Automatic conversion analysis method and system based on implicit common cause failure model

Abstract

The present invention discloses an automatic conversion analysis method and system based on an implicit common cause fault model, which relates to the technical field of power system fault analysis, including: acquiring and marking power equipment to form a plurality of equipment sets, setting a monitoring period for operation monitoring, and storing data in an operation record library; extracting equipment feature data, calculating the correlation index between equipment according to equipment categories, and using different strategies to calculate the correlation between equipment of the same category and different categories; using an implicit common cause fault model to calculate the probability of simultaneous abnormality of a reference equipment and other equipment under various abnormal causes, including independent failure probability, common cause failure probability and comprehensive probability; adjusting the monitoring period for equipment exceeding the threshold according to a comparison between the comprehensive probability and a preset threshold, and also adjusting the monitoring period for all associated equipment according to the abnormal cause, so as to realize effective monitoring and fault prevention of the power system.

Description

Technical Field

[0001] The present invention relates to the technical field of power system fault analysis, and in particular to an automatic conversion analysis method and system based on an implicit common cause fault model. Background Art

[0002] As the complexity of power systems continues to increase, the mutual influence and dependence between power equipment are becoming increasingly significant; in the operation of power systems, failures are often not isolated, but the result of the combined action of multiple factors, among which common cause failure is an important consideration; common cause failure refers to the phenomenon that two or more devices fail at the same time or in a relatively short time interval due to common reasons (such as external impact on the system, internal component failure propagation, etc.); in large and complex systems, such as power systems, common cause failure is an important cause of failure of its internal subsystems or components;

[0003] Traditional fault analysis methods often assume that the failures of subsystems or components within the system are independent of each other. This assumption may be valid in simple systems, but in complex systems, especially power systems with strong coupling relationships, it often leads to deviations in the analysis results. Secondly, in traditional fault analysis methods, the correlation analysis between power equipment may not be in-depth enough. The interactions and influences between equipment are complex, and simple statistical indicators or empirical judgments alone may not accurately reflect the actual relationship between equipment. In addition, traditional fault analysis methods usually use a fixed monitoring cycle, which may result in the inability to capture fault signals in time for highly correlated equipment. Summary of the invention

[0004] 1. Technical issues to be resolved

[0005] In response to the technical problems in the background technology, the present invention proposes an automatic conversion analysis method and system based on an implicit common cause failure model, which calculates the correlation index between devices by extracting the characteristic data of the devices; uses the implicit common cause failure model to calculate the probability of simultaneous failure of devices due to common causes; combines the correlation index and the probability of simultaneous failure due to common causes to dynamically adjust the monitoring period of the power equipment; thereby solving the technical problems recorded in the background technology.

[0006] (II) Technical solution

[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0008] Automatic conversion analysis method based on implicit common cause failure model, including:

[0009] Acquire the power equipment in the power system and combine the marks to obtain several power equipment sets; set the monitoring cycle to monitor the operation of the power equipment; store the operation monitoring data of the power equipment in the operation record library of the power system;

[0010] Extract characteristic data of various types of power equipment; calculate the correlation index between two power equipment based on the types of power equipment; for power equipment in the same power equipment set, implement the same type of correlation calculation strategy; for power equipment in different power equipment sets, implement the different types of correlation calculation strategy;

[0011] Select the reference power equipment and the secondary power equipment, and calculate the probability of the reference power equipment and the secondary power equipment failing independently within the set reference time; and calculate the common cause failure probability of the reference power equipment and the secondary power equipment and the probability of abnormality occurring simultaneously under the influence of the abnormal cause; after combining, the total probability of the reference power equipment and all other power equipment occurring abnormally simultaneously under the influence of each abnormal cause is obtained;

[0012] Calculate the comprehensive probability of two power devices experiencing abnormalities at the same time under the influence of each abnormal cause; based on the preset comprehensive probability threshold, adjust the monitoring cycle of the power equipment whose comprehensive probability exceeds the threshold; based on the cause of the abnormality of the power equipment, adjust the monitoring cycle of all power equipment related to the power equipment under the corresponding abnormal cause.

[0013] Specifically, the operation record library of the power system contains the operation record table of each power device in the power system, each operation record table contains the operation data table corresponding to each historical operation of the power system, and each operation data table stores the historical operation monitoring data of the corresponding power device each time the power system operates;

[0014] The historical operation monitoring data includes each monitoring data value and operation time data, as well as the operation status data and abnormality cause data of the corresponding power equipment obtained based on the analysis of each operation monitoring data.

