CN-121980256-A - Method and device for determining fault data label of charging equipment

CN121980256ACN 121980256 ACN121980256 ACN 121980256ACN-121980256-A

Abstract

The embodiment of the invention discloses a method and a device for determining fault data labels of charging equipment, wherein the method comprises the steps of obtaining a data sample comprising at least one charging abnormal code and fault weights of the charging abnormal codes under different fault conditions, and determining fault scores of the corresponding fault conditions according to the different fault weights corresponding to the fault conditions; the fault condition with the highest fault score is determined as the fault label of the data sample. The fault weight is used for representing importance of the charging anomaly code on fault condition marking, and the fault label comprises gun faults and/or module faults. Therefore, in the embodiment, the scores of all fault situations and the fault labels of the data samples are determined by using the importance of different charge abnormal codes for marking different fault situations, so that the accuracy of marking the fault labels for the data samples is higher, marking is more convenient, and sample marking with marking accuracy and marking efficiency is realized.

Inventors

ZHANG QIDONG
GAO HAI
WANG YUHAO
PAN ZHEN
CHEN YONGJIAN

Assignees

浙江小桔绿色能源科技有限公司

Dates

Publication Date: 20260505
Application Date: 20241024

Claims (13)

1. A method for determining a charging device failure data tag, the method comprising: Acquiring a data sample, wherein the data sample comprises at least one charging anomaly code; acquiring fault weights of the charging anomaly codes under different fault conditions, wherein the fault weights are used for representing the importance of the charging anomaly codes on fault condition labeling; Determining a fault score of the corresponding fault situation according to different fault weights corresponding to the fault situations; and determining the fault situation with the highest fault score as a fault label of the data sample, wherein the fault label comprises gun faults and/or module faults.
2. The method according to claim 1, wherein the method further comprises: acquiring an abnormal code set of the charging equipment, wherein the abnormal code set comprises a plurality of charging abnormal codes; Determining an abnormal weight matrix according to the relative importance of each charging abnormal code to fault labeling, wherein the abnormal weight matrix comprises labeling weights of each charging abnormal code, and the labeling weights are used for representing the importance of the charging abnormal code to fault labeling; Determining a fault weight matrix of each charging anomaly code according to the relative importance of each charging anomaly code to different fault conditions and the labeling weight of each charging anomaly code, wherein the fault weight matrix comprises the fault weights of the corresponding charging anomaly codes under different fault conditions.
3. The method of claim 2, wherein determining an anomaly weight matrix based on the relative importance of each of the charging anomaly codes to fault labeling, the anomaly weight matrix including labeling weights for each of the charging anomaly codes comprises: constructing a first judgment matrix, wherein the first judgment matrix comprises quantized values of the relative importance of different charging abnormal codes to fault labeling, and the quantized values are determined based on a preset scale method; Preprocessing the first judgment matrix, and determining a corresponding initial labeling matrix; Performing consistency check on the first judgment matrix based on the initial labeling matrix, and determining a check result of the first judgment matrix; Responding to the verification result representation verification passing, determining the initial labeling matrix as an abnormal weight matrix, wherein the abnormal weight matrix comprises labeling weights of the charging abnormal codes; and responding to the fact that the verification result representation verification fails, adjusting the first judgment matrix until the verification result representation verification corresponding to the adjusted first judgment matrix passes.
4. The method of claim 3, wherein preprocessing the first decision matrix to determine a corresponding initial labeling matrix comprises: Normalizing each column vector in the first judgment matrix to determine a first standard matrix; summing all row vectors in the first standard matrix to determine a first weight matrix; And normalizing the first weight matrix to determine an initial labeling matrix.
5. The method of claim 2, wherein determining the fault weight matrix for each of the charging anomaly codes based on the relative importance of each of the charging anomaly codes for different fault conditions and the labeling weights for each of the charging anomaly codes comprises: Constructing a second judgment matrix corresponding to a target abnormal code, wherein the target abnormal code is any charging abnormal code in the abnormal code set, the second judgment matrix comprises quantized values of relative importance of the target abnormal code on different fault condition labels, and the quantized values are determined based on a preset scale method; preprocessing the second judgment matrix to determine an initial fault matrix; Performing consistency check on the second judgment matrix based on the initial fault matrix, and determining a check result of the second judgment matrix; responding to the verification result representation verification, and determining an initial fault matrix as a fault labeling matrix of the target abnormal code; Multiplying the fault labeling matrix of the target abnormal code by the labeling weight to determine a fault weight matrix, wherein the fault weight matrix comprises the fault weights of the target abnormal code under different fault conditions.
6. The method of claim 5, wherein preprocessing the second decision matrix to determine an initial failure matrix comprises: Normalizing each column vector in the second judgment matrix to determine a second standard matrix; summing all row vectors in the second standard matrix to determine a second weight matrix; And normalizing the second weight matrix to determine an initial fault matrix.
7. The method of claim 5, wherein determining the fault weight matrix for each of the charging anomaly codes based on the relative importance of each of the charging anomaly codes for different fault conditions and the labeling weights for each of the charging anomaly codes further comprises: And responding to the fact that the verification result representation verification fails, adjusting the second judgment matrix until the verification result representation verification corresponding to the adjusted second judgment matrix passes.
8. The method of claim 1, wherein said determining a fault score for each fault condition based on each of said fault weights comprises: And summing the fault weights of the charging anomaly codes under the same fault condition, and determining the fault score of the corresponding fault condition.
9. The method according to claim 1, wherein the method further comprises: Marking the data sample based on the fault label so as to train a fault diagnosis model based on the marked data sample, wherein the fault diagnosis model is used for determining the fault reason of the charging equipment to be identified.
10. A determination apparatus for a charging device failure data tag, the apparatus comprising: a sample acquisition unit configured to acquire a data sample including at least one charging anomaly code; the weight acquisition unit is used for acquiring the fault weight of each charging anomaly code under different fault conditions, and the fault weight is used for representing the importance of the charging anomaly code on fault condition labeling; the score determining unit is used for determining a fault score of the corresponding fault situation according to different fault weights corresponding to the fault situations; And the label determining unit is used for determining the fault situation with the highest fault score as a fault label of the data sample, wherein the fault label comprises gun faults and/or module faults.
11. A computer program product, characterized in that it comprises a computer program/instruction which, when executed by a processor, implements the method of any of claims 1-9.
12. An electronic device comprising a memory and a processor, wherein the memory is configured to store one or more computer program instructions, wherein the one or more computer program instructions are executed by the processor to implement the method of any of claims 1-9.
13. A computer readable storage medium, characterized in that the computer readable storage medium has stored therein a computer program which, when executed by a processor, implements the method of any of claims 1-9.

Description

Method and device for determining fault data label of charging equipment Technical Field The invention relates to the technical field of computers, in particular to a method and a device for determining fault data labels of charging equipment. Background In the fault diagnosis task of the charging equipment, a high-quality sample set is a key for guaranteeing the performance of a model, and accurate sample labeling is a key for constructing the high-quality sample set. The existing sample labeling method comprises site investigation labeling, expert labeling and rule labeling. The field investigation labeling is to assign a dispatch maintainer to the field investigation fault type and finally label the sample. Expert labeling refers to labeling an expert according to experience on the basis of data in dimensions such as orders, worksheets and the like. Rule labeling refers to mining deterministic rules based on data of dimensions such as historical orders, historical worksheets and the like, and labeling with rules. However, the on-site investigation and labeling are long in time consumption, maintenance staff is required to commute to the station for investigation, expert labeling is long in time consumption and cannot be performed in batches, rule labeling subjectivity is high, the setting of labeling rules is greatly influenced by human factors, and labeling quality is difficult to guarantee. Disclosure of Invention Therefore, an object of the embodiments of the present invention is to provide a method and an apparatus for determining a fault data tag of a charging device, so as to achieve sample labeling with both labeling accuracy and labeling efficiency. In a first aspect, an embodiment of the present invention is directed to a method for determining a fault data tag of a charging device, where the method includes: Acquiring a data sample, wherein the data sample comprises at least one charging anomaly code; acquiring fault weights of the charging anomaly codes under different fault conditions, wherein the fault weights are used for representing the importance of the charging anomaly codes on fault condition labeling; Determining a fault score of the corresponding fault situation according to different fault weights corresponding to the fault situations; and determining the fault situation with the highest fault score as a fault label of the data sample, wherein the fault label comprises gun faults and/or module faults. Further, the method further comprises: acquiring an abnormal code set of the charging equipment, wherein the abnormal code set comprises a plurality of charging abnormal codes; Determining an abnormal weight matrix according to the relative importance of each charging abnormal code to fault labeling, wherein the abnormal weight matrix comprises labeling weights of each charging abnormal code, and the labeling weights are used for representing the importance of the charging abnormal code to fault labeling; Determining a fault weight matrix of each charging anomaly code according to the relative importance of each charging anomaly code to different fault conditions and the labeling weight of each charging anomaly code, wherein the fault weight matrix comprises the fault weights of the corresponding charging anomaly codes under different fault conditions. Further, determining an anomaly weight matrix according to the relative importance of each charging anomaly code to fault annotation, wherein the anomaly weight matrix comprises the annotation weights of each charging anomaly code, and the annotation weights comprise: constructing a first judgment matrix, wherein the first judgment matrix comprises quantized values of the relative importance of different charging abnormal codes to fault labeling, and the quantized values are determined based on a preset scale method; Preprocessing the first judgment matrix, and determining a corresponding initial labeling matrix; Performing consistency check on the first judgment matrix based on the initial labeling matrix, and determining a check result of the first judgment matrix; Responding to the verification result representation verification passing, determining the initial labeling matrix as an abnormal weight matrix, wherein the abnormal weight matrix comprises labeling weights of the charging abnormal codes; and responding to the fact that the verification result representation verification fails, adjusting the first judgment matrix until the verification result representation verification corresponding to the adjusted first judgment matrix passes. Further, the preprocessing the first judgment matrix, and determining the corresponding initial labeling matrix includes: Normalizing each column vector in the first judgment matrix to determine a first standard matrix; summing all row vectors in the first standard matrix to determine a first weight matrix; And normalizing the first weight matrix to determine an initial labeling matrix. Further, the