CN118068132B - Cable anomaly identification method and system based on time-frequency analysis - Google Patents
Cable anomaly identification method and system based on time-frequency analysis Download PDFInfo
- Publication number
- CN118068132B CN118068132B CN202410459647.4A CN202410459647A CN118068132B CN 118068132 B CN118068132 B CN 118068132B CN 202410459647 A CN202410459647 A CN 202410459647A CN 118068132 B CN118068132 B CN 118068132B
- Authority
- CN
- China
- Prior art keywords
- fault
- data
- fuzzy
- signal
- frequency
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000004458 analytical method Methods 0.000 title claims abstract description 84
- 238000000034 method Methods 0.000 title claims abstract description 51
- 238000005457 optimization Methods 0.000 claims abstract description 99
- 238000003745 diagnosis Methods 0.000 claims abstract description 71
- 230000008859 change Effects 0.000 claims abstract description 53
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 49
- 239000002245 particle Substances 0.000 claims abstract description 45
- 238000013507 mapping Methods 0.000 claims abstract description 26
- 238000006243 chemical reaction Methods 0.000 claims abstract description 23
- 238000000354 decomposition reaction Methods 0.000 claims abstract description 13
- 238000012360 testing method Methods 0.000 claims description 82
- 238000004088 simulation Methods 0.000 claims description 55
- 238000012545 processing Methods 0.000 claims description 44
- 238000011156 evaluation Methods 0.000 claims description 41
- 230000011218 segmentation Effects 0.000 claims description 35
- 238000007670 refining Methods 0.000 claims description 25
- 238000012795 verification Methods 0.000 claims description 25
- 230000000694 effects Effects 0.000 claims description 21
- 230000008569 process Effects 0.000 claims description 21
- 230000006872 improvement Effects 0.000 claims description 10
- 238000013461 design Methods 0.000 claims description 8
- 238000011867 re-evaluation Methods 0.000 claims description 7
- 238000010276 construction Methods 0.000 claims description 6
- 238000013480 data collection Methods 0.000 claims description 6
- 239000011159 matrix material Substances 0.000 claims description 6
- 230000009466 transformation Effects 0.000 claims description 4
- 230000001133 acceleration Effects 0.000 claims description 3
- 238000004904 shortening Methods 0.000 claims description 3
- 230000002159 abnormal effect Effects 0.000 abstract description 4
- 238000001228 spectrum Methods 0.000 abstract description 3
- 230000035945 sensitivity Effects 0.000 abstract description 2
- 238000010586 diagram Methods 0.000 description 8
- 238000005516 engineering process Methods 0.000 description 5
- 238000003066 decision tree Methods 0.000 description 4
- 239000000243 solution Substances 0.000 description 4
- 230000006698 induction Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 208000025174 PANDAS Diseases 0.000 description 2
- 208000021155 Paediatric autoimmune neuropsychiatric disorders associated with streptococcal infection Diseases 0.000 description 2
- 240000004718 Panda Species 0.000 description 2
- 235000016496 Panda oleosa Nutrition 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000007477 logistic regression Methods 0.000 description 2
- 230000002265 prevention Effects 0.000 description 2
- 230000003595 spectral effect Effects 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 230000005856 abnormality Effects 0.000 description 1
- 238000007635 classification algorithm Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000002790 cross-validation Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000010191 image analysis Methods 0.000 description 1
- 238000002347 injection Methods 0.000 description 1
- 239000007924 injection Substances 0.000 description 1
- 238000009413 insulation Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000008092 positive effect Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000011269 treatment regimen Methods 0.000 description 1
Landscapes
- Test And Diagnosis Of Digital Computers (AREA)
- Testing And Monitoring For Control Systems (AREA)
Abstract
The invention relates to the technical field of time-frequency analysis, in particular to a cable anomaly identification method and system based on time-frequency analysis. According to the invention, through the decomposition and mapping of the signals in the fuzzy set, the inherent fuzzy boundary and spectrum characteristics of the signals are accurately captured, so that the signal analysis is more careful, the recognition capability of fault modes is enhanced by the conversion of fuzzy values and the matching of a rule base, the fault diagnosis accuracy is improved, the key change points in time-frequency data are automatically recognized by utilizing a time sequence clustering algorithm, the data are effectively divided, abnormal signals are further captured acutely, the recognition sensitivity is improved, the application of a particle swarm optimization algorithm in parameter adjustment is optimized, the fault diagnosis flow is optimized, and the diagnosis efficiency and accuracy are improved.
Description
Technical Field
The invention relates to the technical field of time-frequency analysis, in particular to a cable anomaly identification method and system based on time-frequency analysis.
Background
The technical field of time-frequency analysis is a method for analyzing and processing signals, and is particularly suitable for unsteady signals, namely, the situation that the frequency characteristics of the signals change with time. Time-frequency analysis techniques enable the representation of the time-varying characteristics and spectral structure inherent in a signal by representing the signal simultaneously in time and frequency. This analysis method is important in many applications, such as signal processing, communication systems, and medical image analysis, and can provide more information than conventional time-domain or frequency-domain analysis.
The cable abnormality recognition method based on time-frequency analysis is a method for detecting and diagnosing cable faults by applying a time-frequency analysis technology. The purpose is to accurately and rapidly identify and locate abnormal conditions in the cable, such as short circuits, wire breaks or insulation damage. By carrying out time-frequency analysis on the signals transmitted by the cable, tiny abnormal signals can be detected in a complex cable system, so that the effects of preventing faults in advance, reducing economic loss and guaranteeing safe operation of a power grid are achieved.
The traditional method is not timely or accurate in fault detection due to the fact that time-varying characteristics and frequency spectrum structures of signals are not fully utilized in cable anomaly identification, key anomaly signals are ignored due to the fact that fuzzy logic processing and parameter optimization strategies are not adopted, or normal signals are misjudged to be anomaly, accuracy and processing efficiency of fault diagnosis are affected, an obvious short board is formed in aspects of fault prevention and timely response of a cable system, and power grid safety is affected or economic loss is caused.
Disclosure of Invention
The invention aims to solve the defects in the prior art, and provides a cable anomaly identification method and system based on time-frequency analysis.
In order to achieve the above purpose, the present invention adopts the following technical scheme: a cable anomaly identification method based on time-frequency analysis includes the following steps:
s1: on the basis of time-frequency data during the operation of the cable, carrying out signal decomposition and fuzzy set construction, and matching boundary conditions of fuzzy logic control to generate fuzzy characteristic identification data;
S2: based on the fuzzy characteristic identification data, performing fuzzy value conversion and rule base matching, and evaluating and classifying signal characteristics to generate a fuzzy data processing record;
S3: based on the fuzzy data processing record, applying a time sequence clustering algorithm, dividing the data into multiple sections with the same kind of characteristics by automatically identifying key change points in the time-frequency data, and generating an intelligent segmentation strategy;
S4: based on the intelligent segmentation strategy, analyzing target features in each paragraph, carrying out signal re-evaluation, refining fault type judgment, and generating depth logic evaluation information;
s5: based on the depth logic evaluation information, performing key parameter adjustment by using a particle swarm optimization algorithm, performing optimization test, adjusting analysis parameters by simulating fault conditions, optimizing a fault diagnosis process, and generating parameter optimization measures;
s6: based on the parameter optimization measures, testing the parameter optimization effect by simulating the differential fault situation, verifying the implementation effect of the optimization strategy in the application, and generating a simulation test verification result;
S7: and according to the simulation test verification result, refining the fault type by analyzing the feedback of the fault test, and carrying out induction summary to generate a fault diagnosis log.
