A method, device, equipment and readable storage medium for analyzing and judging electric meter faults

Abstract

The present application discloses a method, device, equipment and readable storage medium for analyzing and judging electric meter faults. The method includes: obtaining historical fault information of each electric meter and fault information of the current faulty electric meter; determining the fault state code corresponding to each fault information in the historical fault information; counting the types of fault state transfer groups and the number of each type of fault state transfer groups in the historical fault information, and generating a fault state transfer matrix based on the statistical results; determining the target fault state code corresponding to the fault information of the current faulty electric meter based on the fault information of the current faulty electric meter; and determining the fault analysis result of the current faulty electric meter based on the fault state transfer matrix and the target fault state code. Based on the fault analysis result, the operation and maintenance personnel will check the possible secondary fault problems in advance to avoid the occurrence of secondary faults, or when checking the secondary faults, give priority to the fault types with high probability to improve the efficiency of fault inspection during maintenance.

Description

Technical Field

[0001] The present application relates to the field of power diagnosis, and more specifically, to a method, apparatus, device, and readable storage medium for analyzing and determining an electric meter fault. Background Art

[0002] Smart meters are one of the basic devices for data collection in smart grids, especially smart distribution networks. They are responsible for collecting, measuring and transmitting raw electric energy data, and are the basis for information integration, analysis and optimization, and information presentation.

[0003] For a long time, after the comprehensive operation of smart meters and the construction of related information systems, smart meter failures have occurred frequently, and the power grid system has also collected a series of fault operation and maintenance work orders for users' smart meters. However, due to incomplete fault discovery or chain reactions between multiple abnormal faults, a large number of meters have other or similar fault events not long after the last fault was handled. This has brought a lot of repetitive work to the on-site inspection and verification of the operation and maintenance personnel.

[0004] Based on the above situation, this application proposes a meter fault analysis solution to achieve early warning of the occurrence of secondary faults. Summary of the Invention

[0005] In view of this, the present application provides a method, device, equipment and readable storage medium for analyzing and judging electric meter faults, so as to achieve an effective comprehensive quantitative assessment of the implementation of production safety responsibilities of various personnel throughout the entire operation process.

[0006] In order to achieve the above objectives, the following solutions are proposed:

[0007] A method for analyzing and judging an electric meter fault, comprising:

[0008] Obtain historical fault information of each meter and fault information of the current faulty meter;

[0009] Determine the fault status code corresponding to each piece of fault information in the historical fault information;

[0010] Counting the types of fault state transition groups and the number of each type of fault state transition groups in the historical fault information, and generating a fault state transition matrix based on the statistical results, wherein the fault state transition group includes the fault state codes and transition directions of two adjacent faults on the same electricity meter, and the fault state transition matrix is ​​composed of transition probabilities between each fault state code;

[0011] Determining a target fault state code corresponding to the fault information of the current faulty electric meter according to the fault information of the current faulty electric meter;

[0012] According to the fault state transfer matrix and the target fault state code, a fault analysis result of the current faulty electric meter is determined, wherein the fault analysis result includes a possible fault type and a corresponding probability of the next fault.

[0013] Preferably, determining the fault status code corresponding to each piece of fault information in the historical fault information includes:

[0014] Determine the fault type and the code of each fault type for each piece of fault information in the historical fault information;

[0015] The code is determined as the fault status code of the corresponding fault information.

[0016] Preferably, respectively determining the fault type and the code of each fault type of each piece of fault information in the historical fault information includes:

[0017] Classifying the historical fault information into data anomaly types and device anomaly types;

[0018] For each fault message belonging to the data anomaly type:

[0019] Obtaining the electricity consumption curve of each electricity meter with data anomalies within a preset time range and generating an electricity consumption curve database;

[0020] Clustering the electricity consumption curve database to obtain a plurality of clusters, using the median curve of the electricity consumption curve in each cluster as a class characteristic curve, and defining a data anomaly subtype and code for each cluster, wherein the anomaly type of the cluster serves as the anomaly type to which each fault information in the cluster belongs, and the code of the cluster serves as the code for each fault information in the cluster;

[0021] For each fault message belonging to the device abnormality type:

[0022] Define the device exception subtype and code for each exception under the device exception type.

[0023] Preferably, before generating the electricity consumption curve database, the method further includes:

[0024] The power consumption curve is normalized.

