Risk assessment method, device and equipment of secondary equipment, storage medium and program product

Abstract

This application relates to a risk assessment method, apparatus, device, storage medium, and program product for secondary equipment. Based on the fluctuations and trends of alarm frequency sequences associated with the inherent risk factors of the target equipment, the risk index of the inherent risk factors of the target equipment at the current moment is calculated. Based on the risk index of the target factors, the dynamic conditional probability of the target factors is determined using a logistic regression model. The target factors are inherent risk factors with parent nodes in a dynamic Bayesian network. The dynamic conditional probability is input into the dynamic Bayesian network, and the systemic risk probability of the target factors is determined through a Bayesian inference algorithm. Based on the systemic risk probability of the target factors, their operational impact factors, economic impact factors, and safety impact factors, the risk assessment result of the secondary equipment is determined using a comprehensive risk index expression. This improves the accuracy of the obtained risk assessment results for secondary equipment.

Description

Technical Field

[0001] This application relates to the field of risk assessment technology, and in particular to a risk assessment method, apparatus, equipment, storage medium, and program product for secondary equipment. Background Technology

[0002] Secondary equipment is a key component in ensuring the safe and stable operation of complex industrial systems such as power systems, and its risk management is directly related to the reliability and security of the entire system. Therefore, accurate and dynamic risk assessment and management of secondary equipment is one of the core tasks of operation and maintenance work, and is of great significance for early detection of potential hazards and prevention of escalation of faults.

[0003] However, traditional risk assessment methods for secondary equipment typically rely solely on static analysis of the equipment's own monitoring data, such as analyzing single alarm messages or equipment ledgers. Therefore, the accuracy of the resulting risk assessments for secondary equipment is often low. Summary of the Invention

[0004] Therefore, it is necessary to provide a method, apparatus, device, storage medium, and program product for risk assessment of secondary equipment that can improve the accuracy of the risk assessment results obtained from the aforementioned technical problems.

[0005] Firstly, this application provides a risk assessment method for secondary equipment, including:

[0006] For each secondary device in the power system, obtain the alarm frequency sequence associated with the external operating condition information of the secondary device and the inherent risk factors of the secondary device within the most recent preset time period;

[0007] Based on the fluctuation and trend of the alarm frequency sequence associated with the inherent risk factors of the target equipment, calculate the risk index of the inherent risk factors of the target equipment at the current moment; the target equipment is any one of the secondary equipment.

[0008] Based on the risk index of the target factors of the target equipment, the dynamic conditional probability of the target factors is determined by a logistic regression model; the target factors are inherent risk factors with parent nodes in a dynamic Bayesian network; the dynamic Bayesian network is constructed with all inherent risk factors of the target equipment as the first node and all external operating condition information of the target equipment as the second node;

[0009] The dynamic conditional probability is input into the dynamic Bayesian network, and the system risk probability of the target factor is determined through the Bayesian inference algorithm.

[0010] Based on the systemic risk probability of the target factors, the operational impact factors, economic impact factors, and safety impact factors of the target factors, the risk assessment results of secondary equipment are determined using the comprehensive risk index expression.

[0011] In one embodiment, inherent risk factors include historical defects in the secondary equipment and discrepancies between the design parameters of the secondary equipment and the test data obtained from the actual application of the secondary equipment.

[0012] In one embodiment, based on the fluctuations and trends of the alarm frequency sequence associated with the inherent risk factors of the target device, the risk index of the inherent risk factors of the target device at the current moment is calculated, including:

[0013] Using any moment in the alarm frequency sequence as a reference point, obtain the first alarm frequency subsequence within a preset time window corresponding to the reference point of the alarm frequency sequence;

[0014] Calculate the slope of the second alarm frequency subsequence corresponding to time s under the first alarm frequency subsequence, and the variance of the second alarm frequency subsequence corresponding to time s. Also calculate the slope information entropy of the slope of the second alarm frequency subsequence corresponding to time s, and the variance information entropy of the variance of the second alarm frequency subsequence corresponding to time s.

[0015] The risk index is calculated using the following first objective formula:

[0016] ;

[0017] in, , F represents the risk index; Indicates the preset time window; This represents the time weighting function; Represents the slope information entropy weight, Represents the variance information entropy weights; The slope of the second alarm frequency subsequence corresponding to time s, Let represent the variance of the second alarm frequency subsequence corresponding to time s; Represents slope information entropy, This represents the variance information entropy.

[0018] In one embodiment, the logistic regression model is represented by the following second objective formula:

[0019] ;

[0020] in, Represents the target factor At any moment The dynamic conditional probability; Represents the target factor At any moment The risk index; It is the time when all parent nodes of the target factor are at time. A set of risk indices; It is the second node connected to the target factor at time [time]. A set of risk indices; It is a bias term; , These are the regression coefficients of the risk index of the parent node and the risk index of the second node, respectively. (Time interval not specified) Indicates the current moment.