The invention improves, the said fuzzy characteristic discernment data includes signal amplitude level, frequency change interval, fuzzy boundary divide, the said fuzzy data processing record includes the degree of matching of the failure mode, signal classification basis, trouble prediction accuracy, the said intelligent segmentation tactics include segmentation starting point, leading frequency characteristic, energy concentration area, the said depth logic assessment information includes subdivided fault type, characteristic matching degree score, signal reevaluation result, the said parameter optimization measure includes window size adjustment, step optimization, signal analysis accuracy promote the goal, the said simulation tests the verification result includes the improvement point of the failure recognition rate, analysis time shortens the point, optimize parameter efficiency comparison, the said fault diagnosis log includes diagnosing the key step record, parameter adjustment history.
The invention improves, based on the time-frequency data of the cable operation, the signal decomposition and fuzzy set construction are carried out, the boundary condition of fuzzy logic control is matched, and the step of generating fuzzy characteristic identification data is specifically as follows:
S101: based on time-frequency data of the cable in operation, scanning the whole data set, detecting signal amplitude and frequency, identifying energy increase or frequency change points, marking the energy increase or frequency change points as key change points in the signal, and generating a key change point mark set;
S102: based on the key change point mark set, decomposing the signals, distributing the signals to the differential fuzzy set according to the change of energy and frequency, distributing corresponding fuzzy grades for each signal segment, and generating a signal fuzzy grade mapping table;
S103: and evaluating each fuzzy grade based on the signal fuzzy grade mapping table, determining the matching degree with a fuzzy logic control boundary, determining the signal fuzzy classification and generating fuzzy characteristic identification data.
The invention improves, based on the fuzzy characteristic identification data, the fuzzy value conversion is matched with a rule base, the evaluation and classification of signal characteristics are carried out, and the step of generating a fuzzy data processing record comprises the following steps:
S201: based on the fuzzy characteristic identification data, carrying out numerical conversion on each fuzzy classified signal, converting a fuzzy grade into a processable fuzzy value, and generating a fuzzy value conversion set;
s202: based on the fuzzy value conversion set, executing a fault mode matching process, comparing each converted fuzzy value with a predefined fault mode rule base, identifying a key fault mode, and generating fault mode identification data;
S203: and classifying the signals according to the matched fault modes based on the fault mode identification data, determining the category of each signal, and generating a fuzzy data processing record.
The invention improves, based on the fuzzy data processing record, the time sequence clustering algorithm is applied, the data is divided into multiple sections with the same kind of characteristics by automatically identifying key change points in the time data, and the steps for generating the intelligent segmentation strategy are specifically as follows:
S301: based on the fuzzy data processing record, analyzing key features in time-frequency data by utilizing wavelet transformation, wherein the key features comprise energy peaks and frequency changes, identifying a change interval of the data and generating key feature interval data;
s302: based on the key characteristic interval data, a time sequence clustering algorithm is applied to divide the time frequency data into a plurality of paragraphs according to the change characteristics, and each paragraph comprises data of the same kind of energy and frequency characteristics to generate segmentation mapping data;
The time sequence clustering algorithm is as follows:
；
Calculating similarity between time-frequency data Generating segment mapping data;
Wherein, AndRespectively representing two time-series data sequences,AndRespectively represent sequencesAndIs provided for the length of (a),Representing sequencesMiddle (f)Individual points and sequencesMiddle (f)The distance between the points of interest is such that,As a matrix of weights, the weight matrix,Is thatCorresponding to (a)AndIs used for the weight value of (a),AndIn order to adjust the parameters of the device,Is a sequenceMiddle (f)The additional feature values of the individual points are,Is a sequenceMiddle (f)Additional feature values for the individual points;
S303: based on the segment map data, applying feature map logic to each data segment, distinguishing signal characteristics of difference types, and refining classification of the data segments to generate an intelligent segment strategy.
The invention improves, based on the intelligent segmentation strategy, by analyzing the target characteristics in each section, the signal is estimated again, the fault type judgment is refined, and the step of generating the depth logic estimation information is specifically as follows:
S401: identifying target features within each segment based on the intelligent segmentation strategy, generating segment feature analysis data by analyzing feature details of signals within segments, including variations in amplitude and frequency;
S402: based on the paragraph feature analysis data, carrying out logic analysis on signals in the paragraphs, determining potential fault types of the paragraphs by evaluating the matching degree of signal features and known fault types, and generating fault type pre-judging data;
S403: and based on the fault type pre-judging data, carrying out signal evaluation by combining the analysis result of the data paragraphs, refining and determining the fault type of each paragraph, and generating depth logic evaluation information.
The invention is improved in that on the basis of the depth logic evaluation information, a particle swarm optimization algorithm is utilized to adjust key parameters, an optimization test is carried out, analysis parameters are adjusted by simulating fault conditions, the fault diagnosis process is optimized, and the steps for generating parameter optimization measures are specifically as follows:
s501: based on the depth logic evaluation information, a fuzzy logic algorithm is adopted to analyze and identify parameter characteristics related to fault types, wherein the parameter characteristics comprise window size and sliding step length of signal processing, parameters are adjusted, a differential fault mode is matched, and adjusted parameter characteristics are generated;
S502: based on the adjusted parameter characteristics, simulating signal processing under the condition of differential faults, evaluating the influence of parameter adjustment on fault diagnosis accuracy, determining optimal parameter setting and generating simulated optimized parameter setting;
S503: based on the simulated optimization parameter setting, a particle swarm optimization algorithm is utilized to adjust a fault diagnosis flow, optimize signal characteristic matching logic, verify diagnosis accuracy after parameter adjustment, and generate parameter optimization measures;
The particle swarm optimization algorithm updates the formula according to the particle speed:
；
and a particle location update formula:
；
adjusting parameters of the signal characteristic matching logic to generate parameter optimization measures;
Wherein, Is a particleIn iterationIs used for the speed of the (c) in the (c),Is a particleIn iterationIs provided in the position of (a),Is a particleAt the iteration numberIs used for the speed of the (c) in the (c),Is a particleAt the iteration numberIs provided in the position of (a),Is a particleThe optimal position to be searched so far,For the current global optimum position,As the weight of the inertia is given,AndIn order for the acceleration constant to be high,AndIn the form of a random number,AndIn order to add the parameters of the device,Scaling factors for location update.