[0025] Preferably, if the fault information of the current faulty meter belongs to the fault information of the data abnormality type, the process of determining the target fault state code corresponding to the fault information of the current faulty meter includes:

[0026] Obtaining a target power consumption curve within a preset time range of the currently faulty electric meter;

[0027] A similarity evaluation is performed on the target power consumption curve and the class characteristic curve of each cluster, and the code of the cluster where the class characteristic curve with the highest similarity belongs is used as the target fault state code.

[0028] Preferably, similarity evaluation is performed on the target power consumption curve and the class characteristic curve of each cluster, and the code of the cluster where the class characteristic curve with the highest similarity belongs is used as the target fault state code, including:

[0029] The Euclidean distance between the target power consumption curve and the class characteristic curve of each cluster is calculated one by one, and the code of the cluster where the class characteristic curve with the smallest Euclidean distance belongs is used as the target fault state code.

[0030] Preferably, defining the encoding of each cluster includes:

[0031] The code of each cluster is determined according to the arrangement order of the class characteristic curves of each cluster.

[0032] An electric meter fault analysis and judgment device, comprising:

[0033] An acquisition unit, used to acquire historical fault information of each electric meter and fault information of the current faulty electric meter;

[0034] A first encoding unit, configured to determine a fault status code corresponding to each piece of fault information in the historical fault information;

[0035] a statistical unit, configured to count the types of fault state transition groups and the number of each type of fault state transition groups in the historical fault information, and generate a fault state transition matrix based on the statistical results, wherein the fault state transition group includes the fault state codes and transition directions of two adjacent faults on the same electric meter, and the fault state transition matrix is ​​composed of transition probabilities between the fault state codes;

[0036] A second encoding unit is used to determine a target fault state code corresponding to the fault information of the current faulty electric meter according to the fault information of the current faulty electric meter;

[0037] The analysis and judgment unit is used to determine the fault analysis result of the current faulty electric meter according to the fault state transfer matrix and the target fault state code, wherein the fault analysis result includes the possible fault type and corresponding probability of the next fault.

[0038] An electric meter fault analysis and judgment device, comprising a memory and a processor;

[0039] The memory is used to store programs;

[0040] The processor is used to execute the program to implement the various steps of the above-mentioned electric meter fault analysis method.

[0041] A readable storage medium stores a computer program, which, when executed by a processor, implements each step of the above-mentioned electric meter fault analysis and judgment method.

[0042] As can be seen from the above technical solutions, the embodiments of the present application provide a method, device, equipment, and readable storage medium for analyzing and determining meter faults. By obtaining historical fault information of each meter and fault information of the current faulty meter, the fault state code corresponding to each piece of fault information in the historical fault information is determined. The historical fault information of each meter is statistically analyzed to obtain the types of fault state transfer groups in the historical fault information and the number of each type of fault state transfer group. A fault state transfer matrix is ​​generated based on the statistical results. Since the fault state transfer group includes the fault state codes and transfer directions of two adjacent faults on the same meter, the number of each type of fault state transfer group is the number of fault transfers with the same fault state code composition and transfer mode for two consecutive faults before and after the occurrence of the fault. The number of times can reflect the correlation between the two faults. The fault state transfer matrix is ​​generated based on the statistical results. The fault state transfer matrix is ​​composed of the transition probabilities between each fault state code, that is, each factor of the fault state transfer matrix represents the probability that one fault will cause another fault to occur next.

[0043] After determining the target fault state code for the currently faulty meter, the fault analysis result for the currently faulty meter can be determined based on the fault state transition matrix. This includes the possible fault types and corresponding probabilities of the current meter's secondary fault. Based on this fault analysis result, operations and maintenance personnel can inspect for potential secondary faults when the current fault occurs to prevent them. Furthermore, after a secondary fault occurs, they can prioritize fault types with a high probability of occurrence, improving the efficiency of fault inspection for the faulty meter during maintenance. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are merely embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without any creative work.

[0045] Figure 1 This is a flow chart of a method for analyzing and judging electric meter faults disclosed in this application;

[0046] Figure 2This is a structural block diagram of an electric meter fault analysis and judgment device disclosed in this application;

[0047] Figure 3 This is a hardware structure block diagram of an electric meter fault analysis and judgment device disclosed in this application. DETAILED DESCRIPTION

[0048] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0049] Next, we will introduce the application scheme. This application proposes the following technical scheme, please see below for details.

