, the risk prediction system performs a comprehensive risk intensity calculation based on the risk intensity adjustment factor and initial risk intensity of each risk scenario to obtain the target risk intensity of each risk scenario. , where target risk intensity The specific formula is: .

[0113] Step 4043, performing level mapping based on the target risk intensity of each risk scenario to obtain the risk level of each risk scenario.

[0114] Optionally, the embodiment of the present invention pre-sets a series of risk level interval boundary values, such as ,like The specific risk level interval boundary value is determined according to the actual system situation and experience.

[0115] Therefore, the risk prediction system maps the target risk intensity of each risk scenario to a level according to the boundary value of the risk level interval, and obtains the risk level of each risk scenario. , the specific formula is: . .

[0116] Indicates rounding down. express The parameter variables, Indicates the number of risk level interval boundary values, Indicates the first risk level interval boundary value, Indicates The boundary value of the risk level interval.

[0117] The embodiment of the present invention accurately obtains the risk level of each risk scenario, and can therefore accurately predict the risk level of power system asset failure based on the matching results of the environmental meteorological data of the power system assets and the risk scenarios, thereby improving the comprehensiveness and accuracy of asset risk prediction, enabling power operation and maintenance personnel to more clearly grasp the risk situation of power system assets and achieve more effective operation, maintenance and management.

[0118] Furthermore, the power system asset risk prediction system based on big data analysis provided by the present invention is described below. The power system asset risk prediction system based on big data analysis described below and the power system asset risk prediction method based on big data analysis described above can be referenced to each other.

[0119] Reference Figure 2 , Figure 2 It is a structural schematic diagram of the power system asset risk prediction system based on big data analysis provided by the present invention. The power system asset risk prediction system based on big data analysis includes.

[0120] The correlation analysis module 210 is used to perform correlation analysis on each data in the power data set to determine the potential correlation between the equipment operation status and the environmental meteorological data; The risk factor screening module 220 is used to analyze the risk impact of environmental meteorological factors on power system assets based on potential correlation relationships and determine key risk factors; A risk scenario construction module 230 is used to construct risk scenarios based on key risk factors; each risk scenario represents a combination of key risk factors at different levels; The risk scenario simulation module 240 is used to simulate the operation of power system assets under each risk scenario, and determine the risk level of each risk scenario based on the probability of failure of power system assets under each risk scenario and the degree of interference of power system asset failure on the power grid system; The risk prediction module 250 is used to match the environmental meteorological data of the power system assets monitored in real time with the risk scenario, and determine the risk level of the matched target risk scenario as the risk level of the power system asset failure.

[0121] The embodiment of the present invention analyzes the potential correlation between the equipment operation status and environmental factors and meteorological factors through correlation, so that the potential correlation can be fully utilized to screen out the key risk factors with greater risk impact on the power system assets, avoiding the interference of irrelevant factors. Further, the operation of the power system assets is simulated under each risk scenario to accurately obtain the risk level of each risk scenario. In the subsequent risk prediction process of the power system assets, the risk level of the power system asset failure can be accurately predicted based on the matching results of the environmental meteorological data of the power system assets and the risk scenario. Therefore, the embodiment of the present invention can comprehensively consider various risk factors affecting the power system assets for risk prediction, improve the comprehensiveness and accuracy of the prediction of asset risks, and enable power operation and maintenance personnel to more clearly grasp the risk situation of the power system assets and realize more effective operation and management of the power system.

[0122] See also Figure 3 , Figure 3 FIG. 1 is an embodiment diagram of an electronic device provided by an embodiment of the present invention. Figure 3 As shown, an embodiment of the present invention provides an electronic device 300, including a memory 310, a processor 320, and a computer program 311 stored in the memory 310 and executable on the processor 320. When the processor 320 executes the computer program 311, the following steps are implemented: Conduct correlation analysis on various data in the power data set to determine the potential correlation between equipment operation status and environmental meteorological data; Analyze the risk impact of environmental meteorological factors on power system assets based on potential correlations and identify key risk factors; Construct risk scenarios based on key risk factors; each risk scenario represents a combination of key risk factors at different levels; Simulate the operation of power system assets under each risk scenario, and determine the risk level of each risk scenario based on the probability of power system asset failure under each risk scenario and the degree of interference of power system asset failure on the power grid system; The environmental meteorological data of the power system assets monitored in real time are matched with the risk scenarios, and the risk level of the matched target risk scenarios is determined as the risk level of failure of the power system assets.