The invention improves, based on the parameter optimization measures, by simulating the differential fault situation, testing the parameter optimization effect, verifying the implementation effect of the optimization strategy in the application, the steps for generating the simulation test verification result are as follows:
s601: based on the parameter optimization measures, designing a fault scenario simulation test, covering faults of different types and degrees, verifying the adaptability of parameter adjustment, and generating a fault simulation test design;
S602: executing the fault simulation test design, collecting test data, including diagnosis time and fault identification accuracy, comparing performance differences before and after the test, and generating a fault simulation test record;
S603: based on the fault simulation test record, evaluating the performance of parameter adjustment under the differential fault situation, determining the application range of the optimization strategy, and generating a simulation test verification result.
The invention improves, according to the result of the simulation test verification, by analyzing the feedback of the fault test, refining the fault type, and summarizing, the step of generating the fault diagnosis log is specifically as follows:
s701: based on the simulation test verification result, collecting and arranging data in a fault simulation test, including fault type, diagnosis accuracy and processing time, and generating fault data collection information;
s702: analyzing each fault type based on the fault data collecting information, identifying commonality and difference in fault diagnosis, refining the fault type, and generating fault type analysis data;
s703: and based on the fault type analysis data, summarizing key points and improvement points in the fault diagnosis process, making logs, including diagnosis steps and optimization measures, and generating fault diagnosis logs.
A cable anomaly identification system based on time-frequency analysis, the system comprising:
the signal decomposition module detects the amplitude and the frequency of the signal based on the time-frequency data when the cable runs, identifies and marks key points of energy increase or frequency change, decomposes the signal according to the key change points, distributes the signal into a differential fuzzy set according to the energy and the frequency change, distributes corresponding grades for signal segments, and generates a signal fuzzy grade mapping table;
The feature identification module executes fuzzy value conversion based on the signal fuzzy level mapping table, matches with a predefined fault mode rule base, classifies signals by identifying key fault modes, and generates fault mode identification data;
The clustering algorithm module is used for dividing the data into a plurality of paragraphs with similar characteristics by automatically identifying energy peaks and frequency changes in the time-frequency data by utilizing a time sequence clustering algorithm based on the fault mode identification data, and generating an intelligent segmentation strategy;
The logic evaluation module carries out logic analysis and evaluation on each data paragraph based on the intelligent segmentation strategy, and generates depth logic evaluation information by analyzing characteristic details of signals in paragraphs, including amplitude and frequency changes, and refining fault type judgment;
the parameter optimization module is used for adjusting and optimizing the key parameters by adopting a particle swarm optimization algorithm based on the depth logic evaluation information, adjusting and analyzing the parameters by simulating fault conditions, optimizing the fault diagnosis process and generating parameter optimization measures;
the simulation test module performs fault simulation test and efficiency verification based on the parameter optimization measures, tests the parameter optimization effect by simulating the differential fault situation, verifies the implementation effect of the optimization strategy in application, refines the fault type according to the implementation result, and generates a fault diagnosis log.
Compared with the prior art, the invention has the advantages and positive effects that:
According to the invention, through the decomposition and mapping of the signals in the fuzzy set, the inherent fuzzy boundary and spectrum characteristics of the signals are accurately captured, so that the signal analysis is finer. The conversion of the fuzzy value and the matching of the rule base enhance the recognition capability of the fault mode and improve the fault diagnosis accuracy. And automatically identifying key change points in the time-frequency data by using a time sequence clustering algorithm, effectively dividing the data, further capturing abnormal signals sharply, and improving the identification sensitivity. The application of the particle swarm optimization algorithm in parameter adjustment optimizes the fault diagnosis flow and improves the diagnosis efficiency and accuracy. And the refinement and induction report of the fault type provides an improved basis for fault prevention and treatment strategies through analysis and summarization.
Drawings
Fig. 1 is a flowchart of a cable anomaly identification method based on time-frequency analysis;
fig. 2 is a schematic diagram of a refining flow of step S1 in a cable anomaly identification method based on time-frequency analysis;
fig. 3 is a schematic diagram of a refining flow of step S2 in the cable anomaly identification method based on time-frequency analysis according to the present invention;
fig. 4 is a schematic diagram of a step S3 refinement flow in a cable anomaly identification method based on time-frequency analysis according to the present invention;
fig. 5 is a schematic diagram of a refining flow of step S4 in the cable anomaly identification method based on time-frequency analysis according to the present invention;
fig. 6 is a schematic diagram of a refining flow of step S5 in the cable anomaly identification method based on time-frequency analysis according to the present invention;
fig. 7 is a schematic diagram of a refining flow of step S6 in a cable anomaly identification method based on time-frequency analysis;
Fig. 8 is a schematic diagram of a refining flow of step S7 in a cable anomaly identification method based on time-frequency analysis according to the present invention;
Fig. 9 is a block diagram of a cable anomaly identification system based on time-frequency analysis according to the present invention.
Detailed Description
The present invention will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present invention more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
In the description of the present invention, it should be understood that the terms "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," etc. indicate or are based on the orientation or positional relationship shown in the drawings, merely to facilitate description of the invention and simplify description, and do not indicate or imply that the devices or elements referred to must have a specific orientation, be configured and operated in a specific orientation, and thus should not be construed as limiting the invention. Furthermore, in the description of the present invention, the meaning of "a plurality" is two or more, unless explicitly defined otherwise.
Examples: referring to fig. 1, the present invention provides a technical solution: a cable anomaly identification method based on time-frequency analysis comprises the following steps:
s1: on the basis of time-frequency data during the operation of the cable, carrying out signal decomposition and fuzzy set construction, and matching boundary conditions of fuzzy logic control to generate fuzzy characteristic identification data;
s2: based on the fuzzy characteristic identification data, performing fuzzy value conversion and rule base matching, and evaluating and classifying signal characteristics to generate a fuzzy data processing record;
s3: based on fuzzy data processing records, applying a time sequence clustering algorithm, dividing the data into multiple sections with similar characteristics by automatically identifying key change points in the time frequency data, and generating an intelligent segmentation strategy;
S4: based on an intelligent segmentation strategy, through analyzing target features in each paragraph, signal re-evaluation is carried out, fault type judgment is refined, and depth logic evaluation information is generated;
s5: based on the depth logic evaluation information, performing key parameter adjustment by using a particle swarm optimization algorithm, performing optimization test, adjusting analysis parameters by simulating fault conditions, optimizing a fault diagnosis process, and generating parameter optimization measures;
s6: based on the parameter optimization measures, testing the parameter optimization effect by simulating the differential fault situation, verifying the implementation effect of the optimization strategy in the application, and generating a simulation test verification result;
S7: and according to the simulation test verification result, refining the fault type by analyzing the feedback of the fault test, and carrying out induction summary to generate a fault diagnosis log.