[0050] Figure 1 This is a flow chart of a method for analyzing and judging an electric meter fault disclosed in an embodiment of the present application. Figure 1 As shown, the method may include:

[0051] Step S1: Obtain historical fault information of each electricity meter and fault information of the current faulty electricity meter.

[0052] Specifically, historical fault information for each meter is obtained, specifically historical fault operation and maintenance work orders for low-voltage user smart meters. This historical fault information includes the asset number of the smart meter that experienced the fault, the operating region, the time of occurrence, and the cause of the fault. Each fault is recorded in the historical fault information.

[0053] For example, if meter 1 experiences its first fault, a fault message is generated, recording the time and cause of the fault. Some time later, meter 1 experiences its second fault, and another fault record is generated, recording the time and cause of the second fault. In this power grid, any meter fault generates a corresponding fault message, and all generated fault messages constitute historical fault information.

[0054] The fault information of the current faulty meter is the current fault information corresponding to the faulty meter that is currently faulty. Based on the current fault information of the meter, the possible fault type and corresponding probability of the next fault of the meter need to be analyzed.

[0055] Step S2: Determine the fault status code corresponding to each piece of fault information in the historical fault information.

[0056] Specifically, for multiple fault information in the historical fault information, different fault status codes are set according to the fault cause contained in each fault information, that is, the cause of each fault information, according to different cause types, and each code can correspond to a specific type of fault.

[0057] As shown in Table 1, there are four fault records in the historical fault information. The equipment codes for TA open circuit, TV blown fuse, and meter box damage are 101, 102, and 103 respectively. The fault status codes corresponding to the four fault records are 101, 103, 101, and 102 respectively.

[0058] Faulty electric meter Cause of failure Fault status code 1 Meter 1 first failure TA open circuit 101 2 Meter 2 first failure Meter box damaged 103 3 Meter 3 first failure TA open circuit 101 4 Meter 1 fails for the second time TV fuse blown 102

[0059] Table 1

[0060] After determining the fault status code corresponding to each piece of fault information in the historical fault information in sequence, the number of faults can be counted based on the fault status code.

[0061] Step S3: Count the types of fault state transfer groups and the number of each type of fault state transfer groups in the historical fault information, and generate a fault state transfer matrix based on the statistical results.

[0062] Specifically, the fault state transfer group includes the fault state codes and transfer directions of two adjacent faults on the same electric meter, and the fault state transfer matrix is ​​composed of the transfer probabilities between the fault state codes.

[0063] For example, for a certain electricity meter, there are 4 fault information in total. According to the order in which the faults occur, the fault status codes are 101, 102, 101, and 103. Then, for this electricity meter, there will be three different fault status transition groups, namely 101→102, 102→101, and 101→103. Fault status transition groups that contain the same fault status code and transition direction are the same fault status transition group.

[0064] First, determine the types of fault state transition groups in historical fault information, and then calculate the number of each type of fault state transition group in the historical fault information. A large number of fault state transition groups indicates a strong correlation between the two faults represented by these fault state transition groups. After a fault occurs, the probability of a different specific fault occurring a second time is high. Therefore, a fault state transition matrix can be generated based on the statistical results. This fault state transition matrix consists of the transition probabilities between each fault state code. Each factor in the fault state transition matrix represents the probability of a different fault occurring the next time a fault occurs.

[0065] For example, if S i to S jThe number of state transitions is shown in the following table. By counting the number of different fault state transition groups, Table 2 can be generated. a,b Indicates S a Transfer to S b The number of transfers, S k and S m Fault characteristic state codes representing two different fault types, N 1,k The number of fault characteristic state codes for S1 to transfer to the first fault type, N 1,m The number of fault characteristic state codes of the mth type when S1 transfers to another fault type.

[0066]

[0067] Table 2

[0068] Number of transfers N a,b The cumulative ratio of the number of transitions from state i to state j can be used to obtain the corresponding probability P a,b Form the fault state transfer matrix T:

[0069]

[0070] Among them, P 1,0 Indicates that the characteristic state S1 is transferred to S j The characteristic state transition probability of .

[0071] Step S4: Determine a target fault state code corresponding to the fault information of the current faulty meter according to the fault information of the current faulty meter.

[0072] Specifically, when a new fault operation and maintenance work order is detected, that is, the fault information of the current faulty meter is generated, the fault status code corresponding to the current fault information can be determined based on the fault cause contained in the fault information of the current meter, that is, the cause of the current fault information.

[0073] Step S5: Determine the fault analysis result of the current faulty electric meter according to the fault state transfer matrix and the target fault state code.

[0074] Specifically, the fault analysis results include the possible fault type and corresponding probability of the next fault. Based on the target fault state code corresponding to the fault information of the current faulty meter, the previously generated fault state transition matrix can be used to determine the probability of other faults occurring within the cycle time or the probability of normal behavior.

[0075] In practical applications, the fault analysis results can be used to avoid secondary faults and quickly detect them when they occur. While addressing the current fault, maintenance personnel can check for potential secondary faults and perform repairs in advance to prevent them. After a secondary fault occurs, when determining the specific cause, they can prioritize testing for fault types with a high probability, quickly identifying the cause and improving the efficiency of fault detection for faulty meters during maintenance.

[0076] As can be seen from the above technical solution, the embodiment of the present application provides a method for analyzing and diagnosing electric meter faults. By obtaining the historical fault information of each electric meter and the fault information of the current faulty electric meter, the fault state code corresponding to each fault information in the historical fault information is determined, and the historical fault information of each electric meter is statistically analyzed to obtain the types of fault state transfer groups in the historical fault information and the number of each type of fault state transfer group. A fault state transfer matrix is ​​generated based on the statistical results. Since the fault state transfer group includes the fault state codes and transfer directions of two adjacent faults on the same electric meter, the number of each type of fault state transfer group is the number of fault transfers with the same fault state code composition and transfer mode for two consecutive faults before and after the occurrence. The number of times can reflect the correlation between the two faults. The fault state transfer matrix is ​​generated based on the statistical results. The fault state transfer matrix is ​​composed of the transition probabilities between each fault state code, that is, each factor of the fault state transfer matrix represents the probability that one fault will lead to another fault occurring next.

[0077] After determining the target fault state code for the currently faulty meter, the fault analysis result for the currently faulty meter can be determined based on the fault state transition matrix. This includes the possible fault types and corresponding probabilities of the current meter's secondary fault. Based on this fault analysis result, operations and maintenance personnel can inspect for potential secondary faults when the current fault occurs to prevent them. Furthermore, after a secondary fault occurs, they can prioritize fault types with a high probability of occurrence, improving the efficiency of fault inspection for the faulty meter during maintenance.

[0078] In some embodiments of the present application, the process of step S2, determining the fault status code corresponding to each piece of fault information in the historical fault information, is introduced, which may specifically include:

[0079] Step S21: Determine the fault type and the code of each fault type for each piece of fault information in the historical fault information.

[0080] Step S22: Determine the code as the fault status code of the corresponding fault information.

[0081] Specifically, for the type of fault, when encoding, it is necessary to determine the fault type to which each fault information in the historical fault information belongs, then determine the encoding of the fault information under the fault type to which it belongs, and determine the encoding as the fault status encoding of the corresponding fault information.

[0082] Optionally, the fault type can be divided into a data anomaly type and a device anomaly type, that is, the historical fault information can be further divided into a data anomaly type and a device anomaly type.

[0083] Specifically, based on the software and hardware characteristics of the fault, fault types can be divided into data anomalies and equipment anomalies. Equipment anomalies may include TA open circuit, TV blown fuse, meter box damage, battery failure, electrical mechanical failure, electrical failure burnout, TA overload, meter burnout due to overload, transformer ratio error, wiring error, pulse sampling failure, clock disorder, system freeze, communication function failure, display screen failure, etc. Other data anomalies are all data anomalies.

[0084] For different fault types, the implementation of step S21, determining the fault type to which each piece of fault information in the historical fault information belongs and encoding under the fault type also varies. The specific implementation process of step S21 is introduced for two different fault types.

[0085] ①For each fault information belonging to data anomaly type:

[0086] The first step is to obtain the electricity consumption curve of each electricity meter with data anomaly within a preset time range and generate an electricity consumption curve database.

[0087] Specifically, the power consumption curve for each meter experiencing data anomalies within a preset time range can be obtained to generate a power consumption curve database. The power consumption curve database contains multiple power consumption curves corresponding to data anomalies. For each data anomaly, a corresponding power consumption curve within the preset range is obtained. The preset time range is manually set, generally from three days before to four days after the data anomaly occurs. The power consumption curve data is generally based on 96 load data points per day. For low-voltage users that do not meet the requirements, the power consumption data is collected at one point per day.