[0123] See also Figure 4 , Figure 4 Detailed description of an embodiment of a computer-readable storage medium provided by an embodiment of the present invention. Figure 4 As shown, this embodiment provides a computer-readable storage medium 400, on which a computer program 311 is stored. When the computer program 311 is executed by a processor, the following steps are implemented: Conduct correlation analysis on various data in the power data set to determine the potential correlation between equipment operation status and environmental meteorological data; Analyze the risk impact of environmental meteorological factors on power system assets based on potential correlations and identify key risk factors; Construct risk scenarios based on key risk factors; each risk scenario represents a combination of key risk factors at different levels; Simulate the operation of power system assets under each risk scenario, and determine the risk level of each risk scenario based on the probability of power system asset failure under each risk scenario and the degree of interference of power system asset failure on the power grid system; The environmental meteorological data of the power system assets monitored in real time are matched with the risk scenarios, and the risk level of the matched target risk scenarios is determined as the risk level of failure of the power system assets.

[0124] On the other hand, the present invention further provides a computer program product, the computer program product includes a computer program, the computer program can be stored on a non-transitory computer readable storage medium, when the computer program is executed by a processor, the computer can perform the following steps: Conduct correlation analysis on various data in the power data set to determine the potential correlation between equipment operation status and environmental meteorological data; Analyze the risk impact of environmental meteorological factors on power system assets based on potential correlations and identify key risk factors; Construct risk scenarios based on key risk factors; each risk scenario represents a combination of key risk factors at different levels; Simulate the operation of power system assets under each risk scenario, and determine the risk level of each risk scenario based on the probability of power system asset failure under each risk scenario and the degree of interference of power system asset failure on the power grid system; The environmental meteorological data of the power system assets monitored in real time are matched with the risk scenarios, and the risk level of the matched target risk scenarios is determined as the risk level of failure of the power system assets.

[0125] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, by hardware. Based on this understanding, the above technical solution can be essentially or in other words, the part that contributes to the prior art can be embodied in the form of a software product, which can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the method described in each embodiment or some parts of the embodiment.

[0126] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for predicting power system asset risks based on big data analysis, characterized in that: include: Conduct correlation analysis on various data in the power data set to determine the potential correlation between equipment operation status and environmental meteorological data; Analyze the risk impact of environmental meteorological factors on power system assets based on the potential correlation relationship and determine key risk factors; Construct risk scenarios based on the key risk factors; Each risk scenario represents a combination of key risk factors at different levels; Simulate the operation of power system assets under each risk scenario, and determine the risk level of each risk scenario based on the probability of power system asset failure under each risk scenario and the degree of interference of power system asset failure on the power grid system; The environmental meteorological data of the power system assets monitored in real time are matched with the risk scenarios, and the risk level of the matched target risk scenarios is determined as the risk level of failure of the power system assets.

2. The power system asset risk prediction method based on big data analysis according to claim 1 is characterized in that: The correlation analysis of each data in the power data set to determine the potential correlation between the equipment operation status and the environmental meteorological data includes: The equipment operation status information, environmental factor data and meteorological factor data in the power data set are mapped into a multi-dimensional space to construct an initial three-dimensional association matrix; each matrix element in the three-dimensional association matrix represents the ratio between the number of data instances containing the equipment operation status information, environmental factor data and meteorological factor data at the same time and the total number of data instances in the power data set; Based on a preset threshold and in combination with each matrix element in the initial three-dimensional correlation matrix, the initial three-dimensional correlation matrix is ​​screened to obtain a screened three-dimensional correlation matrix; Based on a preset time window, each matrix element in the filtered three-dimensional association matrix is ​​updated to obtain an updated three-dimensional association matrix; each matrix element in the updated three-dimensional association matrix represents the ratio between the number of data instances that simultaneously contain equipment operation status information, environmental factor data, and meteorological factor data within the preset time window and the total number of data instances within the preset time window; Based on the screened three-dimensional correlation matrix and the updated three-dimensional correlation matrix, potential correlations between the equipment operating status and environmental factors and meteorological factors are determined.

3. The power system asset risk prediction method based on big data analysis according to claim 2 is characterized in that: The determining of the potential correlation between the equipment operation status and the environmental factors and the meteorological factors based on the screened three-dimensional correlation matrix and the updated three-dimensional correlation matrix includes: Based on each matrix element in the three-dimensional association matrix after the screening, the power data set is screened to obtain a preliminary association candidate set; For each item set in the preliminary association candidate set, a frequency score of each item set is obtained based on the number of occurrences of the item set combined with the first matrix element of the item set in the three-dimensional association matrix after screening and the second matrix element of the three-dimensional association matrix after updating; The item sets whose frequency scores are greater than or equal to the preset minimum support threshold are determined as frequent item sets; For any two target frequent item sets in the frequent item sets, association rules are generated according to the item set causal relationship and item set influence degree between the target frequent item sets, and the association rules are determined as the potential association relationship between the equipment operation status and the environmental factors and meteorological factors.

4. The power system asset risk prediction method based on big data analysis according to claim 1 is characterized in that: Analyzing the risk impact of environmental meteorological factors on power system assets based on the potential correlation relationship and determining key risk factors include: Determine target environmental factors and target meteorological factors that affect the equipment operating status of power system assets based on the potential association relationship; For each factor combination of the target environmental factor and / or the target meteorological factor, when each factor combination occurs, if a disturbance occurs to the power system asset, the disturbance response intensity when the power system asset is disturbed is determined according to the magnitude of change in the operating state of the power system asset under the action of the factor combination; Based on the network topology of the power grid system, determine the number of nodes, the number of edges, the coupling relationship between nodes, and the coupling relationship between edges in the path through which the disturbance propagates, and determine the complexity of the propagation path when a disturbance occurs in the power system assets; Determining the degree of disturbance fluctuation when a disturbance occurs to the power system assets based on the gradient of the operating state change of the power system assets in each local area of ​​the power grid system; The key risk factors are determined based on the disturbance response intensity, propagation path complexity and disturbance fluctuation degree of the power system assets when disturbance occurs under each combination of factors.

5. The power system asset risk prediction method based on big data analysis according to claim 4 is characterized in that: The key risk factors are determined based on the disturbance response intensity, propagation path complexity and disturbance fluctuation degree of the power system assets when disturbance occurs under each combination of factors, including: For each factor combination, simulate the recovery of the disturbance to the power system assets and obtain the hysteresis recovery time for the power system assets to return to the original operating state; Based on the disturbance response intensity, propagation path complexity and disturbance fluctuation degree of power system assets, determine the risk vulnerability of power system assets to each combination of factors; Determine the risk sensitivity of power system assets to each factor combination based on the risk vulnerability and delayed recovery time of power system assets to each factor combination; Determine the comprehensive risk index of each factor combination based on the risk vulnerability and risk sensitivity of each factor combination; A factor combination whose comprehensive risk index is greater than or equal to a preset index threshold is determined as a risk factor combination, and the factors in the risk factor combination are determined as the key risk factors.

6. The power system asset risk prediction method based on big data analysis according to claim 1 is characterized in that: Based on the probability of power system asset failure in each risk scenario and the degree of interference of power system asset failure on the power grid system, the risk level of each risk scenario is determined, including: For each risk scenario, determining the repair and regulation capability of the power grid system based on historical power grid fault data of power system assets; Based on the repair and regulation capability, the interference degree of the power system asset failure on the power grid system is corrected to obtain a corrected interference degree; Determine the initial risk intensity for each risk scenario based on the corrected disturbance level and probability of failure of power system assets; Each risk scenario is evaluated based on the changing trend indicators of each risk scenario over time combined with the initial risk intensity to obtain the risk level of each risk scenario.

7. The method for predicting power system asset risk based on big data analysis according to claim 6 is characterized in that: The risk level of each risk scenario is obtained by evaluating each risk scenario based on the change trend indicator of each risk scenario over time combined with the initial risk intensity, including: Determine the risk adjustment factor based on the trend indicators of each risk scenario over time; Perform a comprehensive risk intensity calculation based on the risk intensity adjustment factor and initial risk intensity of each risk scenario to obtain the target risk intensity of each risk scenario; Based on the target risk intensity of each risk scenario, level mapping is performed to obtain the risk level of each risk scenario.

8. A power system asset risk prediction system based on big data analysis, characterized in that: Applied to the power system asset risk prediction method based on big data analysis as described in any one of claims 1 to 7; The power system asset risk prediction system based on big data analysis includes: The correlation analysis module is used to perform correlation analysis on each data in the power data set to determine the potential correlation between the equipment operation status and the environmental meteorological data; A risk factor screening module is used to analyze the risk impact of environmental meteorological factors on power system assets based on the potential correlation relationship and determine key risk factors; A risk scenario construction module, used to construct risk scenarios based on the key risk factors; each risk scenario represents a combination of key risk factors at different levels; The risk scenario simulation module is used to simulate the operation of power system assets under each risk scenario, and determine the risk level of each risk scenario based on the probability of power system asset failure under each risk scenario and the degree of interference of power system asset failure on the power grid system; The risk prediction module is used to match the environmental meteorological data of the power system assets monitored in real time with the risk scenario, and determine the risk level of the matched target risk scenario as the risk level of the power system asset failure.

9. An electronic device, comprising: Memory for storing computer software programs; A processor, used to read and execute the computer software program, characterized in that when the computer software program is executed by the processor, it implements the power system asset risk prediction method based on big data analysis as described in any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium having a computer software program stored therein, characterized in that: When the computer software program is executed by a processor, it implements the power system asset risk prediction method based on big data analysis as described in any one of claims 1 to 7.

Images