The fuzzy feature identification data comprises a signal amplitude level, a frequency change interval and fuzzy boundary division, the fuzzy data processing record comprises a fault mode matching degree, a signal classification basis and a fault prediction accuracy, the intelligent segmentation strategy comprises a segmentation starting point, a dominant frequency feature and an energy concentration area, the depth logic evaluation information comprises a subdivided fault type, a feature matching degree score and a signal re-evaluation result, the parameter optimization measure comprises a window size adjustment, a step length optimization and a signal analysis accuracy improvement target, the simulation test verification result comprises a fault recognition rate improvement point, an analysis time shortening point and an optimized parameter efficiency comparison, and the fault diagnosis log comprises a diagnosis key step record and a parameter adjustment history.
Referring to fig. 2, based on time-frequency data during cable operation, signal decomposition and fuzzy set construction are performed, boundary conditions of fuzzy logic control are matched, and the step of generating fuzzy feature identification data specifically includes:
S101: based on time-frequency data of the cable in operation, scanning the whole data set, detecting signal amplitude and frequency, identifying energy increase or frequency change points, marking the energy increase or frequency change points as key change points in the signal, and generating a key change point mark set;
S102: based on the key change point mark set, decomposing the signals, distributing the signals to the differential fuzzy set according to the change of energy and frequency, distributing corresponding fuzzy grades for each signal segment, and generating a signal fuzzy grade mapping table;
s103: and evaluating each fuzzy grade based on the signal fuzzy grade mapping table, determining the matching degree with the fuzzy logic control boundary, determining the signal fuzzy classification and generating fuzzy characteristic identification data.
In S101, based on time-frequency data of a cable in operation, a data set is scanned by adopting a Fourier transform algorithm, fourier transform of signals is carried out by utilizing the fft function in NumPy library, energy increase or frequency change points are identified by calculating the amplitude values of the signals under different frequencies, and the points are marked as key change points in the signals by utilizing a mark function mark_key_changes, so that a key change point mark set is generated.
S102, based on a key change point mark set, performing signal decomposition processing by adopting a fuzzy C-means clustering algorithm, setting the number of clustering centers as the number of preset fuzzy sets by utilizing a cmeans function in a scikit-fuzzy library, distributing signals to different fuzzy sets according to the change of the signals in energy and frequency, distributing corresponding fuzzy grades for each signal segment, and generating a signal fuzzy grade mapping table by utilizing an assignment membership function.
In S103, based on a signal fuzzy level mapping table, fuzzy level assessment is performed by adopting a fuzzy logic controller, the fuzzy level of the signal is set by utilizing FuzzyLogicToolbox in MATLAB, the input variable is set as the fuzzy level of the signal, the output variable is the matching degree of the fuzzy logic control boundary, and fuzzy classification of the signal is determined by defining a group of fuzzy rules, such as 'if the fuzzy level of the signal is high, the matching degree is high', so as to generate fuzzy feature identification data.
Referring to fig. 3, based on the fuzzy feature recognition data, performing fuzzy value conversion and rule base matching, and performing evaluation and classification of signal characteristics, the steps of generating a fuzzy data processing record specifically include:
S201: based on the fuzzy characteristic identification data, carrying out numerical conversion on each fuzzy classified signal, converting the fuzzy grade into a processable fuzzy value, and generating a fuzzy value conversion set;
s202: based on the fuzzy value conversion set, executing a fault mode matching process, comparing each converted fuzzy value with a predefined fault mode rule base, identifying a key fault mode, and generating fault mode identification data;
s203: based on the fault mode identification data, classifying signals according to the matched fault modes, determining the category of each signal, and generating a fuzzy data processing record.
In S201, based on the Fuzzy feature identification data, a Fuzzy logic reasoning algorithm is adopted to conduct numerical conversion on each Fuzzy classified signal, a Scikit-Fuzzy library in Python is utilized to execute a fuzzification function fuzzify on Fuzzy grades, the Fuzzy grades are converted into processable Fuzzy values, and a Fuzzy value conversion set is generated by defining a range of the Fuzzy values and a membership function.
And S202, based on the fuzzy value conversion set, executing a fault mode matching process, adopting a rule-based reasoning system, defining a set of predefined fault mode rule base by utilizing experta base in Python, comparing each converted fuzzy value with the fault mode in the rule base, executing the matching process by using a match_rules function, identifying a key fault mode, and generating fault mode identification data.
And S203, classifying signals according to the matched fault modes by adopting a decision tree classification algorithm, setting the depth of a decision tree to be a preset value by utilizing a DecisionTreeClassifier function in a scikit-learn library, performing fit function training according to the matching result of the fault modes, determining the category of each signal by using a predict function, and generating a fuzzy data processing record.
Referring to fig. 4, based on the fuzzy data processing record, a time sequence clustering algorithm is applied, and the data is divided into multiple segments with similar characteristics by automatically identifying key change points in the time-frequency data, so that the steps of generating an intelligent segmentation strategy are specifically as follows:
S301: based on the fuzzy data processing record, analyzing key features in the time-frequency data by utilizing wavelet transformation, including energy peak value and frequency change, identifying a change interval of the data, and generating key feature interval data;
S302: based on the key feature interval data, a time sequence clustering algorithm is applied, the time frequency data is divided into a plurality of paragraphs according to the change features, each paragraph comprises data of the same kind of energy and frequency features, and segmentation mapping data is generated;
S303: based on the segment map data, feature map logic is applied to each data segment to distinguish signal characteristics of the difference types and refine classification of the data segments to generate an intelligent segmentation strategy.
In S301, key features in the time-frequency data are analyzed by discrete wavelet transformation based on fuzzy data processing records, the key features are realized by PYWAVELETS libraries, a wavelet function is set as db1 of Daubechies series, a decomposition level is selected as 5, time-frequency data are analyzed to identify energy peaks and frequency changes, a thresholding method is used for carrying out threshold processing on the wavelet coefficients, the key features are highlighted, the change interval of the data is identified, and key feature interval data are generated.
In S302, based on key feature interval data, a time sequence clustering algorithm is applied to segment the time-frequency data, a TIMESERIESKMEANS function in a tslearn library is adopted, the clustering number is set to be the preset number of segments, a measurement standard adopts DTW (dynamic time warping) to consider the elastic matching of time-sequence data, the time-frequency data is divided into a plurality of segments according to change features, and each segment comprises data of the same type of energy and frequency features, so that segment mapping data are generated.
A time sequence clustering algorithm, according to the formula:
；
Calculating similarity between time-frequency data Generating segment mapping data;
Wherein, AndRespectively representing two time-series data sequences,AndRespectively represent sequencesAndIs provided for the length of (a),Representing sequencesMiddle (f)Individual points and sequencesMiddle (f)The distance between the points of interest is such that,The relative importance of each point in the sequence is represented as a weight matrix,Is thatCorresponding to (a)AndIs used for the weight value of (a),AndFor adjusting parameters, respectively for adjustingAndThe impact weight of a particular feature point in the sequence,Is a sequenceMiddle (f)The additional feature values of the individual points are,Is a sequenceMiddle (f)Additional feature values for the individual points.