[0088] The second step is to cluster the electricity consumption curve database to obtain several clusters, use the median curve of the electricity consumption curve in each cluster as the class characteristic curve, and define the data anomaly subtype and code of each cluster.

[0089] Specifically, clustering is performed on the electricity consumption curve database. K-means clustering algorithm, x-means clustering algorithm, etc. are used for clustering. If x-means clustering algorithm is used, the clustering can preliminarily propose a cluster number interval (k min ,k max ), and according to expert experience, take k min =2, Here, n represents the number of curves in the fault curve database. After clustering generates several clusters, the median curve of the power consumption curve in each cluster can be used as the cluster characteristic curve. The data anomaly subtype and code for each cluster are defined. The anomaly type of the cluster serves as the anomaly type for each fault message within the cluster, and the code of the cluster serves as the code for each fault message within the cluster. At this point, the code for each fault message stored in the power consumption curve database can be determined and is encoded as the code of the cluster to which the corresponding power consumption curve belongs.

[0090] Optionally, before generating the power consumption curve database, the power consumption curve may be normalized.

[0091] Specifically, considering that in actual applications, the capacity of the power lines connected to users is not exactly the same, the units of the smart meters may be different, resulting in that for the power consumption curve with the same change percentage, the large-capacity power consumption curve has a larger upper and lower amplitude range, while the small-capacity power consumption curve has a smaller curve amplitude range. What we actually need to care about is the sudden change of power consumption in the power consumption curve. Therefore, in order to avoid the impact of different power line capacities and different units of smart meters, for data abnormality type faults, it is also necessary to perform unit processing, that is, to perform per-unit processing on them. The per-unit processing can use maximum normalization, Z-score normalization, minimum-maximum value normalization and other methods.

[0092] Optionally, the process of defining the encoding of each cluster may include:

[0093] The code of each cluster is determined according to the arrangement order of the class characteristic curves of each cluster.

[0094] For data anomaly type faults, sequential coding is performed according to the arrangement order of the class characteristic curves of each cluster, and the sequential coding is used as the coding of each cluster.

[0095] ②For each fault information belonging to the equipment abnormality type:

[0096] Define the device exception subtype and code for each exception under the device exception type.

[0097] It is understood that the encoding method of this application is not limited to the above method. Any method that can classify fault information and set a corresponding unique code for each category should fall within the scope of protection of this application. This application uses codes instead of fault causes, that is, classifying fault information according to the fault cause, the purpose of which is to calculate the transition probability between different types of faults.

[0098] Based on the above embodiment, if the fault information of the current faulty meter belongs to the fault information of the data abnormality type, the process of determining the target fault state code corresponding to the fault information of the current faulty meter in step S4 may specifically include:

[0099] Step S41: Obtain a target power consumption curve within a preset time range of the currently faulty electric meter.

[0100] Step S42: performing similarity evaluation on the target power consumption curve and the class characteristic curve of each cluster, and taking the code of the cluster where the class characteristic curve with the highest similarity belongs as the target fault state code.

[0101] Specifically, for fault information belonging to the data anomaly type, when a fault operation and maintenance work order belonging to the data anomaly type is found, the electricity consumption curve of the faulty meter three days before the fault and four days after the fault is selected and normalized, and then a similarity evaluation is performed with the class characteristic curve of each cluster cluster. The class characteristic curve with the highest similarity is taken, and the code of the cluster cluster to which the class characteristic curve belongs is determined, and the code of the cluster cluster is used as the target fault state code.

[0102] Furthermore, the similarity evaluation method can be adopted to calculate the Euclidean distance between the target power consumption curve and the class characteristic curve, specifically:

[0103] The Euclidean distance between the target power consumption curve and the class characteristic curve of each cluster is calculated one by one, and the code of the cluster where the class characteristic curve with the smallest Euclidean distance belongs is used as the target fault state code.

[0104] Specifically, the electricity consumption curves of the faulty electricity meter three days before and four days after the fault are selected and normalized, and the Euclidean distance between the class characteristic curve of each cluster and the electricity consumption curve is calculated one by one. The class characteristic curve with the smallest Euclidean distance is taken, and the code of the cluster to which the class characteristic curve with the smallest Euclidean distance belongs is determined, and the code of the cluster is used as the target fault state code.

[0105] The following describes an electric meter fault analysis device provided in an embodiment of the present application. The electric meter fault analysis device described below and the electric meter fault analysis method described above can be referenced to each other.

[0106] See also Figure 2 , Figure 2 This is a schematic diagram of the structure of an electric meter fault analysis and judgment device disclosed in an embodiment of the present application.

[0107] like Figure 2 As shown, the device may include:

[0108] An acquisition unit 110 is configured to acquire historical fault information of each electric meter and fault information of a currently faulty electric meter;

[0109] The first encoding unit 120 is used to determine the fault status code corresponding to each piece of fault information in the historical fault information;

[0110] a statistical unit 130 configured to count the types of fault state transition groups and the number of each type of fault state transition group in the historical fault information, and generate a fault state transition matrix based on the statistical results, wherein the fault state transition group includes the fault state codes and transition directions of two adjacent faults on the same electric meter, and the fault state transition matrix is ​​composed of transition probabilities between the fault state codes;

[0111] The second encoding unit 140 is configured to determine a target fault state code corresponding to the fault information of the current faulty meter according to the fault information of the current faulty meter;

[0112] The analysis unit 150 is used to determine the fault analysis result of the current faulty electric meter according to the fault state transfer matrix and the target fault state code, wherein the fault analysis result includes the possible fault type and corresponding probability of the next fault.

[0113] As can be seen from the above technical solution, the embodiment of the present application provides an electric meter fault analysis device, which obtains the historical fault information of each electric meter and the fault information of the current fault electric meter, determines the fault state code corresponding to each fault information in the historical fault information, performs statistical analysis on the historical fault information of each electric meter, obtains the type of fault state transfer group in the historical fault information and the number of each type of fault state transfer group, and generates a fault state transfer matrix based on the statistical results. Since the fault state transfer group includes the fault state codes and transfer directions of two adjacent faults on the same electric meter, the number of each type of fault state transfer group is the number of fault transfers with the same fault state code composition and transfer mode for two consecutive faults before and after the occurrence. The number of times can reflect the correlation between the two faults. The fault state transfer matrix is ​​generated based on the statistical results. The fault state transfer matrix is ​​composed of the transition probabilities between each fault state code, that is, each factor of the fault state transfer matrix represents the probability that one fault will cause another fault to occur next.

[0114] After determining the target fault state code for the currently faulty meter, the fault analysis result for the currently faulty meter can be determined based on the fault state transition matrix. This includes the possible fault types and corresponding probabilities of the current meter's secondary fault. Based on this fault analysis result, operations and maintenance personnel can inspect for potential secondary faults when the current fault occurs to prevent them. Furthermore, after a secondary fault occurs, they can prioritize fault types with a high probability of occurrence, improving the efficiency of fault inspection for the faulty meter during maintenance.

[0115] Optionally, the process of the first encoding unit determining the fault status code corresponding to each piece of fault information in the historical fault information may include:

[0116] Determine the fault type and the code of each fault type for each piece of fault information in the historical fault information;

[0117] The code is determined as the fault status code of the corresponding fault information.

[0118] Optionally, the process of the first encoding unit determining the fault type of each piece of fault information in the historical fault information and the encoding of the fault type may include:

[0119] Classifying the historical fault information into data anomaly types and device anomaly types;

[0120] For each fault message belonging to the data anomaly type:

[0121] Obtaining the electricity consumption curve of each electricity meter with data anomalies within a preset time range and generating an electricity consumption curve database;

[0122] Clustering the electricity consumption curve database to obtain a plurality of clusters, using the median curve of the electricity consumption curve in each cluster as a class characteristic curve, and defining a data anomaly subtype and code for each cluster, wherein the anomaly type of the cluster serves as the anomaly type to which each fault information in the cluster belongs, and the code of the cluster serves as the code for each fault information in the cluster;

[0123] For each fault message belonging to the device abnormality type:

[0124] Define the device exception subtype and code for each exception under the device exception type.

[0125] Optionally, the first encoding unit may be further configured to perform normalization processing on the power consumption curve before generating the power consumption curve database.

[0126] Optionally, if the fault information of the current faulty meter is fault information of a data anomaly type, the process of the second encoding unit determining the target fault state code corresponding to the fault information of the current faulty meter may include:

[0127] Obtaining a target power consumption curve within a preset time range of the currently faulty electric meter;

[0128] A similarity evaluation is performed on the target power consumption curve and the class characteristic curve of each cluster, and the code of the cluster where the class characteristic curve with the highest similarity belongs is used as the target fault state code.

[0129] Optionally, the second encoding unit may perform similarity evaluation on the target power consumption curve and the class characteristic curve of each cluster, and use the code of the cluster containing the class characteristic curve with the highest similarity as the target fault state code, which may include:

[0130] The Euclidean distance between the target power consumption curve and the class characteristic curve of each cluster is calculated one by one, and the code of the cluster where the class characteristic curve with the smallest Euclidean distance belongs is used as the target fault state code.

[0131] Optionally, the first encoding unit executing the encoding process of defining each cluster may include:

[0132] The code of each cluster is determined according to the arrangement order of the class characteristic curves of each cluster.

[0133] The electric meter fault analysis and judgment device provided in the embodiment of the present application can be applied to electric meter fault analysis and judgment equipment. Figure 3 The hardware structure diagram of the meter fault analysis equipment is shown. Figure 3 ,The hardware structure of the electric meter fault analysis and judgment device may include: at least one processor 1, at least one communication interface 2, at least one memory 3 and at least one communication bus 4;

[0134] In the embodiment of the present application, the number of the processor 1, the communication interface 2, the memory 3, and the communication bus 4 is at least one, and the processor 1, the communication interface 2, and the memory 3 communicate with each other through the communication bus 4;

[0135] The processor 1 may be a central processing unit (CPU), or an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present invention;

[0136] The memory 3 may include a high-speed RAM memory, and may also include a non-volatile memory, such as at least one disk memory;

[0137] The memory stores a program, and the processor can call the program stored in the memory, wherein the program is used to:

[0138] Obtain historical fault information of each meter and fault information of the current faulty meter;

[0139] Determine the fault status code corresponding to each piece of fault information in the historical fault information;

[0140] Counting the types of fault state transition groups and the number of each type of fault state transition groups in the historical fault information, and generating a fault state transition matrix based on the statistical results, wherein the fault state transition group includes the fault state codes and transition directions of two adjacent faults on the same electricity meter, and the fault state transition matrix is ​​composed of transition probabilities between each fault state code;

[0141] Determining a target fault state code corresponding to the fault information of the current faulty electric meter according to the fault information of the current faulty electric meter;

[0142] According to the fault state transfer matrix and the target fault state code, a fault analysis result of the current faulty electric meter is determined, wherein the fault analysis result includes a possible fault type and a corresponding probability of the next fault.

[0143] Optionally, the refined functions and extended functions of the program may refer to the above description.

[0144] The present application also provides a readable storage medium, which may store a program suitable for execution by a processor, wherein the program is used to:

[0145] Obtain historical fault information of each meter and fault information of the current faulty meter;

[0146] Determine the fault status code corresponding to each piece of fault information in the historical fault information;

[0147] Counting the types of fault state transition groups and the number of each type of fault state transition groups in the historical fault information, and generating a fault state transition matrix based on the statistical results, wherein the fault state transition group includes the fault state codes and transition directions of two adjacent faults on the same electricity meter, and the fault state transition matrix is ​​composed of transition probabilities between each fault state code;

[0148] Determining a target fault state code corresponding to the fault information of the current faulty electric meter according to the fault information of the current faulty electric meter;

[0149] According to the fault state transfer matrix and the target fault state code, a fault analysis result of the current faulty electric meter is determined, wherein the fault analysis result includes a possible fault type and a corresponding probability of the next fault.

[0150] Optionally, the refined functions and extended functions of the program may refer to the above description.

[0151] Finally, it should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or device. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of additional identical elements in the process, method, article, or device comprising the element.