[0015] Specifically, the same category relevance calculation strategy is:

[0016] The same number of historical records of normal operation status are obtained respectively, and the operation record data of the corresponding characteristic data are filtered out, which are recorded as: and ;

[0017] Calculate the mean of each set of running record data and ; Based on the sum of the deviation products and standard deviation of the two sets of operation record data, calculate the correlation index between the characteristic data of the two power equipment , the expression is: ;

[0018] Calculate the correlation index of each type of feature data in the associated feature set in turn to obtain the correlation index of each type of feature data ;

[0019] The correlation index of all feature data Combined, the correlation index of the two power devices is obtained , the expression is: ;in, Represents the total number of feature data in the associated feature set, Indicates the weight ratio coefficient of each feature data.

[0020] Specifically, the non-same category correlation calculation strategy is:

[0021] For two associated feature sets and , filter out similar feature data in two related feature sets and ,in, Indicates the number of similar feature data in two associated feature sets; calculates the correlation index between corresponding feature data according to the same category correlation calculation strategy, and combines the correlation indexes between similar feature data to obtain the first correlation index ;

[0022] The remaining feature data of the two associated feature sets and Combine each feature data in feature data set.

[0023] Further, obtain the running record data of each feature data group and , and record the data at the same position in the two sets of running record data as a data pair;

[0024] Arrange each set of acquired running record data in ascending order of value, and record the rank after sorting; assign the rank according to the actual position of each running record data;

[0025] Calculate the rank difference of each data pair in two sets of running record data , calculate the correlation index between two feature data based on rank difference , the expression is: ;

[0026] The correlation index between two feature data in the remaining feature data groups is calculated in the same way, and the second correlation index is obtained after the mean operation. ;

[0027] The first correlation index The second correlation index Combined, the correlation index of the two power devices is obtained , the expression is: .

[0028] Specifically, obtaining the benchmark power equipment appears All historical operation monitoring data of the abnormality; the total operation time of the historical operation monitoring data obtained by statistics , recorded as the benchmark time, and the benchmark power equipment appears Total duration of the anomaly ;

[0029] Get the total number of abnormalities of secondary power equipment within the reference time and the total duration of the anomaly ;

[0030] Calculate the probability of independent failure of the primary power equipment and the secondary power equipment separately and , the expression is: , .

[0031] Further, the abnormal reason data of each abnormality of the reference power equipment and the secondary power equipment during the reference time are obtained respectively, and after classification, a comprehensive abnormal reason set is obtained;

[0032] Obtain each abnormal reason data in the comprehensive abnormal reason set respectively, and obtain the total number of abnormalities of the reference power equipment and the secondary power equipment due to the abnormal reason within the reference time respectively and , and count the number of times the baseline power equipment and the secondary power equipment have abnormalities at the same time due to the abnormal reason ;

[0033] Get the total number of times the baseline power equipment and secondary power equipment have abnormalities at the same time during the baseline time , calculate the probability of common cause failure , the expression is: ;

[0034] Calculate the probability that both the reference power equipment and the secondary power equipment will be abnormal at the same time under the influence of the abnormal cause , the expression is: ;

[0035] Calculate the total probability of abnormality of the reference power equipment and the secondary power equipment under the influence of each abnormal reason at the same time , the expression is: .

[0036] Furthermore, based on the correlation index between the two power devices , and the probability that the two power equipment will be abnormal at the same time under the influence of each abnormal cause , calculate the comprehensive probability of two power equipment experiencing abnormalities at the same time under the influence of each abnormal cause , the expression is: .

[0037] Furthermore, a comprehensive probability threshold is preset , when under the influence of a certain abnormal reason, the comprehensive probability of two power equipments experiencing abnormalities at the same time , the monitoring period of these two power devices is shortened to ,in, is the length of a monitoring period; otherwise, the monitoring period is not adjusted;

[0038] If an abnormality is detected in the power equipment during the operation of the power system, the cause of the abnormality is analyzed, and the comprehensive probability of the corresponding power equipment having an abnormality at the same time under the abnormal cause is obtained. For all electrical equipment, adjust the monitoring cycle of these electrical equipment to real-time monitoring.

[0039] Automatic conversion analysis system based on implicit common cause failure model, including:

[0040] The data acquisition and storage module is used to acquire the power equipment in the power system and perform tag combination to obtain a number of power equipment sets; set a monitoring cycle to perform operation monitoring on the power equipment; and store the operation monitoring data of the power equipment in the operation record library of the power system;

[0041] The correlation analysis module is used to extract the characteristic data of various types of power equipment; calculate the correlation index between two power equipment based on the types of power equipment; for power equipment in the same power equipment set, the same type of correlation calculation strategy is implemented; for power equipment in different power equipment sets, the non-same type of correlation calculation strategy is implemented;

[0042] The co-occurrence probability analysis module is used to select the reference power equipment and the secondary power equipment, and respectively calculate the probability of the reference power equipment and the secondary power equipment failing independently within the set reference time; and calculate the common cause failure probability of the reference power equipment and the secondary power equipment and the probability of simultaneous abnormality under the influence of the abnormal cause; after combining, the total probability of the reference power equipment and all other power equipment simultaneously failing under the influence of each abnormal cause is obtained;

[0043] The monitoring cycle adjustment module is used to calculate the comprehensive probability of two power devices simultaneously experiencing abnormalities under the influence of each abnormal cause; based on a preset comprehensive probability threshold, the monitoring cycle of the power equipment whose comprehensive probability exceeds the threshold is adjusted; based on the cause of the abnormality of the power equipment, the monitoring cycle of all power equipment related to the power equipment under the corresponding abnormal cause is adjusted.