The execution process comprises the following steps:
determining a time series data sequence AndAnd its lengthAndSetting a weight matrixDetermining weights for data points by data analysis or optimization algorithmsSetting adjustment parametersAndThe influence of the additional features in the DTW calculation is balanced through the determination of cross-validation or parameter optimization technology, the DTW distance is calculated, and the weight of each data point and the additional features are considered to improve the clustering accuracy and efficiency.
In S303, based on the segment mapping data, applying feature mapping logic to each data segment to further analyze, extracting features of each data segment by using a Pandas library of Python, identifying and marking different signal characteristics such as peak frequency and energy level, refining and classifying the data segments according to the extracted features by adopting LogisticRegression functions in a scikit-learn library through a logistic regression algorithm, and generating an intelligent segmentation strategy.
Referring to fig. 5, based on the intelligent segmentation strategy, by analyzing the target features in each segment, signal re-evaluation is performed, fault type judgment is refined, and the step of generating depth logic evaluation information specifically includes:
S401: identifying target features within each paragraph based on the intelligent segmentation strategy, generating paragraph feature analysis data by analyzing feature details of signals within the paragraphs, including variations in amplitude and frequency;
S402: based on paragraph feature analysis data, carrying out logic analysis on signals in the paragraphs, determining potential fault types of the paragraphs by evaluating the matching degree of signal features and known fault types, and generating fault type prejudging data;
S403: based on the fault type pre-judging data, carrying out signal evaluation by combining the analysis result of the data paragraphs, refining and determining the fault type of each paragraph, and generating depth logic evaluation information.
In S401, based on an intelligent segmentation strategy, identifying target features in each paragraph by adopting a signal processing technology, carrying out numerical analysis on signals by utilizing a NumPy library of Python, analyzing feature details of the signals in the paragraphs, including amplitude and frequency changes, and generating paragraph feature analysis data by calculating statistics such as average value, standard deviation, spectral density and the like of the signals.
In S402, based on paragraph feature analysis data, a Fuzzy logic analysis method is adopted to carry out logic analysis on signals in a paragraph, a Fuzzy reasoning system is built by utilizing a Scikit-Fuzzy library of Python, an input variable is defined as amplitude variation and frequency variation of the signals, an output variable is defined as the matching degree of the signals and known fault types, and by setting a Fuzzy rule, for example, if the amplitude variation is large and the frequency variation is small, the fault type may be a type A, the potential fault type of the paragraph is determined, and fault type pre-judging data is generated.
In S403, based on the fault type pre-judging data, combining the analysis result of the data paragraphs, adopting a decision tree algorithm to carry out comprehensive evaluation on signals, using DecisionTreeClassifier functions in scikit-learn libraries of Python, taking the fault type pre-judging data as input, training a model to refine and determine the fault type of each paragraph, setting the depth of the decision tree as a preset value to control the complexity of the model, and generating depth logic evaluation information.
Referring to fig. 6, based on the depth logic evaluation information, the key parameters are adjusted by using a particle swarm optimization algorithm, and an optimization test is performed, so that the fault diagnosis process is optimized by simulating the fault condition to adjust analysis parameters, and the steps for generating the parameter optimization measures are specifically as follows:
S501: based on the depth logic evaluation information, a fuzzy logic algorithm is adopted to analyze and identify parameter characteristics related to fault types, wherein the parameter characteristics comprise window size and sliding step length of signal processing, parameters are adjusted, a differential fault mode is matched, and adjusted parameter characteristics are generated;
S502: based on the adjusted parameter characteristics, simulating signal processing under the condition of differential faults, evaluating the influence of parameter adjustment on fault diagnosis accuracy, determining optimal parameter setting, and generating simulated optimized parameter setting;
s503: based on the simulated optimization parameter setting, a particle swarm optimization algorithm is utilized to adjust the fault diagnosis flow, optimize signal characteristic matching logic, verify the diagnosis accuracy after parameter adjustment, and generate parameter optimization measures.
In S501, based on depth logic evaluation information, a Fuzzy logic algorithm is adopted to analyze and identify parameter characteristics related to fault types, wherein the parameter characteristics comprise window size and sliding step length of signal processing, a Fuzzy reasoning system is constructed by utilizing a Scikit-Fuzzy library of Python, input variables are defined as fault types, output variables are signal processing parameters, and the window size is increased by setting a Fuzzy rule, for example, if the fault types are of type A, parameters are adjusted to match with differential fault modes, so that adjusted parameter characteristics are generated.
And S502, based on the adjusted parameter characteristics, simulating signal processing under the condition of differential faults, adopting a signal reconstruction technology to evaluate the influence of parameter adjustment on fault diagnosis accuracy, utilizing NumPy libraries in Python to perform signal reconstruction simulation, executing a reconstruction process by changing the window size and the sliding step length, evaluating fault diagnosis effects under different parameter settings, determining optimal parameter settings, and generating simulation optimization parameter settings.
In S503, based on the setting of simulation optimization parameters, the fault diagnosis flow is adjusted and optimized by using a particle swarm optimization algorithm, particle swarm optimization is executed by using a PySwarms library of Python, the number of particles is set to be 20, the iteration number is 100, the optimization target is to maximize the accuracy of fault diagnosis, optimization test is performed by adjusting parameters of signal feature matching logic, such as window size and sliding step length, the diagnosis accuracy after parameter adjustment is verified, and a parameter optimization measure is generated.
Particle swarm optimization algorithm, according to the particle velocity updating formula:
；
and a particle location update formula:
；
adjusting parameters of the signal characteristic matching logic to generate parameter optimization measures;
Wherein, Is a particleIn iterationIs used for the speed of the (c) in the (c),Is a particleIn iterationIs provided in the position of (a),Is a particleAt the iteration numberIs used for the speed of the (c) in the (c),Is a particleAt the iteration numberIs provided in the position of (a),Is a particleThe optimal position to be searched so far,For the current global optimum position,In order to improve the inertia weight after the modification,AndIn order for the acceleration constant to be high,AndIs a random number, again ranging from 0,1,AndTo adjust the influence of individual optima and global optima for the parameters introduced,A scaling factor for the location update, for controlling the step size of the location update.
The execution process comprises the following steps:
randomly setting the position of a group of particles Sum speed ofEach particle represents a solution, calculate fitness of each particle, updateAndApplying a velocity update formula and a location update formula to each particle, wherein parameters are introducedAndThe influence of individuals and global optima is adjusted,、AndIn order to adjust the weight of the object,AndIs a random number in the range of 0,1,Controlling the position updating step length until the maximum iteration number or other termination conditions are met, and after the completion,Representing the optimal parameter settings, i.e. the solution that maximizes the fault diagnosis accuracy.