[0152] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0153] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present application. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application is not limited to the embodiments shown herein, but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for analyzing and judging an electric meter fault, characterized in that: include: Obtain historical fault information of each meter and fault information of the current faulty meter; Determining the fault status code corresponding to each piece of fault information in the historical fault information, including respectively determining the fault type and the code under the fault type for each piece of fault information in the historical fault information, and determining the code as the fault status code of the corresponding fault information; Counting the types of fault state transition groups and the number of each type of fault state transition groups in the historical fault information, and generating a fault state transition matrix based on the statistical results, wherein the fault state transition group includes the fault state codes and transition directions of two adjacent faults on the same electricity meter, and the fault state transition matrix is ​​composed of transition probabilities between each fault state code; Determining a target fault state code corresponding to the fault information of the current faulty electric meter according to the fault information of the current faulty electric meter; Determine a fault analysis result of the current faulty electric meter according to the fault state transfer matrix and the target fault state code, wherein the fault analysis result includes a possible fault type and corresponding probability of the next fault; The step of respectively determining the fault type and the code of each piece of fault information in the historical fault information includes: Classifying the historical fault information into data anomaly types and device anomaly types; For each fault message belonging to the data anomaly type: Obtaining the electricity consumption curve of each electricity meter with data anomalies within a preset time range and generating an electricity consumption curve database; Clustering the electricity consumption curve database to obtain a plurality of clusters, using the median curve of the electricity consumption curve in each cluster as a class characteristic curve, and defining a data anomaly subtype and code for each cluster, wherein the anomaly type of the cluster serves as the anomaly type to which each fault information in the cluster belongs, and the code of the cluster serves as the code for each fault information in the cluster; For each fault message belonging to the device abnormality type: Define the device exception subtype and code for each exception under the device exception type.

2. The method according to claim 1, characterized in that Before generating the electricity consumption curve database, it also includes: The power consumption curve is normalized.

3. The method according to any one of claims 1 or 2, characterized in that If the fault information of the current faulty electric meter belongs to the fault information of the data abnormality type, the process of determining the target fault state code corresponding to the fault information of the current faulty electric meter includes: Obtaining a target power consumption curve within a preset time range of the currently faulty electric meter; A similarity evaluation is performed on the target power consumption curve and the class characteristic curve of each cluster, and the code of the cluster where the class characteristic curve with the highest similarity belongs is used as the target fault state code.

4. The method according to claim 3, characterized in that The target power consumption curve and the class characteristic curve of each cluster are evaluated for similarity, and the code of the cluster where the class characteristic curve with the highest similarity belongs is used as the target fault state code, including: The Euclidean distance between the target power consumption curve and the class characteristic curve of each cluster is calculated one by one, and the code of the cluster where the class characteristic curve with the smallest Euclidean distance belongs is used as the target fault state code.

5. The method according to claim 1, wherein The code for defining each cluster includes: The code of each cluster is determined according to the arrangement order of the class characteristic curves of each cluster.

6. An electric meter fault analysis and judgment device, characterized in that: include: An acquisition unit, used to acquire historical fault information of each electric meter and fault information of the current faulty electric meter; a first encoding unit, configured to determine a fault status code corresponding to each piece of fault information in the historical fault information, including determining the fault type and the code for each piece of fault information in the historical fault information, and determining the code as the fault status code of the corresponding fault information; a statistical unit, configured to count the types of fault state transition groups and the number of each type of fault state transition groups in the historical fault information, and generate a fault state transition matrix based on the statistical results, wherein the fault state transition group includes the fault state codes and transition directions of two adjacent faults on the same electric meter, and the fault state transition matrix is ​​composed of transition probabilities between the fault state codes; A second encoding unit is used to determine a target fault state code corresponding to the fault information of the current faulty electric meter according to the fault information of the current faulty electric meter; a judgment unit, configured to determine a fault judgment result of the current faulty electric meter according to the fault state transfer matrix and the target fault state code, wherein the fault judgment result includes a possible fault type and corresponding probability of the next fault; The step of respectively determining the fault type and the code of each piece of fault information in the historical fault information includes: Classifying the historical fault information into data anomaly types and device anomaly types; For each fault message belonging to the data anomaly type: Obtaining the electricity consumption curve of each electricity meter with data anomalies within a preset time range and generating an electricity consumption curve database; Clustering the electricity consumption curve database to obtain a plurality of clusters, using the median curve of the electricity consumption curve in each cluster as a class characteristic curve, and defining a data anomaly subtype and code for each cluster, wherein the anomaly type of the cluster serves as the anomaly type to which each fault information in the cluster belongs, and the code of the cluster serves as the code for each fault information in the cluster; For each fault message belonging to the device abnormality type: Define the device exception subtype and code for each exception under the device exception type.

7. An electric meter fault analysis and judgment device, characterized in that: including memory and processor; The memory is used to store programs; The processor is used to execute the program to implement each step of the electric meter fault analysis method according to any one of claims 1 to 5.

8. A readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, each step of the electric meter fault analysis method according to any one of claims 1 to 5 is implemented.

Images