[0021] In one embodiment, the expression for the comprehensive risk index is:

[0022] ;

[0023] in, Represents the target factor In the The overall risk index at any given moment; , , These are the target factors The corresponding operational impact factors, economic impact factors, and safety impact factors, , , These are the target factors The corresponding historical averages of historical operational impact factors, historical economic impact factors, and historical safety impact factors.

[0024] In one embodiment, the method further includes:

[0025] Based on the causal structure of dynamic Bayesian networks, the intervention quantifier is used to extrapolate each intervention measure in the set of intervention measures, calculate the overall comprehensive risk index of the system after each intervention measure is taken, and determine the intervention measure corresponding to the smallest overall comprehensive risk index as the target intervention measure.

[0026] Secondly, this application also provides a risk assessment device for secondary equipment, comprising:

[0027] The acquisition module is used to acquire, for each secondary device in the power system, the external operating condition information of the secondary device within the most recent preset time period and the alarm frequency sequence associated with the inherent risk factors of the secondary device.

[0028] The first calculation module is used to calculate the risk index of the inherent risk factors of the target device at the current moment based on the fluctuation and trend of the alarm frequency sequence associated with the inherent risk factors of the target device; the target device is any device among the secondary devices.

[0029] The first determination module is used to determine the dynamic conditional probability of the target factor based on the risk index of the target factor of the target device and a logistic regression model. The target factor is an inherent risk factor with a parent node in a dynamic Bayesian network. The dynamic Bayesian network is constructed with all inherent risk factors of the target device as the first node and all external operating condition information of the target device as the second node.

[0030] The second determination module is used to input the dynamic conditional probability into the dynamic Bayesian network and determine the system risk probability of the target factor through the Bayesian inference algorithm.

[0031] The third determination module is used to determine the risk assessment results of secondary equipment based on the system risk probability of the target factor, the operational impact factor, the economic impact factor, and the safety impact factor of the target factor, using a comprehensive risk index expression.

[0032] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of any of the above methods.

[0033] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any of the above methods.

[0034] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of any of the above methods.

[0035] The aforementioned risk assessment methods, devices, equipment, storage media, and program products for secondary equipment calculate the risk index of the inherent risk factors of the target equipment at the current moment by analyzing the fluctuations and trends of the alarm frequency sequence associated with the inherent risk factors of the target equipment. Based on the risk index of the target factors, the dynamic conditional probability of the target factors is determined using a logistic regression model. The target factors are inherent risk factors with parent nodes in a dynamic Bayesian network. The dynamic conditional probability is input into the dynamic Bayesian network, and the systemic risk probability of the target factors is determined through a Bayesian inference algorithm. Based on the systemic risk probability of the target factors, their operational impact factors, economic impact factors, and safety impact factors, the risk assessment result of the secondary equipment is determined using a comprehensive risk index expression. By considering the inherent correlations between various inherent risk factors and the external operating conditions of the secondary equipment, the accuracy of the obtained risk assessment results for the secondary equipment is improved. Attached Figure Description

[0036] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0037] Figure 1 This is a flowchart illustrating a risk assessment method for secondary equipment provided in an embodiment of this application;

[0038] Figure 2 This is a schematic diagram of the alarm frequency sequence of a circuit breaker provided in an embodiment of this application;

[0039] Figure 3 This is a schematic diagram of the structure of a dynamic Bayesian network provided in an embodiment of this application;

[0040] Figure 4 This is a flowchart illustrating a method for determining the risk index of an inherent risk factor provided in an embodiment of this application;

[0041] Figure 5 This is a risk assessment comparison chart of a circuit breaker provided in an embodiment of this application;

[0042] Figure 6 This is a schematic diagram of the structure of a risk assessment device for secondary equipment provided in an embodiment of this application;

[0043] Figure 7 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0044] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0045] It should be noted that the terms "first," "second," etc., used in this application can be used to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish the first element from the second element. The terms "comprising" and "having," and any variations thereof, used in this application, are intended to cover non-exclusive inclusion. The term "multiple" used in this application refers to two or more. The term "and / or" used in this application refers to one of the embodiments, or any combination of multiple embodiments.

[0046] Secondary equipment is a key component in ensuring the safe and stable operation of complex industrial systems such as power systems, and its risk management is directly related to the reliability and security of the entire system. Therefore, accurate and dynamic risk assessment and management of secondary equipment is one of the core tasks of operation and maintenance work, and is of great significance for early detection of potential hazards and prevention of escalation of faults.

[0047] However, traditional risk assessment methods for secondary equipment typically rely solely on static analysis of the equipment's own monitoring data, such as analyzing single alarm messages or equipment ledgers. This approach ignores the inherent relationships between different risk factors and fails to fully consider the dynamic impact of external operating scenarios, such as severe weather and grid load, on the evolution of equipment risk. Therefore, its assessment results are often one-sided and fail to accurately reflect the true risk level of the equipment under specific complex scenarios, resulting in relatively low accuracy in the risk assessment results for secondary equipment.