Referring to fig. 7, based on the parameter optimization measures, by simulating the differential fault scenario, testing the parameter optimization effect, verifying the implementation effect of the optimization strategy in the application, the steps of generating the simulation test verification result are specifically as follows:
S601: based on the parameter optimization measures, designing a fault scenario simulation test, covering faults of different types and degrees, verifying the adaptability of parameter adjustment, and generating a fault simulation test design;
S602: executing fault simulation test design, collecting test data, including diagnosis time and fault identification accuracy, comparing performance differences before and after the test, and generating a fault simulation test record;
S603: based on the fault simulation test record, evaluating the performance of parameter adjustment under the differential fault situation, determining the application range of the optimization strategy, and generating a simulation test verification result.
In S601, based on parameter optimization measures, fault scenario simulation tests are designed to cover faults of different types and degrees, fault injection technology is utilized to simulate faults in a signal processing system, python script is adopted to dynamically adjust signal characteristics, such as noise is artificially increased or signal amplitude is artificially changed, adaptability of parameter adjustment is verified, and fault simulation test designs are generated.
And S602, performing fault simulation test design, collecting test data, measuring diagnosis time by adopting a time library in Python, evaluating fault recognition accuracy by using an accuracy calculation formula, comparing performance differences before and after the test, and generating a fault simulation test record by recording time consumption before and after the diagnosis and change of the recognition accuracy.
In S603, based on the fault simulation test record, the performance of the evaluation parameter adjustment under the differential fault scenario is determined by using a statistical analysis method, such as calculating an average improvement rate, and the application range and effect of the optimization strategy are determined, and the simulation test verification result is generated by comprehensively analyzing the shortening of the diagnosis time and the improvement of the accuracy.
Referring to fig. 8, according to the simulation test verification result, by analyzing the feedback of the fault test, the fault type is refined, and the steps of generating the fault diagnosis log are specifically:
S701: based on the simulation test verification result, collecting and arranging data in the fault simulation test, including fault type, diagnosis accuracy and processing time, and generating fault data collection information;
s702: based on fault data collection information, analyzing each fault type, identifying commonality and difference in fault diagnosis, refining the fault types, and generating fault type analysis data;
s703: and (3) based on the fault type analysis data, summarizing key points and improvement points in the fault diagnosis process, making logs, including diagnosis steps and optimization measures, and generating fault diagnosis logs.
Based on the simulation test verification result, the data in the fault simulation test is collected and organized by adopting a data statistics analysis method, wherein the data comprises fault type, diagnosis accuracy and processing time, and the data is organized and analyzed by utilizing a Pandas library of Python to generate fault data collection information.
In S702, based on fault data collection information, each fault type is analyzed by adopting a clustering analysis method, KMeans functions in a Scikit-learn library of Python are used for clustering, commonalities and differences in fault diagnosis are identified, the fault types are refined, and fault type analysis data are generated.
And S703, based on the fault type analysis data, adopting a text analysis method to summarize key points and improved points in the fault diagnosis process, utilizing a NLTK library of Python to extract key words and analyze subjects, making logs, including diagnosis steps and optimization measures, and generating a fault diagnosis log.
Referring to fig. 9, a cable anomaly identification system based on time-frequency analysis, the system includes:
the signal decomposition module detects the amplitude and the frequency of the signal based on the time-frequency data when the cable runs, identifies and marks key points of energy increase or frequency change, decomposes the signal according to the key change points, distributes the signal into a differential fuzzy set according to the energy and the frequency change, distributes corresponding grades for signal segments, and generates a signal fuzzy grade mapping table;
the feature identification module performs fuzzy value conversion based on the signal fuzzy level mapping table, matches with a predefined fault mode rule base, classifies signals by identifying key fault modes, and generates fault mode identification data;
The clustering algorithm module is used for dividing the data into a plurality of paragraphs with similar characteristics by automatically identifying energy peaks and frequency changes in the time-frequency data by utilizing a time sequence clustering algorithm based on the fault mode identification data, and generating an intelligent segmentation strategy;
the logic evaluation module carries out logic analysis and evaluation on each data paragraph based on an intelligent segmentation strategy, and generates depth logic evaluation information by analyzing characteristic details of signals in the paragraphs, including amplitude and frequency changes, and refining fault type judgment;
The parameter optimization module is used for adjusting and optimizing the key parameters by adopting a particle swarm optimization algorithm based on the depth logic evaluation information, adjusting the analysis parameters by simulating the fault condition, optimizing the fault diagnosis process and generating parameter optimization measures;
the simulation test module performs fault simulation test and efficiency verification based on parameter optimization measures, tests parameter optimization effects by simulating differential fault situations, verifies implementation effects of optimization strategies in application, refines fault types according to implementation results, and generates fault diagnosis logs.
The present invention is not limited to the above embodiments, and any equivalent embodiments which can be changed or modified by the technical disclosure described above can be applied to other fields, but any simple modification, equivalent changes and modification made to the above embodiments according to the technical matter of the present invention will still fall within the scope of the technical disclosure.
Claims (8)
1. The cable anomaly identification method based on time-frequency analysis is characterized by comprising the following steps of:
on the basis of time-frequency data during the operation of the cable, carrying out signal decomposition and fuzzy set construction, and matching boundary conditions of fuzzy logic control to generate fuzzy characteristic identification data;
Based on the fuzzy characteristic identification data, performing fuzzy value conversion and rule base matching, and evaluating and classifying signal characteristics to generate a fuzzy data processing record;
Based on the fuzzy data processing record, applying a time sequence clustering algorithm, dividing the data into multiple sections with the same kind of characteristics by automatically identifying key change points in the time-frequency data, and generating an intelligent segmentation strategy;
based on the intelligent segmentation strategy, analyzing target features in each paragraph, carrying out signal re-evaluation, refining fault type judgment, and generating depth logic evaluation information;
Based on the depth logic evaluation information, performing key parameter adjustment by using a particle swarm optimization algorithm, performing optimization test, adjusting analysis parameters by simulating fault conditions, optimizing a fault diagnosis process, and generating parameter optimization measures;
based on the parameter optimization measures, testing the parameter optimization effect by simulating the differential fault situation, verifying the implementation effect of the optimization strategy in the application, and generating a simulation test verification result;
According to the simulation test verification result, the fault type is refined by analyzing the feedback of the fault test, and the fault diagnosis log is generated by summarizing;
Based on the fuzzy data processing record, a time sequence clustering algorithm is applied, and the data is divided into multiple sections with the same kind of characteristics by automatically identifying key change points in the time-frequency data, so that the intelligent segmentation strategy is generated by the following steps:
based on the fuzzy data processing record, analyzing key features in time-frequency data by utilizing wavelet transformation, wherein the key features comprise energy peaks and frequency changes, identifying a change interval of the data and generating key feature interval data;
