CN-121997151-A - Cross-equipment power adapter fault diagnosis method and system

CN121997151ACN 121997151 ACN121997151 ACN 121997151ACN-121997151-A

Abstract

The invention relates to the technical field of fault diagnosis and discloses a cross-equipment power adapter fault diagnosis method and a system, wherein the method comprises the steps of collecting operation data sets of multi-model equipment; extracting general feature vectors, calculating deviation values of the general feature vectors and preset normal operation vectors, marking the general feature vectors as abnormal vectors and classifying the abnormal vectors if the deviation values exceed a preset deviation tolerance threshold value to obtain fault classification labels, uniformly correcting fault feature vectors matched with the labels across equipment, extracting fault distribution features, combining real-time feature vectors, extracting standardized fault modes and integrating the standardized fault modes to obtain initial diagnosis standards, carrying out matching analysis on real-time operation parameters and the initial diagnosis standards, obtaining optimal diagnosis standards according to analysis results, monitoring the operation states of all the equipment, and diagnosing the equipment by using the optimal diagnosis standards to obtain final fault diagnosis results. The method can unify the characteristic data of different devices under the same type of faults and flexibly cope with the differences among the devices.

Inventors

LI JIANJUN
LU JINGUI
LU TAOYUAN

Assignees

深圳市华锦电子有限公司

Dates

Publication Date: 20260508
Application Date: 20251211

Claims (9)

1. A method for diagnosing faults of a power adapter of a cross-device, comprising: Collecting operation data and historical operation records of multi-model equipment, and combining the operation data and the historical operation records and correcting to obtain an operation data set, wherein the operation data comprises voltage data, current data and temperature data; Extracting a core parameter combination from the operation data set, performing cross-equipment association analysis on the core parameter combination, generating a general feature set according to an analysis result, and performing vectorization processing to obtain a general feature vector; calculating the deviation value of the general feature vector and a preset normal operation vector, and if the deviation value exceeds a preset deviation tolerance threshold, marking the general feature vector as an abnormal vector and classifying by using a support vector machine to obtain a fault classification label; Obtaining a fault feature vector matched with the fault classification label, carrying out cross-equipment unified correction on the fault feature vector, and extracting fault distribution features from the corrected vector; according to the fault distribution characteristics, combining real-time characteristic vectors obtained in real time, extracting and integrating a standardized fault mode to obtain an initial diagnosis standard suitable for cross-equipment diagnosis; Acquiring real-time operation parameters, carrying out matching analysis on the real-time operation parameters and the initial diagnosis standard, and optimizing the initial diagnosis standard according to an analysis result to obtain an optimized diagnosis standard; and monitoring the running state of each device and diagnosing by using the optimized diagnosis standard to obtain a final fault diagnosis result.
2. The method for diagnosing a power adapter failure according to claim 1, wherein said collecting operation data and history operation records of a plurality of types of devices, combining said operation data with said history operation records and correcting them to obtain an operation data set, comprises: acquiring voltage data, current data and temperature data from multi-model equipment according to a pre-established equipment identification mapping table, and recording equipment model identification to obtain first operation data; Extracting a history operation record corresponding to the equipment model identifier from a pre-established history record table, and merging the first operation data in the history operation record according to time sequence to obtain second operation data; If the value of the data in the second operation data exceeds a preset operation data threshold, correcting through linear interpolation to obtain third operation data, and if the value of the data in the second operation data does not exceed the preset operation data threshold, directly taking the second operation data as the third operation data; and carrying out format unified conversion on the third operation data to obtain an operation data set.
3. The power adapter fault diagnosis method according to claim 1, wherein the extracting a core parameter combination from the operation data set, performing cross-device association analysis on the core parameter combination, generating a general feature set according to an analysis result, and performing vectorization processing to obtain a general feature vector, includes: performing dimension reduction transformation on the voltage data, the current data and the temperature data in the operation data set by adopting a principal component analysis algorithm to obtain a core parameter combination; calculating the relevance value of each component in the core parameter combination; If the relevance value is larger than a preset relevance threshold, the universal feature set is included, and if the relevance value is smaller than the preset relevance threshold, the universal feature set is not included; assigning a corresponding general feature weight coefficient to each feature component of the general feature set to obtain a weighted feature sequence, wherein the general feature weight coefficient is subjected to normalization processing; and combining and packaging the weighted feature sequences according to a predefined vector format to obtain a universal feature vector.
4. The power adapter fault diagnosis method according to claim 1, wherein the calculating a deviation value between the general feature vector and a preset normal operation vector, if the deviation value exceeds a preset deviation tolerance threshold, marking the general feature vector as an abnormal vector and classifying by using a support vector machine to obtain a fault classification label includes: calculating the deviation value of the general feature vector and a preset normal operation vector; If the deviation value exceeds a preset deviation tolerance threshold, marking the universal feature vector as an abnormal vector; And classifying the abnormal vector by using a support vector machine to obtain a fault classification label.
5. The method for diagnosing a power adapter fault according to claim 1, wherein the step of performing cross-device unified correction on the fault feature vector and extracting fault distribution features from the corrected vector comprises: acquiring a hardware configuration parameter matrix of each device, and if the feature dimensions of different devices are not matched, performing zero padding operation on the low-dimensional feature vector to obtain a device feature matrix with uniform dimensions; Calculating characteristic offset among different types of equipment according to the equipment characteristic matrix, determining parameters of a linear transformation matrix, and performing iterative optimization if the characteristic offset exceeds a preset alignment precision threshold value until the characteristic offset meets the requirement, so as to obtain a transformation matrix aligned across equipment characteristics; according to the transformation matrix, performing space mapping transformation on the fault feature vector, and calculating the transformed distribution density to obtain an aligned feature distribution density map; And calculating the abnormality degree value of each feature vector relative to the neighborhood point according to the feature distribution density map to obtain fault distribution features.
6. The power adapter fault diagnosis method according to claim 1, wherein the extracting and integrating the standardized fault mode according to the fault distribution feature and in combination with the real-time feature vector obtained in real time to obtain the initial diagnosis standard suitable for cross-device diagnosis comprises: Classifying and combining similar fault modes through clustering according to the fault distribution characteristics to obtain limited types of fault modes, and normalizing characteristic parameters of the fault modes to obtain a standardized fault mode library; acquiring real-time operation parameters of equipment in operation, acquiring a time window segment according to the standardized fault mode library and the real-time operation parameters, and extracting corresponding real-time feature vectors from the time window segment; calculating the similarity between the standardized fault mode library and the real-time feature vector and carrying out weighted scoring to obtain a matching degree matrix; and summing the weighted characteristics of each fault mode in the matching degree matrix to obtain an initial diagnosis standard suitable for cross-equipment diagnosis.
7. The power adapter fault diagnosis method according to claim 1, wherein the performing a matching analysis on the real-time operation parameter and the initial diagnosis standard, optimizing the initial diagnosis standard according to the analysis result, and obtaining an optimized diagnosis standard includes: Calculating the matching degree value of the real-time operation parameter and the initial diagnosis standard; when the matching degree value is lower than a preset matching degree threshold value, calculating the deviation value of each component, and combining to obtain an initial parameter deviation vector; Step optimization processing is carried out on the initial parameter deviation vector by adopting a gradient descent optimization strategy, and the deviation change trend is monitored through an error feedback mechanism to obtain an optimal parameter updating step value; Setting a corresponding updating weight coefficient for key parameters in the initial diagnosis standard, carrying out iterative updating on the initial diagnosis standard according to the optimal parameter updating step length value, and obtaining an optimized diagnosis standard when the matching degree value of the updated diagnosis standard and the real-time operation parameter exceeds the matching degree threshold value.
8. The power adapter fault diagnosis method according to claim 1, wherein the monitoring the operation state of each device and diagnosing using the optimized diagnosis criteria to obtain a final fault diagnosis result comprises: Acquiring operation parameters to be diagnosed of multi-model equipment; Calculating the matching degree value of the operation parameters to be diagnosed and the optimized diagnosis standard; Acquiring historical operation data corresponding to equipment models, and calculating the association degree of the historical operation data and the operation parameters to be diagnosed to obtain historical data reference weight coefficients; weighting and fusing the matching degree value and the historical data reference weight coefficient to obtain a comprehensive diagnosis index; And when the comprehensive diagnosis index exceeds a preset fault judgment threshold value, calculating the confidence coefficient of each type of fault, and when the highest confidence coefficient value is larger than the preset confidence threshold value and the difference value between the highest confidence coefficient value and the second highest confidence coefficient exceeds the preset distinction threshold value, taking the corresponding fault type as a final fault diagnosis result.
9. A cross-device power adapter fault diagnosis system, comprising: the data acquisition and preprocessing module is used for acquiring operation data and historical operation records of the multi-model equipment, combining the operation data and the historical operation records and correcting the operation data to obtain an operation data set, wherein the operation data comprises voltage data, current data and temperature data; the feature extraction module is used for extracting a core parameter combination from the operation data set, analyzing the cross-equipment association degree of the core parameter combination, generating a general feature set according to an analysis result, and carrying out vectorization processing to obtain a general feature vector; The fault classification module is used for calculating the deviation value of the general feature vector and a preset normal operation vector, and if the deviation value exceeds a preset deviation tolerance threshold, the general feature vector is marked as an abnormal vector and classified by using a support vector machine to obtain a fault classification label; the cross-equipment unification module is used for acquiring fault feature vectors matched with the fault classification labels, performing cross-equipment unification correction on the fault feature vectors, and extracting fault distribution features from the corrected vectors; The primary diagnosis module is used for extracting and integrating a standardized fault mode according to the fault distribution characteristics and combining real-time feature vectors acquired in real time to obtain an initial diagnosis standard suitable for cross-equipment diagnosis; The diagnosis standard optimization module is used for acquiring real-time operation parameters, carrying out matching analysis on the real-time operation parameters and the initial diagnosis standard, and optimizing the initial diagnosis standard according to an analysis result to obtain an optimized diagnosis standard; and the final diagnosis module is used for monitoring the running state of each device and diagnosing by using the optimized diagnosis standard to obtain a final fault diagnosis result.

Description

Cross-equipment power adapter fault diagnosis method and system Technical Field The invention relates to the technical field of fault diagnosis, in particular to a cross-equipment power adapter fault diagnosis method and system. Background In modern industry and daily life, power adapters are used as energy supply cores for various electronic devices, and stable operation of the power adapters is critical to the safety and efficiency of the devices. The MCU (micro control unit) is used as a core control and data processing unit of the power adapter, and further determines the accurate monitoring and instruction execution capacity of the running state of the adapter. Therefore, aiming at the fault diagnosis technology of the power adapter, the fault diagnosis technology not only needs to cover the abnormal identification of key parameters such as voltage, current and the like, but also needs to combine the operation characteristics of the MCU to realize the imaging and accurate fault positioning of the adapter, and the fault positioning is directly related to the service life of equipment and the user experience. In the current mainstream method, the historical data of a specific device or fixed rules are generally relied on, in practical application, a model is trained by collecting the historical data of the single-model device, or fixed diagnosis rules or thresholds are formulated according to the physical principle and operation experience of the single-model device, and the running device data is judged based on the rules or thresholds. When the method is applied to cross-equipment, the influence of the control logic difference and the data output format difference of the MCU in the adapters with different models on fault characteristics is ignored, and the flexible coping capability of the difference between the equipment is lacking, so that the strategy needs to be frequently adjusted in actual use, the time consumption is increased, and the problem of low coordination efficiency exists. Disclosure of Invention The invention provides a fault diagnosis method and system for a cross-equipment power adapter, which are used for unifying characteristic data of different equipment under the same type of faults under the environment of multiple types of equipment and can flexibly cope with the difference among the equipment. In order to solve the above technical problems, the present invention provides a method for diagnosing faults of a power adapter of a cross-device, including: Collecting operation data and historical operation records of multi-model equipment, and combining the operation data and the historical operation records and correcting to obtain an operation data set, wherein the operation data comprises voltage data, current data and temperature data; Extracting a core parameter combination from the operation data set, performing cross-equipment association analysis on the core parameter combination, generating a general feature set according to an analysis result, and performing vectorization processing to obtain a general feature vector; calculating the deviation value of the general feature vector and a preset normal operation vector, and if the deviation value exceeds a preset deviation tolerance threshold, marking the general feature vector as an abnormal vector and classifying by using a support vector machine to obtain a fault classification label; Obtaining a fault feature vector matched with the fault classification label, carrying out cross-equipment unified correction on the fault feature vector, and extracting fault distribution features from the corrected vector; according to the fault distribution characteristics, combining real-time characteristic vectors obtained in real time, extracting and integrating a standardized fault mode to obtain an initial diagnosis standard suitable for cross-equipment diagnosis; Acquiring real-time operation parameters, carrying out matching analysis on the real-time operation parameters and the initial diagnosis standard, and optimizing the initial diagnosis standard according to an analysis result to obtain an optimized diagnosis standard; and monitoring the running state of each device and diagnosing by using the optimized diagnosis standard to obtain a final fault diagnosis result. In a second aspect, an embodiment of the present invention further provides a system for diagnosing faults of a power adapter of a cross-device, including: the data acquisition and preprocessing module is used for acquiring operation data and historical operation records of the multi-model equipment, combining the operation data and the historical operation records and correcting the operation data to obtain an operation data set, wherein the operation data comprises voltage data, current data and temperature data; the feature extraction module is used for extracting a core parameter combination from the operation data set, analyzing the cross-equipment association degr