[0048] To address the aforementioned technical problems, this application provides a risk assessment method for secondary equipment. The method is as follows: Figure 1 As shown, Figure 1 This is a flowchart illustrating a risk assessment method for secondary equipment provided in an embodiment of this application. The method includes the following steps:

[0049] S101: For each secondary device in the power system, obtain the alarm frequency sequence associated with the external operating condition information of the secondary device and the inherent risk factors of the secondary device within the most recent preset time period.

[0050] Inherent risk factors may include equipment defects in the secondary equipment, as well as discrepancies between the design parameters of the secondary equipment and the test data obtained from the actual application of the secondary equipment.

[0051] For each inherent risk factor of each secondary device, a preliminary risk score is obtained based on its records at different times. If all inherent risk factor records are numerical, all elements of the inherent risk factor are divided by its maximum value to obtain the preliminary risk score. If all inherent risk factor records are words, each word is assigned a value according to its severity to obtain the preliminary risk score. For example, taking the inherent risk factor "equipment defect" as an example, which includes three records: "general," "serious," and "urgent," "general" can be assigned a value of 0.3, "serious" a value of 0.6, and "urgent" a value of 0.9.

[0052] like Figure 2 As shown, Figure 2 This is a schematic diagram of the alarm frequency sequence of a circuit breaker provided in an embodiment of this application. The alarm frequency sequence of the circuit breaker includes the alarm frequency sequence of defects existing in the equipment defect database.

[0053] Multiple inherent risk factors for each secondary device can be obtained from the relevant business systems, along with the alarm frequency sequence associated with each inherent risk factor. Furthermore, multiple external operating condition information for each secondary device can be obtained from external information systems. This external operating condition information may include weather severity indices from meteorological APIs, system load levels from SCADA systems, and network security threat levels from network intrusion detection systems. For example, the equipment defect history for a circuit breaker shows a medium defect of "delayed opening and closing response time," and the meteorological API indicates a current weather severity index of "severe."

[0054] The preset duration is the duration that should be able to reflect the risk changes of secondary equipment. For example, the preset duration is 1 month. The business system is the power company's production management system, equipment asset management system, and dispatch automation system, etc. The external information system can be the National Meteorological Information Center, the power grid energy management system, etc. The inherent risk factors are automatically extracted from the structured data through the database interface.

[0055] S102, calculate the risk index of the inherent risk factor of the target equipment at the current moment based on the fluctuation and trend of the alarm frequency sequence associated with the inherent risk factor of the target equipment; the target equipment is any one of the secondary equipment.

[0056] In this embodiment, considering that the variance of the alarm frequency sequence can characterize the fluctuation of the sequence, and the linear regression result of the alarm frequency sequence can characterize the changing trend of the alarm frequency sequence, and based on this, since the importance of fluctuation and changing trend are different, their impact on the risk index of the inherent risk factor of secondary equipment is also different. Considering that information entropy is a characteristic that measures information uncertainty, the larger the information entropy of the alarm frequency sequence, the more chaotic and unreliable the corresponding information, while the smaller the information entropy, the more deterministic the sequence information is. Therefore, the corresponding sequence should be given a larger weight, so that the risk index of the inherent risk factor of the secondary equipment is more accurate in the final calculation. Therefore, this application further combines the variance, linear regression result, and information entropy of the alarm frequency sequence associated with the inherent risk factor of the secondary equipment to calculate the risk index of any inherent risk factor of any secondary equipment.

[0057] The alarm frequency sequence associated with any inherent risk factor of any secondary device can be denoted as the target sequence; taking any moment of the target sequence as a reference point, the alarm frequency subsequence corresponding to the reference point of the target sequence within a preset time window is obtained and denoted as the target subsequence; the trend slope and variance of the alarm frequency subsequence corresponding to each moment within the target subsequence are calculated, and the slope information entropy of the trend slope and the variance information entropy of the variance of the alarm frequency subsequence corresponding to each moment within the target subsequence are calculated; the risk index is calculated according to the following first objective formula:

[0058] ;

[0059] in, , F represents the risk index; Indicates the preset time window; This represents the time weighting function; Represents the slope information entropy weight, Represents the variance information entropy weights; The slope of the second alarm frequency subsequence corresponding to time s, Let represent the variance of the second alarm frequency subsequence corresponding to time s; The slope information entropy corresponding to time s is represented by... Let represent the variance information entropy at time s. The slope reflects short-term changes in alarm frequency, while the variance measures the fluctuations in alarm frequency.

[0060] For example, the preset time window includes five times: time 1, time 2, ..., time 5. These five times correspond to a first alarm frequency sequence, i.e., a target subsequence. Time 1 corresponds to a second alarm frequency sequence, time 2 corresponds to a second alarm frequency sequence, ..., time 5 corresponds to a second alarm frequency sequence, i.e., a total of five second alarm frequency subsequences. The slope and variance of the second alarm frequency subsequences corresponding to these five times can be calculated, as well as the slope information entropy of the slope and the variance information entropy of the variance of the second alarm frequency subsequences corresponding to these five times. The risk index is calculated by sampling the first target formula mentioned above.

[0061] It should be noted that the risk index can also be calculated based on a modified formula of the first objective formula. For example, the risk index can be obtained by multiplying the formula on the right side of the equal sign in the first objective formula by a first preset coefficient.

[0062] The time weighting function can be a Gaussian function, which makes the feature contribution of the closer the time point is to the time, so that the risk assessment focuses more on the recent state.

[0063] In the formula, This means that the trend slope and variance information entropy of the alarm frequency subsequence corresponding to each time point within the target subsequence are used as the weighting basis. Data with greater information entropy has a smaller weight, and the corresponding data has higher reference value, thus contributing more to the risk index of the inherent risk factor of secondary equipment. This means that time weighting is used as a coefficient, so that the trend slope and variance of the time point closer to the reference point contribute more to the risk index. Therefore, the closer the time point is to the reference point, the greater the trend slope and variance, and the greater the risk index of the inherent risk factor of the secondary equipment.

[0064] S103, Based on the risk index of the target factor of the target equipment, determine the dynamic conditional probability of the target factor using a logistic regression model.

[0065] The target factor is an inherent risk factor with a parent node in the dynamic Bayesian network; the dynamic Bayesian network is constructed with all inherent risk factors of the target device as the first node and all external operating condition information of the target device as the second node.

[0066] Considering that Bayesian networks can effectively model causal relationships between variables, this invention uses Bayesian networks as the basic framework. Since the conditional probability of traditional Bayesian networks is static, this invention introduces external working condition information as a regulating factor to construct a causal risk network that can adapt to the scenario.

[0067] Specifically, any secondary device is designated as the target device, and all inherent risk factors of the target device are designated as nodes. A Bayesian network graph of the target device is constructed, with all external operating condition information as special nodes in the Bayesian network graph. Based on expert knowledge and a data-driven causal discovery algorithm, all special nodes are connected in the Bayesian network graph. It should be noted that causal discovery algorithms such as the PC algorithm and the LiNGAM model are existing technologies. For example, a directed edge can be established between the two inherent risk factor nodes, "non-standard loop information" and "equipment defect history," while the external operating condition information node "system load level" and the external operating condition information node "weather severity" are both designated as parent nodes of the "equipment defect history" node. Figure 3 As shown, Figure 3 This is a schematic diagram of a dynamic Bayesian network provided in an embodiment of this application. The consistency deviation between the design parameters of the secondary equipment and the detection data obtained from the secondary equipment in actual application refers to the discrepancy between these two sets of data.

[0068] To enable external operating condition information to dynamically adjust the strength of causal relationships among risk factors, this invention modifies the traditional static conditional probability table into a dynamic function. Considering that the logistic regression function can map any linear combination of inputs to a probability interval of (0,1), it is suitable for modeling conditional probabilities. Therefore, this invention combines the logistic regression function to establish dynamic conditional probabilities.

[0069] Preferably, any inherent risk factor with a parent node in the Bayesian network graph is denoted as the target factor. A logistic regression model for the target factor is established. The logistic regression model takes the risk index of all parent nodes of the target factor and the risk index of all connected external working condition information as input, and calculates the dynamic conditional probability value based on the real-time state value as the output.

[0070] The logistic regression model is expressed by the following formula for the second objective:

[0071] ;

[0072] in, Represents the target factor At any moment The dynamic conditional probability; Represents the target factor At any moment The risk index; It is the time when all parent nodes of the target factor are at time. A set of risk indices; It is the second node connected to the target factor at time [time]. A set of risk indices; It is a bias term; , These are the regression coefficients of the risk index of the parent node and the risk index of the second node, respectively. (Time interval not specified) Indicates the current moment. This represents the normalization function.

[0073] It should be noted that the dynamic conditional probability can also be calculated based on a modified version of the second objective formula. For example, the product of the formula on the right side of the equal sign in the second objective formula and a second preset coefficient can be used as the dynamic conditional probability.

[0074] in, express The dynamic conditional probability depends on its parent node. and external operating condition information ; Represents a linear combination, where the parent node and external operating condition information The current state value is summed by weighted regression coefficients, which integrates the overall strength of the effects of all influencing factors; This means mapping the linear combination value to a range of 0 to 1, thus obtaining a dynamic conditional probability value that varies with the scenario.

[0075] S104 inputs the dynamic conditional probability into the dynamic Bayesian network, and determines the system risk probability of the target factor through the Bayesian inference algorithm.

[0076] Optionally, the risk indices of each inherent risk factor of each secondary device can be transformed into probability values ​​through a mapping function, and input into the Bayesian network graph as observation evidence at the current moment. The system risk probability of each inherent risk factor is then calculated using a Bayesian inference algorithm. It should be noted that this step transforms the risk indices calculated in S102 into evidence in a probabilistic form that the Bayesian network can understand. For example, a Sigmoid function can be used to map the risk indices to the (0,1) interval. Using a Bayesian inference algorithm, combining the posterior probabilities of each node at the previous moment and the observation evidence at the current moment, the system risk probability of each inherent risk factor at the current moment is inferred and calculated. The system risk probability refers to the posterior probability of the inherent risk factor occurring after considering the evidence from all other nodes in the network, the causal transmission relationships between nodes, and the dynamic influence of external operating conditions.

[0077] S105. Based on the systemic risk probability of the target factors, the operational impact factors, economic impact factors, and safety impact factors of the target factors, the risk assessment results of the secondary equipment are determined using the comprehensive risk index expression.

[0078] Based on the systemic risk probability, operational impact factors, economic impact factors, and safety impact factors of the target factors, a comprehensive risk index can be determined using a comprehensive risk index expression. Based on this comprehensive risk index, the risk assessment result for the secondary equipment is determined, which can indicate whether the secondary equipment faces risk or not. For example, if the sum of the comprehensive risk indices of all target factors is greater than or equal to a preset value, then the secondary equipment is determined to face risk; if the sum of the comprehensive risk indices of all target factors is less than the preset value, then the secondary equipment is determined to face no risk.

[0079] This application can introduce consequence quantification analysis and causal intervention-based virtual extrapolation decision-making to achieve support for risk-optimized decision-making, moving from probabilistic early warning to risk management. The essence of risk management is to prioritize threats that may cause serious consequences. This step aims to combine the probability and severity of failures to obtain a more comprehensive risk measurement. Considering the different dimensions of consequences, such as operational, economic, and safety dimensions, this invention normalizes the impact by comparing it with historical averages. Operational impacts include, for example, load shedding, and economic impacts include, for example, power outage losses.

[0080] Specifically, the operational, economic, and safety impacts of inherent risk factors are assessed, and the comprehensive risk index of each inherent risk factor for each secondary equipment is calculated.

[0081] For each inherent risk factor of each secondary device, assess the potential operational, economic, and safety impacts of its failure. It should be noted that, for example, the operational impact is defined as the expected load loss caused by the failure of a specific inherent risk factor, measured in megawatt-hours (MWh), reflecting the degree of impact of equipment failure on the core functions of the power grid. The economic impact is defined as direct economic loss, measured in monetary value, which comprehensively considers both direct costs (such as labor, spare parts, and equipment costs required for repair) and indirect costs (such as lost electricity sales revenue due to power outages, i.e., opportunity costs). Safety and environmental impacts are addressed by introducing a Risk Priority Number (RPN), which comprehensively assesses the potential risks of personal injury, penalties for violating environmental regulations, and damage to the company's reputation.

[0082] The composite risk index of the target factor satisfies the expression:

[0083] ;

[0084] In the formula, This represents the comprehensive risk index of the target factor at time t; This represents the dynamic conditional probability value of the target factor at time t; , , These are the operational, economic, and security impacts of the target factors; , , This represents the historical average of the operational, economic, and safety impacts.

[0085] In the formula, , , These represent the normalization of the operational, economic, and safety impacts of the target factor, respectively. The greater the operational, economic, and safety impacts of the target factor, and the greater the dynamic conditional probability value of the target factor at time t, the greater the comprehensive risk index of the target factor at time t.

[0086] In this embodiment, the risk index of the inherent risk factors of the target equipment is calculated at the current moment by analyzing the fluctuations and trends of the alarm frequency sequence associated with the inherent risk factors of the target equipment. Based on the risk index of the target factors, the dynamic conditional probability of the target factors is determined using a logistic regression model. The target factors are inherent risk factors with parent nodes in a dynamic Bayesian network. The dynamic conditional probability is input into the dynamic Bayesian network, and the systemic risk probability of the target factors is determined using a Bayesian inference algorithm. Based on the systemic risk probability of the target factors, their operational impact factors, economic impact factors, and safety impact factors, the risk assessment result of the secondary equipment is determined using a comprehensive risk index expression. By considering the inherent correlations between various inherent risk factors and the external operating conditions of the secondary equipment, the accuracy of the obtained risk assessment results for the secondary equipment is improved.

[0087] In one exemplary embodiment, inherent risk factors include historical defects of the secondary equipment and discrepancies between the design parameters of the secondary equipment and the test data obtained from the secondary equipment in actual application.

[0088] In this embodiment, by considering multiple inherent risk factors, the accuracy of the risk assessment results for the secondary equipment can be improved.

[0089] In one exemplary embodiment, such as Figure 4 As shown, Figure 4 This is a flowchart illustrating a method for determining the risk index of an inherent risk factor according to an embodiment of this application. The method includes the following steps:

[0090] S401, using any moment in the alarm frequency sequence as a reference point, obtain the first alarm frequency subsequence of the preset time window corresponding to the reference point of the alarm frequency sequence.

[0091] S402, calculate the slope of the second alarm frequency subsequence corresponding to time s under the first alarm frequency subsequence, and the variance of the second alarm frequency subsequence corresponding to time s, and calculate the slope information entropy of the slope of the second alarm frequency subsequence corresponding to time s, and the variance information entropy of the variance of the second alarm frequency subsequence corresponding to time s.

[0092] S403, the risk index is calculated using the following first objective formula.

[0093] ;

[0094] in, , F represents the risk index; Indicates the preset time window; This represents the time weighting function; Represents the slope information entropy weight, Represents the variance information entropy weights; The slope of the second alarm frequency subsequence corresponding to time s, Let represent the variance of the second alarm frequency subsequence corresponding to time s; Represents slope information entropy, This represents the variance information entropy.

[0095] In this embodiment, taking any moment of the alarm frequency sequence as a reference point, the first alarm frequency subsequence corresponding to the reference point of the alarm frequency sequence within a preset time window is obtained. The slope of the second alarm frequency subsequence corresponding to moment s under the first alarm frequency subsequence and the variance of the second alarm frequency subsequence corresponding to moment s are calculated. The slope information entropy of the slope of the second alarm frequency subsequence corresponding to moment s and the variance information entropy of the variance of the second alarm frequency subsequence corresponding to moment s are also calculated. The risk index is calculated using the following first objective formula, which improves the accuracy of risk index determination.

[0096] In one exemplary embodiment, the logistic regression model is represented by the following second objective formula:

[0097] ;

[0098] in, Represents the target factor At any moment The dynamic conditional probability; Represents the target factor At any moment The risk index; It is the time when all parent nodes of the target factor are at time. A set of risk indices; It is the second node connected to the target factor at time [time]. A set of risk indices; It is a bias term; , These are the regression coefficients of the risk index of the parent node and the risk index of the second node, respectively.

[0099] In this embodiment, the accuracy of the dynamic conditional probability of the target factor can be improved by using a logistic regression model.

[0100] In an exemplary embodiment, the comprehensive risk index is expressed as follows:

[0101] ;

[0102] in, Represents the target factor In the The overall risk index at any given moment; , , These are the target factors The corresponding operational impact factors, economic impact factors, and safety impact factors, , , These are the target factors The corresponding historical averages of historical operational impact factors, historical economic impact factors, and historical safety impact factors.

[0103] In this embodiment, the accuracy of the comprehensive risk index of the target factor can be improved by using a logistic regression model, thereby improving the accuracy of the risk assessment of the secondary equipment.

[0104] In one exemplary embodiment, the method further includes:

[0105] Based on the causal structure of dynamic Bayesian networks, the intervention quantifier is used to extrapolate each intervention measure in the set of intervention measures, calculate the overall comprehensive risk index of the system after each intervention measure is taken, and determine the intervention measure corresponding to the smallest overall comprehensive risk index as the target intervention measure.

[0106] This invention utilizes a causal network structure to virtually simulate different operational decisions, i.e., What-if analysis, thereby predicting consequences and selecting the optimal action before implementation. For example... Figure 5 As shown, Figure 5 This is a risk assessment comparison diagram of a circuit breaker provided in an embodiment of this application. It shows the comparison results of the risk assessment effect of circuit breaker A based on dynamic multi-source secondary equipment in this invention and the traditional risk assessment effect based on static single source secondary equipment.

[0107] Preferably, the causal structure of a dynamic Bayesian network is utilized, and an intervention predictor is introduced to realize intervention-based causal queries. This allows operations and maintenance personnel to quantitatively extrapolate and compare different handling plans before taking action. The intervention predictor is a do-operator. The selection of the optimal intervention measure can be achieved by solving the following optimization problem:

[0108] ;

[0109] In the formula, Indicates the optimal intervention measure; Represents a set of intervention measures; Indicates the simulated implementation of intervention measures The action; Indicates the implementation of intervention measures Then, the overall comprehensive risk index of the system; the overall comprehensive risk index of the system is defined as the sum of the comprehensive risk indices of all inherent risk factors.

[0110] The optimal intervention measure, which corresponds to the minimum overall comprehensive risk index, is taken as the target intervention measure. This target intervention measure is then distributed to the maintenance personnel's terminals to complete the management and control of secondary equipment risks.

[0111] It should be understood that although the steps in the flowcharts of the above embodiments are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the above embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages in other steps. It is understood that the steps in different embodiments can be freely combined as needed, and all non-contradictory solutions formed by such combinations are within the scope of protection of this application.

[0112] Based on the same inventive concept, this application also provides a risk assessment device for secondary equipment to implement the risk assessment method for secondary equipment described above. The solution provided by this device is similar to the solution described in the above method; therefore, the specific limitations in one or more embodiments of the risk assessment device for secondary equipment provided below can be found in the limitations of the risk assessment method for secondary equipment described above, and will not be repeated here.

[0113] In one exemplary embodiment, such as Figure 6 As shown, Figure 6This is a schematic diagram of the structure of a risk assessment device for secondary equipment provided in an embodiment of this application. The risk assessment device 600 includes:

[0114] The acquisition module 601 is used to acquire, for each secondary device in the power system, the external operating condition information of the secondary device within the most recent preset time period and the alarm frequency sequence associated with the inherent risk factors of the secondary device.

[0115] The first calculation module 602 is used to calculate the risk index of the inherent risk factor of the target device at the current moment based on the fluctuation and trend of the alarm frequency sequence associated with the inherent risk factor of the target device; the target device is any device among the secondary devices.

[0116] The first determining module 603 is used to determine the dynamic conditional probability of the target factor based on the risk index of the target factor of the target device and a logistic regression model. The target factor is an inherent risk factor with a parent node in a dynamic Bayesian network. The dynamic Bayesian network is constructed with all inherent risk factors of the target device as the first node and all external operating condition information of the target device as the second node.

[0117] The second determining module 604 is used to input the dynamic conditional probability into the dynamic Bayesian network and determine the system risk probability of the target factor through the Bayesian inference algorithm.

[0118] The third determining module 605 is used to determine the risk assessment result of secondary equipment based on the system risk probability of the target factor, the operational impact factor, the economic impact factor, and the safety impact factor of the target factor, using a comprehensive risk index expression.

[0119] In one exemplary embodiment, inherent risk factors include historical defects of the secondary equipment and discrepancies between the design parameters of the secondary equipment and the test data obtained from the secondary equipment in actual application.

[0120] In an exemplary embodiment, the first calculation module 602 is specifically configured to: obtain a first alarm frequency subsequence corresponding to a preset time window of the alarm frequency sequence, using any time point of the alarm frequency sequence as a reference point; calculate the slope of the second alarm frequency subsequence corresponding to time s under the first alarm frequency subsequence, and the variance of the second alarm frequency subsequence corresponding to time s; calculate the slope information entropy of the slope of the second alarm frequency subsequence corresponding to time s, and the variance information entropy of the variance of the second alarm frequency subsequence corresponding to time s; and calculate the risk index using the following first objective formula:

[0121] ;

[0122] in, , F represents the risk index; Indicates the preset time window; This represents the time weighting function; Represents the slope information entropy weight, Represents the variance information entropy weights; The slope of the second alarm frequency subsequence corresponding to time s, Let represent the variance of the second alarm frequency subsequence corresponding to time s; Represents slope information entropy, This represents the variance information entropy.

[0123] In one embodiment, the logistic regression model is represented by the following second objective formula:

[0124] ;

[0125] in, Represents the target factor At any moment The dynamic conditional probability; Represents the target factor At any moment The risk index; It is the time when all parent nodes of the target factor are at time. A set of risk indices; It is the second node connected to the target factor at time [time]. A set of risk indices; It is a bias term; , These are the regression coefficients of the risk index of the parent node and the risk index of the second node, respectively. (Time interval not specified) Indicates the current moment.

[0126] In one embodiment, the expression for the comprehensive risk index is:

[0127] ;

[0128] in, Represents the target factor In the The overall risk index at any given moment; , , These are the target factors The corresponding operational impact factors, economic impact factors, and safety impact factors, , , These are the target factors The corresponding historical averages of historical operational impact factors, historical economic impact factors, and historical safety impact factors.

[0129] In one exemplary embodiment, the risk assessment device 600 may further include:

[0130] The second calculation module is used to extrapolate each intervention measure in the set of intervention measures based on the causal structure of the dynamic Bayesian network, and to calculate the overall comprehensive risk index of the system after taking each intervention measure. The intervention measure corresponding to the smallest overall comprehensive risk index is determined as the target intervention measure.

[0131] Each module in the risk assessment device for the aforementioned secondary equipment can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0132] In one exemplary embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 7 As shown, the computer device includes a processor, memory, input / output interface, communication interface, display unit, and input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interface. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The input / output interface is used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, Near Field Communication (NFC), or other technologies. When executed by the processor, the computer program implements a risk assessment method for secondary devices. The display unit is used to form a visually visible image and can be a display screen, projection device, or virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0133] Those skilled in the art will understand that Figure 7The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0134] In one exemplary embodiment, a computer device is provided, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps of any of the method embodiments described above. The technical principles and effects are similar and will not be repeated here.

[0135] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored, which, when executed by a processor, implements the steps of any of the above method embodiments. The technical principles and effects are similar and will not be repeated here.

[0136] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps of any of the method embodiments described above. Its technical principles and effects are similar and will not be repeated here.

[0137] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0138] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0139] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0140] The above embodiments are merely illustrative of several implementation methods of this application, and their descriptions are relatively specific and detailed. However, they should not be construed as limiting the scope of this application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A risk assessment method for secondary equipment, characterized in that, The method includes: For each secondary device in the power system, obtain the alarm frequency sequence associated with the external operating condition information of the secondary device and the inherent risk factors of the secondary device within the most recent preset time period; Based on the fluctuation and trend of the alarm frequency sequence associated with the inherent risk factors of the target device, calculate the risk index of the inherent risk factors of the target device at the current moment; the target device is any one of the secondary devices. Based on the risk index of the target factor of the target device, the dynamic conditional probability of the target factor is determined based on a logistic regression model; the target factor is an inherent risk factor with a parent node in a dynamic Bayesian network; the dynamic Bayesian network is constructed with all the inherent risk factors of the target device as first nodes and all the external operating condition information of the target device as second nodes; The dynamic conditional probability is input into a dynamic Bayesian network, and the system risk probability of the target factor is determined through a Bayesian inference algorithm. Based on the system risk probability of the target factor, the operational impact factor, economic impact factor, and safety impact factor of the target factor, the risk assessment result of the secondary equipment is determined using a comprehensive risk index expression.

2. The method according to claim 1, characterized in that, The inherent risk factors include the historical defects of the secondary equipment and the consistency deviation between the design parameters of the secondary equipment and the test data obtained from the actual application of the secondary equipment.

3. The method according to claim 1 or 2, characterized in that, The calculation of the risk index of the inherent risk factors of the target equipment at the current moment, based on the fluctuation and trend of the alarm frequency sequence associated with the inherent risk factors of the target equipment, includes: Using any moment in the alarm frequency sequence as a reference point, obtain the first alarm frequency subsequence of the preset time window corresponding to the reference point in the alarm frequency sequence; Calculate the slope of the second alarm frequency subsequence corresponding to time s under the first alarm frequency subsequence, and the variance of the second alarm frequency subsequence corresponding to time s. Also calculate the slope information entropy of the slope of the second alarm frequency subsequence corresponding to time s, and the variance information entropy of the variance of the second alarm frequency subsequence corresponding to time s. The risk index is calculated using the following first objective formula: in, , , This indicates the risk index; This refers to the preset time window; This represents the time weighting function; The slope information entropy weight represents the slope information weight. The variance information entropy weights are represented by the weights. The slope of the second alarm frequency subsequence corresponding to time s is represented by... Let represent the variance of the second alarm frequency subsequence corresponding to time s; The slope information entropy, This represents the variance information entropy.

4. The method according to claim 1 or 2, characterized in that, The logistic regression model is expressed by the following second objective formula: ； in, Represents the target factor At any moment The dynamic conditional probability; Represents the target factor At any moment The risk index; It is the time when all parent nodes of the target factor are at time. A set of risk indices; It is the second node connected to the target factor at time 1. A set of risk indices; It is a bias term; , These are the regression coefficients of the risk index of the parent node and the risk index of the second node, respectively.

5. The method according to claim 1 or 2, characterized in that, The expression for the comprehensive risk index is as follows: ； in, Represents the target factor In the The overall risk index at any given moment; , , These are the target factors The corresponding operational impact factors, economic impact factors, and safety impact factors, , , These are the target factors The corresponding historical averages of historical operational impact factors, historical economic impact factors, and historical safety impact factors.

6. The method according to claim 1 or 2, characterized in that, The method further includes: Based on the causal structure of the dynamic Bayesian network, the intervention quantifier is used to extrapolate each intervention measure in the set of intervention measures, calculate the overall comprehensive risk index of the system after taking each intervention measure, and determine the intervention measure corresponding to the smallest overall comprehensive risk index as the target intervention measure.

7. A risk assessment device for secondary equipment, characterized in that, The device includes: The acquisition module is used to acquire, for each secondary device in the power system, the external operating condition information of the secondary device within the most recent preset time period and the alarm frequency sequence associated with the inherent risk factors of the secondary device. The first calculation module is used to calculate the risk index of the inherent risk factor of the target device at the current moment based on the fluctuation and trend of the alarm frequency sequence associated with the inherent risk factor of the target device; the target device is any one of the secondary devices. The first determining module is used to determine the dynamic conditional probability of the target factor based on the risk index of the target factor of the target device using a logistic regression model; the target factor is an inherent risk factor with a parent node in a dynamic Bayesian network; the dynamic Bayesian network is constructed with all the inherent risk factors of the target device as first nodes and all the external operating condition information of the target device as second nodes. The second determining module is used to input the dynamic conditional probability into the dynamic Bayesian network and determine the system risk probability of the target factor through a Bayesian inference algorithm. The third determining module is used to determine the risk assessment result of the secondary equipment based on the system risk probability of the target factor, the operational impact factor, the economic impact factor, and the safety impact factor of the target factor, using a comprehensive risk index expression.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images