based on the key characteristic interval data, a time sequence clustering algorithm is applied to divide the time frequency data into a plurality of paragraphs according to the change characteristics, and each paragraph comprises data of the same kind of energy and frequency characteristics to generate segmentation mapping data;
The time sequence clustering algorithm is as follows:
；
Wherein, AndRespectively representing two time-series data sequences,AndRespectively represent sequencesAndIs provided for the length of (a),Representing sequencesMiddle (f)Individual points and sequencesMiddle (f)The distance between the points of interest is such that,As a matrix of weights, the weight matrix,Is thatCorresponding to (a)AndIs used for the weight value of (a),AndIn order to adjust the parameters of the device,Is a sequenceMiddle (f)The additional feature values of the individual points are,Is a sequenceMiddle (f)Additional feature values for the individual points;
Based on the segmentation mapping data, applying feature mapping logic to each data segment, distinguishing signal characteristics of difference types, and refining classification of the data segments to generate an intelligent segmentation strategy;
Based on the depth logic evaluation information, a particle swarm optimization algorithm is utilized to adjust key parameters, an optimization test is carried out, analysis parameters are adjusted through simulating fault conditions, a fault diagnosis process is optimized, and the step of generating parameter optimization measures specifically comprises the following steps:
Based on the depth logic evaluation information, a fuzzy logic algorithm is adopted to analyze and identify parameter characteristics related to fault types, wherein the parameter characteristics comprise window size and sliding step length of signal processing, parameters are adjusted, a differential fault mode is matched, and adjusted parameter characteristics are generated;
Based on the adjusted parameter characteristics, simulating signal processing under the condition of differential faults, evaluating the influence of parameter adjustment on fault diagnosis accuracy, determining optimal parameter setting and generating simulated optimized parameter setting;
based on the simulated optimization parameter setting, a particle swarm optimization algorithm is utilized to adjust a fault diagnosis flow, optimize signal characteristic matching logic, verify diagnosis accuracy after parameter adjustment, and generate parameter optimization measures;
The particle swarm optimization algorithm updates the formula according to the particle speed:
；
and a particle location update formula:
；
adjusting parameters of the signal characteristic matching logic to generate parameter optimization measures;
Wherein, Is a particleIn iterationIs used for the speed of the (c) in the (c),Is a particleIn iterationIs provided in the position of (a),Is a particleAt the iteration numberIs used for the speed of the (c) in the (c),Is a particleAt the iteration numberIs provided in the position of (a),Is a particleThe optimal position to be searched so far,For the current global optimum position,As the weight of the inertia is given,AndIn order for the acceleration constant to be high,AndIn the form of a random number,AndIn order to add the parameters of the device,Scaling factors for location update.
2. The cable anomaly identification method based on time-frequency analysis according to claim 1, wherein the fuzzy feature identification data comprises signal amplitude level, frequency variation interval and fuzzy boundary division, the fuzzy data processing records comprise fault mode matching degree, signal classification basis and fault prediction accuracy, the intelligent segmentation strategy comprises segmentation starting points, dominant frequency features and energy concentration areas, the depth logic evaluation information comprises subdivided fault types, feature matching degree scores and signal re-evaluation results, the parameter optimization measures comprise window size adjustment, step optimization and signal analysis accuracy improvement targets, the simulation test verification results comprise fault identification rate improvement points, analysis time shortening points and optimized parameter efficiency comparison, and the fault diagnosis logs comprise diagnosis key step records and parameter adjustment history.
3. The method for identifying cable anomalies based on time-frequency analysis according to claim 1, wherein the steps of performing signal decomposition and fuzzy set construction based on time-frequency data during cable operation, and matching boundary conditions of fuzzy logic control to generate fuzzy feature identification data are specifically as follows:
based on time-frequency data of the cable in operation, scanning the whole data set, detecting signal amplitude and frequency, identifying energy increase or frequency change points, marking the energy increase or frequency change points as key change points in the signal, and generating a key change point mark set;
Based on the key change point mark set, decomposing the signals, distributing the signals to the differential fuzzy set according to the change of energy and frequency, distributing corresponding fuzzy grades for each signal segment, and generating a signal fuzzy grade mapping table;
And evaluating each fuzzy grade based on the signal fuzzy grade mapping table, determining the matching degree with a fuzzy logic control boundary, determining the signal fuzzy classification and generating fuzzy characteristic identification data.
4. The method for identifying cable anomalies based on time-frequency analysis according to claim 1, wherein the step of performing fuzzy value conversion and rule base matching based on the fuzzy feature identification data, performing evaluation and classification of signal characteristics, and generating a fuzzy data processing record is specifically as follows:
based on the fuzzy characteristic identification data, carrying out numerical conversion on each fuzzy classified signal, converting a fuzzy grade into a processable fuzzy value, and generating a fuzzy value conversion set;
based on the fuzzy value conversion set, executing a fault mode matching process, comparing each converted fuzzy value with a predefined fault mode rule base, identifying a key fault mode, and generating fault mode identification data;
And classifying the signals according to the matched fault modes based on the fault mode identification data, determining the category of each signal, and generating a fuzzy data processing record.
5. The cable anomaly identification method based on time-frequency analysis according to claim 1, wherein the step of generating depth logic evaluation information by analyzing target features in each section based on the intelligent segmentation strategy, performing signal re-evaluation, refining fault type judgment specifically comprises the steps of:
Identifying target features within each segment based on the intelligent segmentation strategy, generating segment feature analysis data by analyzing feature details of signals within segments, including variations in amplitude and frequency;
Based on the paragraph feature analysis data, carrying out logic analysis on signals in the paragraphs, determining potential fault types of the paragraphs by evaluating the matching degree of signal features and known fault types, and generating fault type pre-judging data;
And based on the fault type pre-judging data, carrying out signal evaluation by combining the analysis result of the data paragraphs, refining and determining the fault type of each paragraph, and generating depth logic evaluation information.
6. The method for identifying cable anomalies based on time-frequency analysis according to claim 1, wherein the step of verifying the implementation effect of the optimization strategy in the application by simulating a differential fault scenario, testing the parameter optimization effect, based on the parameter optimization measure, and generating a simulation test verification result is specifically as follows:
based on the parameter optimization measures, designing a fault scenario simulation test, covering faults of different types and degrees, verifying the adaptability of parameter adjustment, and generating a fault simulation test design;
executing the fault simulation test design, collecting test data, including diagnosis time and fault identification accuracy, comparing performance differences before and after the test, and generating a fault simulation test record;
Based on the fault simulation test record, evaluating the performance of parameter adjustment under the differential fault situation, determining the application range of the optimization strategy, and generating a simulation test verification result.
7. The method for identifying cable anomalies based on time-frequency analysis according to claim 1, wherein the step of generating a fault diagnosis log by analyzing feedback of a fault test, refining fault types, and summarizing is specifically as follows:
based on the simulation test verification result, collecting and arranging data in a fault simulation test, including fault type, diagnosis accuracy and processing time, and generating fault data collection information;
Analyzing each fault type based on the fault data collecting information, identifying commonality and difference in fault diagnosis, refining the fault type, and generating fault type analysis data;
And based on the fault type analysis data, summarizing key points and improvement points in the fault diagnosis process, making logs, including diagnosis steps and optimization measures, and generating fault diagnosis logs.