[0044] (III) Beneficial effects

[0045] The present invention provides an automatic conversion analysis method and system based on an implicit common cause failure model, which has the following beneficial effects:

[0046] 1. By systematically acquiring, marking, monitoring and storing the operating data of power equipment, it provides a comprehensive, accurate and reliable data basis for subsequent common cause failure analysis, effectively improving the accuracy and efficiency of fault location and diagnosis;

[0047] 2. By extracting the characteristic data of power equipment in a refined manner and using different strategies to calculate the correlation index according to the type of power equipment, the similarity between equipment of the same category and the correlation between equipment of different categories are taken into account, providing a scientific quantitative basis for accurately identifying common cause failures of power equipment, and effectively improving the accuracy and comprehensiveness of fault analysis;

[0048] 3. By selecting benchmark power equipment and secondary power equipment, and carefully calculating the probability of their independent failure and common cause failure within the set benchmark time, as well as the probability of abnormalities occurring simultaneously under the influence of different abnormal causes, the fault correlation between power equipment is fully revealed, providing a quantitative basis for accurately assessing the common cause failure risk of the power system, which helps to improve the reliability and safety of the power system;

[0049] 4. By calculating the comprehensive probability of two power equipment experiencing abnormalities at the same time under the influence of each abnormal cause, adjusting the monitoring period according to the preset threshold, and conducting real-time monitoring of relevant power equipment, the accuracy and response speed of fault prediction are effectively improved, providing a strong guarantee for the safe and stable operation of the power system and reducing the system risk caused by common cause failures. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 A schematic diagram of the steps of the automatic conversion analysis method based on the implicit common cause failure model provided by the present invention;

[0051] Figure 2 A schematic diagram of the structure of an automatic conversion analysis system based on an implicit common cause failure model provided by the present invention. DETAILED DESCRIPTION

[0052] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0053] refer to Figure 1 The present invention provides an automatic conversion analysis method based on an implicit common cause failure model, comprising:

[0054] Step 1: Obtain the power equipment in the power system and perform tag combination to obtain a number of power equipment sets; set a monitoring cycle to perform operation monitoring on the power equipment; store the operation monitoring data of the power equipment in the operation record library of the power system;

[0055] The step 1 comprises the following steps:

[0056] Step 101: Acquire various power equipment in the power system and mark and combine them according to the types of power equipment; the power equipment in the power system includes generators, transformers, circuit breakers, capacitors, etc., and the number of various types of power equipment is not all 1;

[0057] Starting from the starting point of the power system, various types of power equipment are marked in the order in which the electric energy passes through. The power equipment that the electric energy passes first is marked with a letter. If the electrical equipment that passes immediately after is not the same type of equipment, use letters If they are the same type of equipment, letters are used. Indicates that the power equipment The mark is converted to , and so on;

[0058] After marking all the power equipment in the power system, the marks of the same equipment are put into the same power equipment set, and multiple power equipment sets in the power system are obtained, specifically: , wait;

[0059] Step 102: When the power system starts to operate, the operation of each power device in the power system is monitored; a monitoring cycle is set, and the current, voltage or other data of each power device is monitored and recorded at the end of each monitoring cycle using a monitoring sensor group, wherein the other data includes the number of operation times of the circuit breaker and the oil temperature of the transformer; the monitoring sensor group is designed as independent modules such as current sensor, voltage sensor, counter and temperature sensor, and integrated into the same device based on the interface and protocol of each module, and the device is the integrated sensor group;

[0060] Step 103: The operation record library of the power system contains the operation record table of each power device in the power system, each operation record table contains the operation data table corresponding to each historical operation of the power system, and each operation data table stores the historical operation monitoring data of the corresponding power device when the power system is operated each time; wherein the historical operation monitoring data includes each monitoring data value and operation time data, and the operation status data and abnormal cause data of the corresponding power device obtained based on the analysis of each operation monitoring data;

[0061] The operation status data includes normal operation and abnormal operation, and the abnormal cause data refers to the cause of abnormal operation. When the power equipment is monitored to have abnormal operation, the cause of the problem is located and determined, including environmental factors, interaction factors between equipment and equipment factors, wherein environmental factors include abnormal temperature, abnormal humidity, electromagnetic interference, rain and snow interference, and lightning interference, etc., interaction factors between equipment include electromagnetic coupling and connection problems between power equipment, etc., and equipment factors include equipment aging, equipment defects, etc.; the abnormal cause data are used as data sources for subsequent common cause failure analysis;

[0062] Step 104: Whenever a monitoring cycle ends, preprocessing operations are performed on various types of received data, including data cleaning, smoothing, and outlier processing operations, which can effectively remove noise and outliers, improve data accuracy and reliability, and provide a solid foundation for subsequent data analysis and fault diagnosis.

[0063] When using, combine the contents in steps 101 to 104:

[0064] By systematically acquiring, marking, monitoring and storing the operating data of power equipment, a comprehensive, accurate and reliable data foundation is provided for subsequent common cause failure analysis, effectively improving the accuracy and efficiency of fault location and diagnosis.

[0065] Step 2: extract characteristic data of various types of power equipment; calculate the correlation index between two power equipment based on the types of power equipment; for power equipment in the same power equipment set, execute the same type correlation calculation strategy; for power equipment in different power equipment sets, execute the non-same type correlation calculation strategy;

[0066] The step 2 includes the following steps:

[0067] Step 201: Obtain all power equipment sets in the power system, take out one power equipment from all power equipment sets in turn, and calculate its correlation index with other power equipment;

[0068] Extract characteristic data of various types of power equipment, i.e., characteristic data used to calculate correlation index; for example, characteristic data of generator equipment includes voltage, current, power factor, insulation resistance, etc.; characteristic data of transformer equipment includes voltage, current, power loss, oil temperature, oil level, etc.; characteristic data of circuit breaker equipment includes voltage, current, number of operations, etc.; characteristic data of insulation resistance includes capacitance value, voltage, current, insulation resistance, etc.;

[0069] The characteristic data of various types of power equipment are stored in the associated characteristic set of each power equipment set, and are represented by lowercase letters, specifically , The power equipment sets corresponding to uppercase and lowercase letters correspond to the associated feature sets, such as the power equipment set The associated feature set is ;

[0070] Step 202: For the power equipment in the same power equipment set, the same category correlation calculation strategy is executed: the correlation index of two power equipments of the same type is calculated based on the correlation index between all corresponding feature data, including the following steps:

[0071] Step 2021, respectively taking out the associated feature sets of the power equipment set to which the corresponding power equipment belongs, and calculating the correlation index between each feature data in the associated feature set in turn;

[0072] Obtain the same number of historical record data from the operation record table of the corresponding power equipment respectively, and filter out the operation record data of the corresponding characteristic data, and the selected characteristic data should be the characteristic data monitored when the corresponding power equipment is operating normally;

[0073] Based on the two sets of running record data obtained and , the data at the same position in the two sets of running record data are monitored at the same time and recorded as a data pair; calculate the mean of each set of running record data and ;

[0074] Step 2022: Calculate the corresponding sum of deviation products based on the mean of each set of running record data. The expression is: ;

[0075] The standard deviation of each set of running record data is calculated based on the mean of each set of running record data. The expressions are and ;

[0076] Based on the sum of the deviation products and the standard deviation of the two sets of operation record data, the correlation index between the characteristic data of the two power equipment is calculated. , the expression is: ;

[0077] Step 2023: Calculate the correlation index of each type of feature data in the associated feature set in turn according to the above method to obtain the correlation index of each type of feature data. ;

[0078] The correlation index of all feature data Combined, the correlation index of the two power devices is obtained , the expression is: ;in, Represents the total number of feature data in the associated feature set, It represents the weight ratio coefficient of each feature data. The specific value is calculated based on the entropy weight method. The entropy weight method measures the degree of variation or the amount of information by calculating the information entropy of each indicator, and then determines the weight of each indicator.

[0079] Step 203: For the power equipment in different power equipment sets, a non-same-category correlation calculation strategy is executed: first, the correlation index between the same type of feature data is calculated according to step 202, and for the feature data of different types, the correlation index between the feature data is calculated based on the permutation and combination method. The specific steps are as follows:

[0080] Step 2031: For two associated feature sets and , filter out similar feature data in two related feature sets and ,in, and correspond, and Correspondingly, and so on, Indicates the number of similar feature data in two associated feature sets; calculate the correlation index between the corresponding feature data according to step 202, and combine the correlation indexes between the similar feature data according to step 203 to obtain a first correlation index ;

[0081] Step 2032: For the remaining feature data in the two associated feature sets and ,Will Each feature data in Combine each feature data in feature data sets;

[0082] Step 2033: For each feature data group, obtain the running record data of each feature data from the historical record data. and ,The data at the same position in the two sets of running record data are also monitored at the same time;

[0083] Arrange each set of running record data in ascending order and record the position after sorting, i.e., the rank; assign the rank according to the actual position of each running record data. middle ,but The rank of is 3, The rank of is 1, The rank is 2, that is, the running record data The rank of ;

[0084] Calculate the rank difference of each data pair in two sets of running record data , calculate the correlation index between two feature data based on rank difference , the expression is: ; Calculate the correlation index between two feature data in the remaining feature data groups in the same way, and obtain the second correlation index after performing mean operation ;

[0085] Step 2034: The first correlation index The second correlation index Combined, the correlation index of the two power devices is obtained , the expression is: .

[0086] When using, combine the contents in steps 201 to 203:

[0087] By finely extracting the characteristic data of power equipment and using different strategies to calculate the correlation index according to the category of power equipment, we take into account both the similarities between equipment of the same category and the correlation between equipment of different categories, providing a scientific quantitative basis for accurately identifying common cause failures of power equipment and effectively improving the accuracy and comprehensiveness of fault analysis.

[0088] Step 3: Select the reference power equipment and the secondary power equipment, and calculate the probability of the reference power equipment and the secondary power equipment failing independently within the set reference time; and calculate the common cause failure probability of the reference power equipment and the secondary power equipment and the probability of abnormality occurring simultaneously under the influence of the abnormal cause; after combining, the total probability of the reference power equipment and all other power equipment occurring abnormally simultaneously under the influence of each abnormal cause is obtained;

[0089] The step three includes the following steps:

[0090] Step 301: Take out one power device in turn according to the order in which the electric energy passes through and record it as the reference power device. Based on the implicit common cause failure model, calculate the probability of the reference power device and any other power device in the power system having an abnormality at the same time under each abnormal cause.

[0091] Select the benchmark power equipment and obtain the power equipment from the operation record table of the benchmark power equipment. All historical operation monitoring data of the abnormality, that is, starting from the latest historical operation monitoring data, until the abnormality appears in the acquired data Stop acquisition when an abnormality occurs; count the total running time of the historical operation monitoring data obtained , recorded as the benchmark time, and the benchmark power equipment appears Total duration of the anomaly ; The reference time only includes the time when the power system is in operation;

[0092] Step 302: Take out all the power equipments in the power system that are behind the reference power equipment in turn, and take out one power equipment in turn according to the order in which the electric energy passes through and record it as a secondary power equipment;

[0093] Based on the operation record table of the secondary power equipment, obtain the total number of abnormalities of the secondary power equipment within the reference time and the total duration of the anomaly ;

[0094] Based on the total duration of abnormalities of the reference power equipment and the secondary power equipment, the probability of independent failure of these two equipment is calculated respectively. and , the expression is: , ;

[0095] Step 303: respectively obtain the abnormal reason data of each abnormality of the reference power equipment and the secondary power equipment during the reference time, and classify them into categories to obtain a comprehensive abnormal reason set, which includes all abnormal reason data of the abnormality of the reference power equipment and the secondary power equipment;

[0096] Obtain each abnormal reason data in the comprehensive abnormal reason set respectively, and obtain the total number of abnormalities of the reference power equipment and the secondary power equipment due to the abnormal reason within the reference time respectively. and , and count the number of times the baseline power equipment and the secondary power equipment have abnormalities at the same time due to the abnormal reason The said simultaneous abnormality refers to the situation where the abnormal time of the reference power equipment and the secondary power equipment overlaps. For example, if the reference power equipment is abnormal within 20-21 minutes after the power system is put into operation, and the secondary power equipment is abnormal within 19-21 minutes after the power system is put into operation, it is counted as one simultaneous abnormality;

[0097] Get the total number of times the baseline power equipment and secondary power equipment have abnormalities at the same time during the baseline time , calculate the probability of common cause failure , the expression is: ;

[0098] Calculate the probability that both the reference power equipment and the secondary power equipment will be abnormal at the same time under the influence of the abnormal cause , the expression is: ;

[0099] Step 304: Calculate the total probability of the reference power equipment and the secondary power equipment being abnormal at the same time under the influence of each abnormal reason. , the expression is: ;

[0100] Calculate and record the total probability of the benchmark power equipment and all other power equipment experiencing abnormalities at the same time under the influence of each abnormal cause.

[0101] When used, combine the contents in steps 301 to 304:

[0102] By selecting benchmark power equipment and secondary power equipment, and carefully calculating the probability of their independent failure and common cause failure within the set benchmark time, as well as the probability of simultaneous abnormalities under the influence of different abnormal reasons, the failure correlation between power equipment is fully revealed, providing a quantitative basis for accurately assessing the common cause failure risk of the power system, which helps to improve the reliability and safety of the power system.

[0103] Step 4: Calculate the comprehensive probability of two power devices having abnormalities at the same time under the influence of each abnormal reason; based on a preset comprehensive probability threshold, adjust the monitoring cycle of the power equipment whose comprehensive probability exceeds the threshold; based on the cause of the abnormality of the power equipment, adjust the monitoring cycle of all power equipment related to the power equipment under the corresponding abnormal cause;

[0104] The step 4 includes the following steps:

[0105] Step 401: Obtain the correlation index between any two power devices based on step 2 Based on step 3, the probability of simultaneous abnormality between any two power devices under the influence of each abnormal cause is obtained. , after combining, we get the comprehensive probability of two power equipments having abnormalities at the same time under the influence of each abnormal cause. , the expression is: ;

[0106] Step 402: The power system manager presets the comprehensive probability threshold , when the comprehensive probability of two power equipment experiencing abnormalities at the same time under the influence of a certain abnormal reason is calculated , it means that the probability of the two power equipment being abnormal at the same time is high under the influence of the corresponding abnormal reasons. The monitoring cycle of the two power equipment is adjusted to shorten the monitoring cycle to ,in, is the length of a monitoring period; otherwise, the monitoring period is not adjusted;

[0107] Step 403: If an abnormality is detected in the power equipment during the operation of the power system, the cause of the abnormality is analyzed, and the comprehensive probability of the abnormality occurring simultaneously with the corresponding power equipment under the abnormal cause is obtained. For all electrical equipment, adjust the monitoring cycle of these electrical equipment to real-time monitoring.

[0108] When used, combine the contents in steps 401 to 403:

[0109] By calculating the comprehensive probability of two power equipment experiencing abnormalities at the same time under the influence of each abnormal cause, adjusting the monitoring period according to the preset threshold, and conducting real-time monitoring of relevant power equipment, the accuracy and response speed of fault prediction are effectively improved, providing a strong guarantee for the safe and stable operation of the power system and reducing the system risks caused by common cause failures.

[0110] refer to Figure 2 The present invention also provides an automatic conversion analysis system based on an implicit common cause failure model, comprising:

[0111] The data acquisition and storage module is used to acquire the power equipment in the power system and perform tag combination to obtain a number of power equipment sets; set a monitoring cycle to perform operation monitoring on the power equipment; and store the operation monitoring data of the power equipment in the operation record library of the power system;

[0112] The correlation analysis module is used to extract the characteristic data of various types of power equipment; calculate the correlation index between two power equipment based on the types of power equipment; for power equipment in the same power equipment set, the same type of correlation calculation strategy is implemented; for power equipment in different power equipment sets, the non-same type of correlation calculation strategy is implemented;

[0113] The co-occurrence probability analysis module is used to select the reference power equipment and the secondary power equipment, and respectively calculate the probability of the reference power equipment and the secondary power equipment failing independently within the set reference time; and calculate the common cause failure probability of the reference power equipment and the secondary power equipment and the probability of simultaneous abnormality under the influence of the abnormal cause; after combining, the total probability of the reference power equipment and all other power equipment simultaneously failing under the influence of each abnormal cause is obtained;

[0114] The monitoring cycle adjustment module is used to calculate the comprehensive probability of two power devices simultaneously experiencing abnormalities under the influence of each abnormal cause; based on a preset comprehensive probability threshold, the monitoring cycle of the power equipment whose comprehensive probability exceeds the threshold is adjusted; based on the cause of the abnormality of the power equipment, the monitoring cycle of all power equipment related to the power equipment under the corresponding abnormal cause is adjusted.

[0115] In the above embodiments, all or part of the embodiments may be implemented by software, hardware, firmware or any combination thereof. When implemented by software, all or part of the embodiments may be implemented in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of the present invention are generated. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer storage medium or transmitted via a computer storage medium.

[0116] Computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. Computer storage media can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that includes one or more available media. Available media can be magnetic media (e.g., floppy disks, hard disks, tapes), optical media (e.g., DVDs), or semiconductor media (e.g., solid state drives (SSDs)).

[0117] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. An automatic conversion analysis method based on an implicit common cause failure model, characterized by: The steps include: Obtain the power equipment in the power system, and perform tag combination to obtain a number of power equipment sets; set a monitoring cycle to perform operation monitoring on the power equipment; The operation monitoring data of the power equipment is stored in the operation record library of the power system; Extract characteristic data of various types of power equipment; calculating a correlation index between two electric devices based on the categories of the electric devices; For the power equipment in the same power equipment set, the same category correlation calculation strategy is implemented; for the power equipment in different power equipment sets, the non-same category correlation calculation strategy is implemented; The same category relevance calculation strategy is: Get the same number of historical records of normal operation status respectively, and filter out the operation records of corresponding feature data, which are recorded as: {X1,X2,...,X n } and {Y1,Y2,...,Y n }, where n represents the total number of filtered running record data; Calculate the mean of each set of running record data and Based on the sum of the deviation products and the standard deviation of the two sets of operation record data, the correlation index Ci between the corresponding characteristic data of the two power equipment is calculated, and the expression is: Calculate the correlation index of each type of feature data in the associated feature set in turn to obtain the correlation index Ci of each type of feature data j ; The correlation index Ci of all feature data j Combined, the correlation index ZCi of the two power equipment is obtained, and the expression is: Among them, m represents the total number of feature data in the associated feature set, α j Represents the weight ratio coefficient of each feature data; The non-same category correlation calculation strategy is: For two associated feature sets {a1, a2, ...} and {b1, b2, ...}, filter out the same feature data {a1, a2, ..., a m1 } and {b1,b2,...,b m1 }, where m1 represents the number of similar feature data in the two associated feature sets; the correlation index between the corresponding feature data is calculated according to the same category correlation calculation strategy, and the correlation index between the similar feature data is combined to obtain the first correlation index FCi; The remaining feature data {a m1+1 ,a m1+2 ,...,a ma } and {b m1+1 ,b m1+2 ,...,b mb }, and a total of (ma-m1)*(mb-m1) feature data groups are obtained, where ma and mb represent the number of feature data in the two associated feature sets respectively; Get the running record data of each characteristic data group {U1,U2,...,U n } and {V1,V2,...,V n }, and record the data at the same position in the two sets of running record data as a data pair; Arrange each set of acquired running record data in ascending order of value, and record the rank after sorting; assign the rank according to the actual position of each running record data; Calculate the rank difference d of each data pair in the two sets of running record data i , the correlation index Ci between two feature data is calculated based on the rank difference, and the expression is: The correlation index between two feature data in the remaining feature data groups is calculated in the same way, and the second correlation index SCi is obtained after the mean operation; The first correlation index FCi is combined with the second correlation index SCi to obtain the correlation index ZCi corresponding to the two power devices, which is expressed as: Select the reference power equipment and the secondary power equipment, and calculate the probability of the reference power equipment and the secondary power equipment failing independently within the set reference time; and calculate the common cause failure probability of the reference power equipment and the secondary power equipment and the probability of abnormality occurring simultaneously under the influence of each abnormal cause; after combining, the total probability of the reference power equipment and all other power equipment occurring abnormally simultaneously under the influence of each abnormal cause is obtained; Calculate the comprehensive probability of two power devices experiencing abnormalities at the same time under the influence of each abnormal cause; based on the preset comprehensive probability threshold, adjust the monitoring cycle of the power equipment whose comprehensive probability exceeds the threshold; based on the cause of the abnormality of the power equipment, adjust the monitoring cycle of all power equipment related to the power equipment under the corresponding abnormal cause.

2. The automatic conversion analysis method based on the implicit common cause failure model according to claim 1, characterized in that: The operation record library of the power system contains the operation record table of each power equipment in the power system, each operation record table contains the operation data table corresponding to each historical operation of the power system, and each operation data table stores the historical operation monitoring data of the corresponding power equipment each time the power system operates; The historical operation monitoring data includes each monitoring data value and operation time data, as well as the operation status data and abnormality cause data of the corresponding power equipment obtained based on the analysis of each operation monitoring data.

3. The automatic conversion analysis method based on the implicit common cause failure model according to claim 1, characterized in that: Obtain all historical operation monitoring data of benchmark power equipment with N1 abnormalities; The total operation time t of the historical operation monitoring data obtained is counted and recorded as the reference time, and the total duration t1 of N1 abnormalities of the reference power equipment is counted; Obtain the total number of abnormalities N2 and the total duration t2 of the abnormalities occurring in the secondary power equipment within the reference time; The probabilities of independent failure of the reference power equipment and the secondary power equipment, P1 and P2, are calculated respectively, and the expressions are:

4. The automatic conversion analysis method based on the implicit common cause failure model according to claim 3 is characterized in that: Obtain the abnormal cause data of each abnormality of the reference power equipment and the secondary power equipment during the reference time, and obtain a comprehensive abnormal cause set after classifying them; Obtain each abnormal reason data in the comprehensive abnormal reason set respectively, and obtain the total number of times n1 and n2 of abnormality of the reference power equipment and the secondary power equipment due to the abnormal reason within the reference time respectively, and count the number of times n3 of abnormality of the reference power equipment and the secondary power equipment due to the abnormal reason at the same time; Obtain the total number of times N3 when the reference power equipment and the secondary power equipment are abnormal at the same time within the reference time, and calculate the common cause failure probability P3, which is expressed as: Calculate the probability P4 that the reference power equipment and the secondary power equipment will be abnormal at the same time under the influence of the abnormal cause. The expression is: The total probability P of the reference power equipment and the secondary power equipment simultaneously appearing abnormal under the influence of each abnormal cause is calculated respectively, and the expression is: P = P1*P2+P3*P4.

5. The automatic conversion analysis method based on the implicit common cause failure model according to claim 4, characterized in that: Based on the correlation index ZCi between the two power devices and the probability P that the two power devices will be abnormal at the same time under the influence of each abnormal cause, the comprehensive probability p that the two power devices will be abnormal at the same time under the influence of each abnormal cause is calculated, and the expression is:

6. The automatic conversion analysis method based on the implicit common cause failure model according to claim 5, characterized in that: A comprehensive probability threshold p0 is preset. When the comprehensive probability p≥p0 of two power devices having abnormalities at the same time under the influence of a certain abnormal reason, the monitoring period of the two power devices is shortened to p0*T, where T is the length of a monitoring period; otherwise, the monitoring period is not adjusted; If an abnormality is detected in power equipment during the operation of the power system, the cause of the abnormality is analyzed, and all power equipment with a comprehensive probability p≥p0 of having an abnormality at the same time as the corresponding power equipment under the abnormal cause are obtained, and the monitoring period of these power equipment is adjusted to real-time monitoring.

7. Automatic conversion analysis system based on implicit common cause failure model, characterized in that: include: The data acquisition and storage module is used to acquire the power equipment in the power system and perform tag combination to obtain several power equipment sets; Set monitoring cycles to monitor the operation of power equipment; The operation monitoring data of the power equipment is stored in the operation record library of the power system; The correlation analysis module is used to extract characteristic data of various types of power equipment; calculate the correlation index between two power equipment based on the types of power equipment; For the power equipment in the same power equipment set, the same category correlation calculation strategy is implemented; for the power equipment in different power equipment sets, the non-same category correlation calculation strategy is implemented; The same category relevance calculation strategy is: Get the same number of historical records of normal operation status respectively, and filter out the operation records of corresponding feature data, which are recorded as: {X1,X2,...,X n } and {Y1,Y2,...,Y n }, where n represents the total number of filtered running record data; Calculate the mean of each set of running record data and Based on the sum of the deviation products and the standard deviation of the two sets of operation record data, the correlation index Ci between the corresponding characteristic data of the two power equipment is calculated, and the expression is: Calculate the correlation index of each type of feature data in the associated feature set in turn to obtain the correlation index Ci of each type of feature data j ; The correlation index Ci of all feature data j Combined, the correlation index ZCi of the two power equipment is obtained, and the expression is: Among them, m represents the total number of feature data in the associated feature set, α j Represents the weight ratio coefficient of each feature data; The non-same category correlation calculation strategy is: For two associated feature sets {a1, a2, ...} and {b1, b2, ...}, filter out the same feature data {a1, a2, ..., a m1 } and {b1,b2,...,b m1 }, where m1 represents the number of similar feature data in the two associated feature sets; the correlation index between the corresponding feature data is calculated according to the same category correlation calculation strategy, and the correlation index between the similar feature data is combined to obtain the first correlation index FCi; The remaining feature data {a m1+1 ,a m1+2 ,...,a ma } and {b m1+1 ,b m1+2 ,...,b mb }, and a total of (ma-m1)*(mb-m1) feature data groups are obtained, where ma and mb represent the number of feature data in the two associated feature sets respectively; Get the running record data of each characteristic data group {U1,U2,...,U n } and {V1,V2,...,V n }, and record the data at the same position in the two sets of running record data as a data pair; Arrange each set of acquired running record data in ascending order of value, and record the rank after sorting; assign the rank according to the actual position of each running record data; Calculate the rank difference d of each data pair in the two sets of running record data i , the correlation index Ci between two feature data is calculated based on the rank difference, and the expression is: The correlation index between two feature data in the remaining feature data groups is calculated in the same way, and the second correlation index SCi is obtained after the mean operation; The first correlation index FCi is combined with the second correlation index SCi to obtain the correlation index ZCi corresponding to the two power devices, which is expressed as: The co-occurrence probability analysis module is used to select the reference power equipment and the secondary power equipment, and respectively calculate the probability of the reference power equipment and the secondary power equipment failing independently within the set reference time; and calculate the common cause failure probability of the reference power equipment and the secondary power equipment and the probability of simultaneous abnormality under the influence of each abnormal cause; after combining, the total probability of the reference power equipment and all other power equipment simultaneously failing under the influence of each abnormal cause is obtained; The monitoring cycle adjustment module is used to calculate the comprehensive probability of two power devices simultaneously experiencing abnormalities under the influence of each abnormal cause; based on a preset comprehensive probability threshold, the monitoring cycle of the power equipment whose comprehensive probability exceeds the threshold is adjusted; based on the cause of the abnormality of the power equipment, the monitoring cycle of all power equipment related to the power equipment under the corresponding abnormal cause is adjusted.

Images