8. A cable anomaly identification system based on time-frequency analysis, characterized in that it is performed according to the cable anomaly identification method based on time-frequency analysis of any one of claims 1-7, said system comprising:
the signal decomposition module detects the amplitude and the frequency of the signal based on the time-frequency data when the cable runs, identifies and marks key points of energy increase or frequency change, decomposes the signal according to the key change points, distributes the signal into a differential fuzzy set according to the energy and the frequency change, distributes corresponding grades for signal segments, and generates a signal fuzzy grade mapping table;
The feature identification module executes fuzzy value conversion based on the signal fuzzy level mapping table, matches with a predefined fault mode rule base, classifies signals by identifying key fault modes, and generates fault mode identification data;
The clustering algorithm module is used for dividing the data into a plurality of paragraphs with similar characteristics by automatically identifying energy peaks and frequency changes in the time-frequency data by utilizing a time sequence clustering algorithm based on the fault mode identification data, and generating an intelligent segmentation strategy;
The logic evaluation module carries out logic analysis and evaluation on each data paragraph based on the intelligent segmentation strategy, and generates depth logic evaluation information by analyzing characteristic details of signals in paragraphs, including amplitude and frequency changes, and refining fault type judgment;
the parameter optimization module is used for adjusting and optimizing the key parameters by adopting a particle swarm optimization algorithm based on the depth logic evaluation information, adjusting and analyzing the parameters by simulating fault conditions, optimizing the fault diagnosis process and generating parameter optimization measures;
the simulation test module performs fault simulation test and efficiency verification based on the parameter optimization measures, tests the parameter optimization effect by simulating the differential fault situation, verifies the implementation effect of the optimization strategy in application, refines the fault type according to the implementation result, and generates a fault diagnosis log.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410459647.4A CN118068132B (en) | 2024-04-17 | 2024-04-17 | Cable anomaly identification method and system based on time-frequency analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410459647.4A CN118068132B (en) | 2024-04-17 | 2024-04-17 | Cable anomaly identification method and system based on time-frequency analysis |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118068132A CN118068132A (en) | 2024-05-24 |
| CN118068132B true CN118068132B (en) | 2024-06-28 |
Family
ID=91099434
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410459647.4A Active CN118068132B (en) | 2024-04-17 | 2024-04-17 | Cable anomaly identification method and system based on time-frequency analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118068132B (en) |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109324236A (en) * | 2018-09-29 | 2019-02-12 | 国网山西省电力公司太原供电公司 | A kind of assessment of fault method based on cable connector typical defect |
Family Cites Families (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070191697A1 (en) * | 2006-02-10 | 2007-08-16 | Lynn Lawrence A | System and method for SPO2 instability detection and quantification |
| KR100544366B1 (en) * | 2003-10-17 | 2006-01-23 | 곽희로 | system and method for detecting the failures in gas insulated switchgear using fuzzy inference |
| CN206659651U (en) * | 2016-11-28 | 2017-11-24 | 广东伊莱特电器有限公司 | A kind of IH electric cookers ceramic liner |
| CN107884706B (en) * | 2017-11-09 | 2020-04-07 | 合肥工业大学 | Analog circuit fault diagnosis method based on vector value regular kernel function approximation |
| CN109444728A (en) * | 2018-09-21 | 2019-03-08 | 国网河南省电力公司济源供电公司 | A kind of circuit breaker failure diagnostic method based on dynamic weighting Hybrid Clustering Algorithm |
| CN113805018A (en) * | 2021-09-15 | 2021-12-17 | 陕西省地方电力（集团）有限公司 | Intelligent identification method for partial discharge fault type of 10kV cable of power distribution network |
| CN116819251A (en) * | 2023-06-27 | 2023-09-29 | 国网上海市电力公司 | High-voltage switch cabinet fault diagnosis method based on photoelectric monitoring |
| CN117493961A (en) * | 2023-11-07 | 2024-02-02 | 国网上海市电力公司 | Partial discharge signal separation and identification method, device and storage medium |
-
2024
- 2024-04-17 CN CN202410459647.4A patent/CN118068132B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109324236A (en) * | 2018-09-29 | 2019-02-12 | 国网山西省电力公司太原供电公司 | A kind of assessment of fault method based on cable connector typical defect |
Non-Patent Citations (1)
| Title |
|---|
| 智能电网电缆节点异常故障自动检测技术;陈阵;自动化与仪器仪表;20190825;146-149 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118068132A (en) | 2024-05-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110567720B (en) | Method for diagnosing depth confrontation of fault of fan bearing under unbalanced small sample scene | |
| CN110263846B (en) | Fault diagnosis method based on fault data deep mining and learning | |
| CN111666169B (en) | Improved isolated forest algorithm and Gaussian distribution-based combined data anomaly detection method | |
| CN110070060B (en) | Fault diagnosis method for bearing equipment | |
| CN111275024B (en) | Fault diagnosis method for rolling bearing of variable-working-condition fan driven by limited data containing noise label | |
| CN104330721B (en) | IC Hardware Trojan detecting method and system | |
| Arbin et al. | Comparative analysis between k-means and k-medoids for statistical clustering | |
| CN105275833A (en) | CEEMD (Complementary Empirical Mode Decomposition)-STFT (Short-Time Fourier Transform) time-frequency information entropy and multi-SVM (Support Vector Machine) based fault diagnosis method for centrifugal pump | |
| CN101738998B (en) | System and method for monitoring industrial process based on local discriminatory analysis | |
| CN102589884A (en) | Method for diagnosing failure of airplane generator bearing based on GentleBoost | |
| CN114707571B (en) | Credit data anomaly detection method based on enhanced isolation forest | |
| CN107830996A (en) | A kind of vehicle rudder diagnosis method for system fault | |
| US11331024B2 (en) | Calibration method for critical point of mental fatigue based on self-organized criticality | |
| CN114429152A (en) | Rolling bearing fault diagnosis method based on dynamic index antagonism self-adaption | |
| CN114997218A (en) | Recognition and detection method for polygonal abrasion of wheels of railway vehicle | |
| CN114462475A (en) | Unsupervised machine abnormal sound detection method and unsupervised machine abnormal sound detection device based on single classification algorithm | |
| Song et al. | Intelligent diagnosis method for machinery by sequential auto-reorganization of histogram | |
| CN115273791A (en) | Industrial equipment fault detection method and device based on acoustic feature driving | |
| CN118068132B (en) | Cable anomaly identification method and system based on time-frequency analysis | |
| Kumbhare et al. | A novel tri-component scheme for classifying neuronal discharge patterns | |
| CN117407662A (en) | Sensor data processing method and system | |
| CN116720073A (en) | Abnormality detection extraction method and system based on classifier | |
| CN117014193A (en) | Unknown Web attack detection method based on behavior baseline | |
| CN110057588A (en) | The bearing incipient fault detection and diagnostic method and system merged based on singular value with characteristics of graph theory | |
| CN114903442A (en) | Gating cycle network-based electroencephalogram signal inter-effect connectivity detection method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant |