Breaker operating mechanism mechanical state fault diagnosis method

By extracting and evaluating features of the circuit breaker operating mechanism using multi-scale entropy and intuitive hierarchical analysis, the problem of fault diagnosis in existing technologies is solved, and accurate diagnosis and early warning of the mechanical state of the circuit breaker operating mechanism are realized.

CN118965010AActive Publication Date: 2024-11-15HUAIAN OF JIANGSU ELECTRIC POWER CO POWER SUPPLY

Patent Information

Application Number
CN202410870662.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-01
Publication Date
2024-11-15
Estimated Expiration
2044-07-01

AI Technical Summary

Technical Problem

Existing technology can only determine the motion state of the circuit breaker operating mechanism, but cannot perform fault diagnosis based on its operating state, resulting in the inability to provide corresponding early warnings.

Method used

Multi-scale entropy is used to extract features of the mechanical state of the circuit breaker operating mechanism, and the mechanical state is evaluated by intuitionistic hierarchical analysis to establish a health model. Feature parameters are extracted and diagnosed by real-time data acquisition from sensors.

Benefits of technology

It enables accurate diagnosis of mechanical condition faults in circuit breaker operating mechanisms, avoids redundancy in characteristic quantity decomposition, and improves the accuracy and reliability of fault early warning.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118965010A_ABST
    Figure CN118965010A_ABST
Patent Text Reader

Abstract

The invention discloses a mechanical state fault diagnosis method for a circuit breaker operating mechanism, and belongs to the technical field of circuit breakers. The mechanical state fault diagnosis method for the circuit breaker operating mechanism comprises the following steps: step 1, acquiring historical mechanical performance parameters of the circuit breaker operating mechanism; 2, establishing a health model of the circuit breaker operating mechanism; step 3, collecting operation data of the circuit breaker operating mechanism; 4, establishing a feature extraction algorithm according to the extracted data; 5, obtaining mechanical performance parameters of the circuit breaker operating mechanism; and step 6, diagnosing the fault of the circuit breaker operating mechanism. According to the method, the problem that the fault cannot be diagnosed according to the operating state of the circuit breaker operating mechanism in the prior art, so that corresponding early warning cannot be carried out according to the fault condition of the circuit breaker operating mechanism is solved, and the richness and criticality of the characteristic information amount extracted from the mechanical state characteristics of the circuit breaker operating mechanism are ensured; and the accuracy of fault diagnosis is ensured.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the technical field of circuit breakers, and in particular to a method for diagnosing mechanical state faults of circuit breaker operating mechanisms. Background Art

[0002] Circuit breakers are important components of power grids and have a great impact on regional power supply. If they fail to operate, it may cause power shortages or power outages in most areas, and more seriously, it may cause great losses to society and the economy.

[0003] A Chinese patent with publication number CN105930622B discloses a method for determining the motion state of a circuit breaker spring operating mechanism, comprising: 1. determining the three parameters of the opening and closing spring stiffness, free length and initial pressure length; 2. at any time of the circuit breaker closing operation, calculating the calculated force of the closing spring and the released operating work according to the closing spring stiffness, free length and initial pressure length; 3. calculating the calculated force and consumed work of the opening spring according to the opening spring stiffness, free length and initial pressure length; 4. using Fluent software to simulate and obtain the buffer resistance of the buffer, and calculating its calculated force and consumed work; 5. calculating the calculated motion resistance of the arc extinguishing chamber and the calculated work consumed; 6. obtaining the resultant force at the calculated point and the residual work of the system according to steps 2 to 5; 7. determining the motion state of the spring operating mechanism according to the calculation result of step 6. This patented mathematical model accurately and effectively describes the motion state of the spring operating mechanism during the circuit breaker closing operation, overcoming the limitations of previous methods such as low efficiency, low accuracy and inability to reflect the dynamic changes of the closing process, and providing a good theoretical basis for exploring the reliable operation of the spring operating mechanism.

[0004] In actual use, the above patent can only determine the movement state of the circuit breaker operating mechanism, but cannot diagnose the fault according to the running state of the circuit breaker operating mechanism, resulting in failure to issue corresponding early warning according to the fault condition of the circuit breaker operating mechanism; therefore, it does not meet the existing needs. In this regard, we propose a method for diagnosing mechanical state faults of circuit breaker operating mechanisms. Summary of the invention

[0005] The purpose of the present invention is to provide a method for diagnosing a mechanical state fault of a circuit breaker operating mechanism. The method uses multi-scale entropy to extract features of the mechanical state of the circuit breaker operating mechanism, while also avoiding the redundancy of feature decomposition when extracting features of the mechanical state of the circuit breaker operating mechanism. The method uses an intuitive hierarchical analysis method to evaluate the mechanical state of the circuit breaker operating mechanism, thereby ensuring the accuracy of fault diagnosis and solving the problems raised in the above-mentioned background technology.

[0006] To achieve the above object, the present invention provides the following technical solution: a method for diagnosing mechanical state faults of a circuit breaker operating mechanism, comprising the following steps:

[0007] Step 1: Obtain historical mechanical characteristics of the circuit breaker operating mechanism, analyze the obtained mechanical characteristics of the circuit breaker operating mechanism, and obtain historical mechanical performance parameters of the circuit breaker operating mechanism under normal working conditions;

[0008] Step 2: Establish a health model of the circuit breaker operating mechanism according to the historical mechanical performance parameters of the circuit breaker operating mechanism under normal working conditions;

[0009] Step 3: The operating data of the circuit breaker operating mechanism is collected in real time through sensors, and then the collected data is preprocessed, which includes cleaning, transformation and abnormality detection of the collected data;

[0010] Step 4: extracting characteristic parameters related to the mechanical properties of the circuit breaker operating mechanism from the processed circuit breaker operating mechanism operation data, and establishing a feature extraction algorithm based on the extracted characteristic parameters related to the mechanical properties of the circuit breaker operating mechanism;

[0011] Step 5: Quantitatively analyze the mechanical characteristics according to the established feature extraction algorithm, and obtain the real-time mechanical performance parameters of the circuit breaker operating mechanism through simulation;

[0012] Step six: Analyze the spectrum and envelope of the real-time mechanical performance parameters of the circuit breaker operating mechanism, extract the real-time characteristic parameters of the circuit breaker operating mechanism, input the extracted real-time characteristic parameters into the health model of the circuit breaker operating mechanism, calculate the health score of the circuit breaker operating mechanism, and diagnose the circuit breaker operating mechanism fault according to the health score of the circuit breaker operating mechanism.

[0013] Preferably, establishing a circuit breaker operating mechanism health model according to historical mechanical performance parameters of the circuit breaker operating mechanism under normal working conditions specifically includes:

[0014] The historical mechanical performance parameters of the circuit breaker operating mechanism under normal working conditions are obtained, and the model is trained using the historical mechanical performance parameters. During the training process, the weights and parameters of the model are adjusted using the optimization algorithm by minimizing the loss function. The historical mechanical performance parameters are the transmission ratio, transmission efficiency, and motion stability of the circuit breaker operating mechanism under normal working conditions. The historical mechanical performance parameters are divided into a training set and a validation set, and the validation set is used to evaluate the performance indicators of the model.

[0015] According to the loss function on the training set, adjust the weights and parameters of the model so that the model gradually converges and fits the historical data;

[0016] During the training process, the performance indicators on the validation set are used to evaluate the effect of the model, and the hyperparameters and parameters of the model are adjusted based on the evaluation results;

[0017] Repeat parameter adjustment and verification until the optimal parameter setting is found and the circuit breaker operating mechanism health model is obtained.

[0018] Preferably, the repeated parameter adjustment and verification until the best parameter setting is found also includes adjusting the hyperparameters and parameters of the model according to the performance of the verification set, and finding the best parameter setting by trying different parameter combinations and using methods such as grid search or random search.

[0019] Preferably, the analyzing the spectrum and envelope of the real-time mechanical performance parameters of the circuit breaker operating mechanism to extract the real-time characteristic parameters of the circuit breaker operating mechanism specifically includes:

[0020] Obtain the historical types of mechanical faults of the circuit breaker operating mechanism, and obtain the characteristic parameters of the multi-dimensional sensing signal of the operating mechanism under the fault;

[0021] A characteristic parameter database corresponding to different fault types is established based on the characteristic parameters of the dimensional sensing signal, and the real-time mechanical performance parameters of the circuit breaker operating mechanism collected in real time are compared with the data in the characteristic parameter database corresponding to different fault types;

[0022] The data matching the data in the characteristic parameter database corresponding to different fault types are extracted to obtain the characteristic parameters of the circuit breaker operating mechanism.

[0023] Preferably, the step of obtaining characteristic parameters of multi-dimensional sensing signals of the operating mechanism under fault conditions specifically includes:

[0024] Extract the energy storage current data, and judge whether the energy storage motor of the circuit breaker operating mechanism is idling according to the change of the energy storage current;

[0025] Extract the frequency spectrum amplitude of the sound and vibration signal of the circuit breaker operating mechanism, the frequency spectrum waveform similarity or the speed in the giant magnetoresistance sensor monitoring signal, and judge whether the circuit breaker operating mechanism is not in place according to the frequency spectrum amplitude of the sound and vibration signal of the circuit breaker operating mechanism, the frequency spectrum waveform similarity or the speed change curve in the giant magnetoresistance sensor monitoring signal;

[0026] Extract the speed change curve or the duration of the spectrum waveform of the sound and vibration signal in the giant magnetoresistance sensor monitoring signal, and judge whether the opening and closing speed of the circuit breaker operating mechanism is too low according to the speed change curve or the duration of the spectrum waveform of the sound and vibration signal in the giant magnetoresistance sensor monitoring signal;

[0027] Extract the speed change curve or the spectrum waveform data of the sound and vibration signals in the giant magnetoresistance sensor monitoring signal, and judge whether the circuit breaker transmission mechanism of the circuit breaker operating mechanism is stuck, the circuit breaker spring energy storage is insufficient, or the circuit breaker operating mechanism is abnormally mechanically worn according to the speed change curve or the spectrum waveform similarity of the sound and vibration signals in the giant magnetoresistance sensor monitoring signal;

[0028] According to the results of the discrimination, the characteristic parameters of the multi-dimensional sensing signal of the operating mechanism under fault are obtained.

[0029] Preferably, establishing a feature extraction algorithm based on the extracted data specifically includes:

[0030] The frequency domain features of the time series in the operation data of the circuit breaker operating mechanism are extracted, and six types of information entropy, including energy entropy, singular entropy, approximate entropy, sample entropy and fuzzy entropy of the vibration signal, are obtained as characteristic entropy.

[0031] The global features in the time domain and the frequency domain in the operation data of the circuit breaker operating mechanism are extracted to form a feature matrix, and the global features of the vibration signal are selected as the skewness coefficient, the kurtosis coefficient, and the peak coefficient.

[0032] Use feature entropy dimensionality reduction to perform normalization processing, and subtract the average value of the corresponding dimension from each feature of the feature matrix;

[0033] Calculate the covariance matrix, where the diagonal and off-diagonal elements represent the autocovariance coefficients and cross-covariance coefficients, respectively;

[0034] Calculate the eigenvectors and eigenvalues ​​of the covariance matrix and sort them from large to small, and use the eigenvectors corresponding to the first n largest eigenvalues ​​as the new orthogonal eigenmatrix values.

[0035] Preferably, the sensor specifically includes:

[0036] Hall sensor, used to extract mechanical state information of the circuit breaker operating mechanism and changes in load characteristics;

[0037] Voiceprint and vibration sensors are used to perform spectrum and envelope analysis on the sound and vibration signals of the circuit breaker operating mechanism;

[0038] Image sensor, used to paste marks on the moving parts of the circuit breaker, use a high-speed camera to record the displacement of the operating rod of the circuit breaker operating mechanism during the breaking process, and obtain its stroke and time characteristics;

[0039] The giant magnetoresistance sensor is used to detect changes in angle and acceleration by sensing changes in the magnetic field based on the movement of the transmission shaft of the circuit breaker operating mechanism during opening and closing.

[0040] Preferably, the workflow of the voiceprint and vibration sensor specifically includes:

[0041] Perform feature analysis on the vibration signal of the circuit breaker operating mechanism, compare the time domain waveforms of different fault types, and analyze the spectrum of the circuit breaker operating mechanism signal after filtering;

[0042] Determine whether the signal spectrum of the circuit breaker operating mechanism meets the fault characteristics. If yes, determine the fault type according to the fault characteristics.

[0043] The features of all fault types are combined to generate a feature matrix, and the feature matrix is ​​compressed using principal component reduction (PCA).

[0044] Diagnose circuit breaker operating mechanism faults based on the compressed special diagnosis matrix.

[0045] Preferably, it also includes:

[0046] Extract the motion profile of the circuit breaker operating mechanism based on the Hall sensor;

[0047] Optimizing the action profile based on the optimization model;

[0048]

[0049] Where W(k) is the discrete signal of the motion profile before optimization processing; W N (k) is the optimized signal of the motion profile obtained after N iterations in the optimization model; S is the motion profile of the circuit breaker operating mechanism before optimization; M is the number of pixel points on the motion profile of the circuit breaker operating mechanism; T N-1 (k+i) is the weight at the i-th pixel at the N-1th iteration in the optimization model, T N-1 (k+i)>0;W N-1 (k+i) is the optimization signal at the i-th pixel at the N-1th iteration in the optimization model;

[0050] Selecting any pixel point other than the central pixel point on the optimized motion contour as a segmentation pixel point, segmenting the motion contour into a first sub-motion contour and a second sub-motion contour based on the segmentation pixel point, and determining the center points of the first sub-motion contour and the second sub-motion contour as the first feature point and the second feature point respectively;

[0051] Obtaining a first posture angle corresponding to the first sub-motion profile according to the first feature point and the segmented pixel point;

[0052]

[0053] Among them, θ a(j) is the first posture angle corresponding to the first sub-motion profile; y(j) is the ordinate of the segmented pixel point j; x(j) is the abscissa of the segmented pixel point j; y a (j) is the ordinate of the first feature point; x a (j) is the horizontal coordinate of the first feature point;

[0054] Obtaining a second posture angle corresponding to the second sub-motion profile according to the second feature point and the segmented pixel point;

[0055]

[0056] Among them, θ b (j) is the second posture angle corresponding to the second sub-motion profile; y b (j) is the ordinate of the second feature point; x b (j) is the horizontal coordinate of the second feature point;

[0057] The preset posture data table is queried according to the first posture angle corresponding to the first sub-motion profile and the second posture angle corresponding to the second sub-motion profile to determine the motion posture information of the circuit breaker operating mechanism as the mechanical state information of the circuit breaker operating mechanism.

[0058] Preferably, the step of inputting the extracted real-time characteristic parameters into the health model of the circuit breaker operating mechanism to calculate the health score of the circuit breaker operating mechanism specifically includes:

[0059] When the health status score is 100, the assessment level is no maintenance;

[0060] When the health status score is 90≤X<100, the assessment level is D-level maintenance;

[0061] When the health status score is 80≤X<90, the assessment level is C-level maintenance;

[0062] When the health status score is 60≤X<80, the assessment level is B-level maintenance;

[0063] When the health status score is 0≤X<60, the assessment level is A-level maintenance.

[0064] Compared with the prior art, the present invention has the following beneficial effects:

[0065] The present invention uses multi-scale entropy to extract features of the mechanical state of the circuit breaker operating mechanism, thereby ensuring the richness and criticality of the feature information of the mechanical state feature extraction of the circuit breaker operating mechanism, while also avoiding the redundancy of feature quantity decomposition when extracting features of the mechanical state of the circuit breaker operating mechanism. The intuitive hierarchical analysis method is used to evaluate the mechanical state of the circuit breaker operating mechanism, thereby ensuring the accuracy of fault diagnosis. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] Figure 1 It is a schematic diagram of a method for diagnosing mechanical state faults of a circuit breaker operating mechanism according to the present invention;

[0067] Figure 2 A flow chart of characteristic parameter acquisition of a method for diagnosing mechanical state faults of a circuit breaker operating mechanism of the present invention;

[0068] Figure 3 It is a fault diagnosis schematic diagram of the circuit breaker operating mechanism mechanical state fault diagnosis method of the present invention. DETAILED DESCRIPTION

[0069] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0070] In order to solve the problem that the prior art can only judge the motion state of the circuit breaker operating mechanism in actual use, but cannot diagnose the fault according to the running state of the circuit breaker operating mechanism, resulting in failure to issue corresponding warnings according to the fault conditions of the circuit breaker operating mechanism, please refer to Figure 1-Figure 3 , this embodiment provides the following technical solutions:

[0071] A method for diagnosing mechanical state faults of a circuit breaker operating mechanism comprises the following steps:

[0072] Step 1: Obtain the historical mechanical characteristics of the circuit breaker operating mechanism, analyze the mechanical characteristics of the circuit breaker operating mechanism, such as transmission ratio, transmission efficiency, motion stability, etc., and obtain the historical mechanical performance parameters of the circuit breaker operating mechanism under normal working conditions.

[0073] Step 2: Establish a health model of the circuit breaker operating mechanism based on the historical mechanical performance parameters of the circuit breaker operating mechanism under normal working conditions.

[0074] Step 3: The operating data of the circuit breaker operating mechanism is collected in real time through sensors, and then the collected data is preprocessed. The preprocessing includes filtering, denoising, signal analysis, etc. The preprocessing includes cleaning, transformation and anomaly detection processing of the collected data, accurately obtaining fault-related information, selecting a suitable model, and using historical data to train and optimize it to improve prediction accuracy and stability. The operating data of the circuit breaker operating mechanism is collected in real time through sensors, and then the collected data is preprocessed. Characteristic parameters related to mechanical properties are extracted from the preprocessed data, such as frequency domain characteristics, time domain characteristics, energy characteristics, etc. of sensor signals, so as to accurately describe the mechanical state of the circuit breaker operating mechanism.

[0075] Step 4: Extract characteristic parameters related to the mechanical properties of the circuit breaker operating mechanism from the processed operation data of the circuit breaker operating mechanism, such as the frequency domain characteristics, time domain characteristics, energy characteristics, etc. of the sensor signal, and establish a feature extraction algorithm based on the extracted characteristic parameters related to the mechanical properties of the circuit breaker operating mechanism to accurately describe the mechanical state of the circuit breaker operating mechanism.

[0076] Step 5: Quantitatively analyze the mechanical characteristics, such as transmission ratio, transmission efficiency, motion stability, etc., according to the established feature extraction algorithm. Through simulation, the real-time mechanical performance parameters of the circuit breaker operating mechanism are obtained. By analyzing the mechanical characteristics of the circuit breaker operating mechanism, such as transmission ratio, transmission efficiency, motion stability, etc., and performing simulation, the real-time mechanical performance parameters of the circuit breaker operating mechanism are obtained. The common failure modes and failure characteristics of the circuit breaker operating mechanism can be understood, such as transmission part breakage, transmission part looseness, transmission mechanism jamming, etc. By collecting and summarizing existing failure cases and experiences, a fault diagnosis knowledge base is formed to provide a basis for fault diagnosis of the mechanical state of the circuit breaker mechanism.

[0077] Step six: Analyze the spectrum and envelope of the real-time mechanical performance parameters of the circuit breaker operating mechanism, extract the real-time characteristic parameters of the circuit breaker operating mechanism, the characteristic parameters include vibration frequency, amplitude, kurtosis, etc., input the extracted real-time characteristic parameters into the health model of the circuit breaker operating mechanism, calculate the health score of the circuit breaker operating mechanism, diagnose the circuit breaker operating mechanism fault according to the health score of the circuit breaker operating mechanism, extract characteristic parameters that can reflect the health status and fault characteristics of the circuit breaker operating mechanism, analyze and process multi-dimensional perception data, establish a mechanism health status evaluation model, judge the current health status of the mechanism, model and analyze known fault modes, and accurately diagnose the mechanism fault according to the matching results of characteristic parameters and models.

[0078] Diagnose faults based on health scores. Figure 3, where T is time, N is state, in the initial stage, the mechanical state of the circuit breaker operating mechanism is relatively stable, and the curve is a horizontal line. After a period of time, the state begins to decline at point A, and it can be detected when it declines significantly at point P. It develops into a functional fault at point F, and the state before point F (functional fault point) is the potential fault state), point S is a power grid fault, T is the interval from potential fault to functional fault, Ts is the interval from functional fault to power grid fault, and Tc is the interval for maintenance and inspection. It can be seen from the figure that only when Tc is less than T can the potential fault be detected before the functional fault occurs, and only when Tc is less than Ts can the functional fault be detected before the power grid fault occurs.

[0079] The health model of the circuit breaker operating mechanism is established based on the historical mechanical performance parameters of the circuit breaker operating mechanism under normal working conditions, including:

[0080] The historical mechanical performance parameters of the circuit breaker operating mechanism under normal working conditions are obtained, the model is trained through the historical mechanical performance parameters, and the historical mechanical performance parameters are divided into a training set and a validation set. The historical mechanical performance parameters are the transmission ratio, transmission efficiency, and motion stability of the circuit breaker operating mechanism under normal working conditions.

[0081] According to the loss function on the training set, the weights and parameters of the model are adjusted so that the model gradually converges and fits the historical data.

[0082] During the training process, the effect of the model is evaluated by the performance indicators on the validation set, and the hyperparameters and parameters of the model are adjusted according to the evaluation results.

[0083] Repeat parameter adjustment and verification until the optimal parameter setting is found and the circuit breaker operating mechanism health model is obtained to improve the prediction accuracy and stability of the model.

[0084] Model training using historical mechanical performance parameters also includes: during the training process, by minimizing the loss function, using the optimization algorithm to adjust the model's weights and parameters, and the validation set is used to evaluate the model's performance indicators, such as accuracy, precision, recall, F1 value, etc.

[0085] Repeat parameter adjustment and verification until the best parameter setting is found. This also includes adjusting the model's hyperparameters and parameters based on the performance of the validation set. By experimenting with different parameter combinations and using grid search or random search methods to find the best parameter setting, cross-validation can further verify the model's stability and generalization ability.

[0086] The spectrum and envelope of the real-time mechanical performance parameters of the circuit breaker operating mechanism are analyzed to extract the real-time characteristic parameters of the circuit breaker operating mechanism, including:

[0087] Obtain the historical types of mechanical faults of the circuit breaker operating mechanism, and obtain the characteristic parameters of the multi-dimensional perception signal of the operating mechanism under fault.

[0088] A characteristic parameter database corresponding to different fault types is established based on the characteristic parameters of the dimensional perception signal.

[0089] The real-time mechanical performance parameters of the circuit breaker operating mechanism collected in real time are compared with the data in the characteristic parameter database corresponding to different fault types.

[0090] The data matching the data in the characteristic parameter database corresponding to different fault types are extracted to obtain the characteristic parameters of the circuit breaker operating mechanism. By statistically analyzing and trending the operating data of the circuit breaker operating mechanism, the occurrence of anomalies and faults can be detected. By analyzing the historical data, when the real-time data differs greatly from the health model of the circuit breaker operating mechanism, it can be determined that there may be a fault. By establishing a characteristic parameter database corresponding to different fault types, it is helpful to classify and diagnose mechanism faults. By collecting and organizing the information of known fault samples, constructing fault patterns and matching rules, it can be compared with mechanism data during real-time monitoring to further improve the accuracy and reliability of early warning.

[0091] By studying the relationship between different state types of data of the circuit breaker operating mechanism and combining the experience and knowledge of mechanism failure, a state evaluation system for the mechanical performance of the circuit breaker operating mechanism is established. By online monitoring of the mechanical state of the circuit breaker operating mechanism and analyzing the observation data of various characteristic parameters reflecting the mechanical state of the circuit breaker operating mechanism, the health state of the mechanical state of the circuit breaker operating mechanism is obtained to provide technical support for maintenance decisions. The same state characteristic parameters can reflect the status of the equipment from different angles. At the same time, the health score method is used to simply and intuitively reflect the health assessment results of the circuit breaker, thereby ensuring the accuracy of fault diagnosis.

[0092] Obtain the characteristic parameters of the multi-dimensional sensing signal of the operating mechanism under fault, including:

[0093] Extract the energy storage current data, and judge whether the energy storage motor of the circuit breaker operating mechanism is idling according to the change of the energy storage current;

[0094] Extract the frequency spectrum amplitude of the sound and vibration signal of the circuit breaker operating mechanism, the frequency spectrum waveform similarity or the speed in the giant magnetoresistance sensor monitoring signal, and judge whether the circuit breaker operating mechanism is not in place according to the frequency spectrum amplitude of the sound and vibration signal of the circuit breaker operating mechanism, the frequency spectrum waveform similarity or the speed change curve in the giant magnetoresistance sensor monitoring signal;

[0095] Extract the speed change curve or the duration of the spectrum waveform of the sound and vibration signal in the giant magnetoresistance sensor monitoring signal, and judge whether the opening and closing speed of the circuit breaker operating mechanism is too low according to the speed change curve or the duration of the spectrum waveform of the sound and vibration signal in the giant magnetoresistance sensor monitoring signal;

[0096] Extract the speed change curve or the spectrum waveform data of the sound and vibration signals in the giant magnetoresistance sensor monitoring signal, and judge whether the circuit breaker transmission mechanism of the circuit breaker operating mechanism is stuck, the circuit breaker spring energy storage is insufficient, or the circuit breaker operating mechanism is abnormally mechanically worn according to the speed change curve or the spectrum waveform similarity of the sound and vibration signals in the giant magnetoresistance sensor monitoring signal;

[0097] According to the results of the discrimination, the characteristic parameters of the multi-dimensional sensing signal of the operating mechanism under fault are obtained.

[0098] Acquire the characteristic parameters of the multi-dimensional perception signal of the operating mechanism under fault, establish a characteristic parameter database corresponding to different fault types, conduct a study on the relationship between data of different state types, and establish an expert system for state judgment of the mechanical performance of the circuit breaker operating mechanism in combination with the experience knowledge of mechanism failure. The mechanical state of the circuit breaker operating mechanism is monitored online through the sensor, and the observation data reflecting various characteristic parameters of the circuit breaker operating mechanism are analyzed to evaluate the health state of the circuit breaker operating mechanism, providing technical support for maintenance decisions. Different state characteristic parameters can reflect the status of the equipment from different angles. At the same time, the health score method is used to simply and intuitively reflect the health assessment results of the circuit breaker.

[0099] Extract characteristic parameters related to the mechanical properties of the circuit breaker operating mechanism from the processed circuit breaker operating mechanism operation data, including:

[0100] Based on the information entropy theory, the concept of cross-correlation entropy is proposed to extract the frequency domain features of the time series in the operation data of the circuit breaker operating mechanism, and six types of information entropies, including energy entropy, singular entropy, approximate entropy, sample entropy and fuzzy entropy of the vibration signal, are obtained as characteristic entropies.

[0101] The global features of the time domain and the frequency domain features in the operation data of the circuit breaker operating mechanism are extracted to form a feature matrix. The global features of the vibration signal are selected as the skewness coefficient, kurtosis coefficient, and peak coefficient. These features can complement the local features of the time domain and can more completely characterize the differences between different fault waveforms.

[0102] The feature entropy dimensionality reduction is used for normalization, and the average value of the corresponding dimension is subtracted from each feature of the feature matrix.

[0103] Compute the covariance matrix, where the diagonal and off-diagonal elements represent the autocovariance coefficients and cross-covariance coefficients, respectively.

[0104] Calculate the eigenvectors and eigenvalues ​​of the covariance matrix and sort them from large to small. Use the eigenvectors corresponding to the first n largest eigenvalues ​​as new orthogonal eigenmatrix values, which are the characteristic parameters related to the mechanical properties of the circuit breaker operating mechanism. The value of n can be determined based on the variance contribution rate.

[0105] Sensors, including:

[0106] The Hall sensor is used to extract the mechanical state information of the circuit breaker operating mechanism and the changes in load characteristics.

[0107] The soundprint and vibration sensor is used to detect the fault type by performing spectrum and envelope analysis on the sound and vibration signals of the circuit breaker operating mechanism: idling of the circuit breaker energy storage motor, inadequate opening and closing, too low opening and closing speed, stuck circuit breaker transmission mechanism, insufficient energy storage of the circuit breaker spring, abnormal mechanical wear of the circuit breaker operating mechanism. By analyzing the spectrum and envelope of the vibration signal of the circuit breaker operating mechanism, characteristic parameters such as vibration frequency, amplitude, kurtosis, etc. can be extracted to evaluate the health of the mechanism. According to the fault mode and mechanical characteristics, appropriate feature extraction methods can be selected, such as time domain features, frequency domain features, wavelet transform, modal analysis, etc., to capture the characteristics of the fault signal and perform quantitative analysis.

[0108] The image sensor is used to paste marks on the moving parts of the circuit breaker, and use a high-speed camera to record the displacement of the operating rod of the circuit breaker operating mechanism during the breaking process, obtain its stroke and time characteristics, and realize the detection of the mechanical characteristics of the high-voltage switch.

[0109] The giant magnetoresistance sensor is used to detect changes in angle and acceleration by sensing changes in the magnetic field according to the action of the transmission shaft of the circuit breaker operating mechanism during opening and closing. It can also detect fault types: inadequate opening and closing, too low opening and closing speed, stuck circuit breaker transmission mechanism, insufficient energy storage of circuit breaker springs, and abnormal mechanical wear of the circuit breaker operating mechanism.

[0110] The working principles of voiceprint and vibration sensors include:

[0111] The characteristics of the vibration signal of the circuit breaker operating mechanism are analyzed, the time domain waveforms of different fault types are compared, and the spectrum of the circuit breaker operating mechanism signal is analyzed after filtering.

[0112] Determine whether the signal spectrum of the circuit breaker operating mechanism meets the fault characteristics. If yes, determine the fault type according to the fault characteristics.

[0113] The features of all fault types are combined to generate a feature matrix, and the feature matrix is ​​compressed using principal component reduction (PCA).

[0114] Diagnose circuit breaker operating mechanism faults based on the compressed special diagnosis matrix.

[0115] The extracted real-time characteristic parameters are input into the health model of the circuit breaker operating mechanism to calculate the health score of the circuit breaker operating mechanism, including:

[0116] When the health status score is 100, the assessment level is no maintenance.

[0117] When the health status score is 90≤X<100, the assessment level is D-level maintenance.

[0118] When the health status score is 80≤X<90, the assessment level is C-level maintenance.

[0119] When the health status score is 60≤X<80, the assessment level is B-level maintenance.

[0120] When the health status score is 0≤X<60, the assessment level is A-level maintenance.

[0121] Preferably, it also includes:

[0122] Extract the motion profile of the circuit breaker operating mechanism based on the Hall sensor;

[0123] Optimizing the action profile based on the optimization model;

[0124]

[0125] Where W(k) is the discrete signal of the motion profile before optimization processing; W N (k) is the optimized signal of the motion profile obtained after N iterations in the optimization model; S is the motion profile of the circuit breaker operating mechanism before optimization; M is the number of pixel points on the motion profile of the circuit breaker operating mechanism; T N-1 (k+i) is the weight at the i-th pixel at the N-1th iteration in the optimization model, T N-1 (k+i)>0;W N-1 (k+i) is the optimization signal at the i-th pixel at the N-1th iteration in the optimization model;

[0126] Selecting any pixel point other than the central pixel point on the optimized motion contour as a segmentation pixel point, segmenting the motion contour into a first sub-motion contour and a second sub-motion contour based on the segmentation pixel point, and determining the center points of the first sub-motion contour and the second sub-motion contour as the first feature point and the second feature point respectively;

[0127] Obtaining a first posture angle corresponding to the first sub-motion profile according to the first feature point and the segmented pixel point;

[0128]

[0129] Among them, θ a (j) is the first posture angle corresponding to the first sub-motion profile; y(j) is the ordinate of the segmented pixel point j; x(j) is the abscissa of the segmented pixel point j; y a (j) is the ordinate of the first feature point; x a (j) is the horizontal coordinate of the first feature point;

[0130] Obtaining a second posture angle corresponding to the second sub-motion profile according to the second feature point and the segmented pixel point;

[0131]

[0132] Among them, θ b (j) is the second posture angle corresponding to the second sub-motion profile; y b (j) is the ordinate of the second feature point; x b (j) is the horizontal coordinate of the second feature point;

[0133] The preset posture data table is queried according to the first posture angle corresponding to the first sub-motion profile and the second posture angle corresponding to the second sub-motion profile to determine the motion posture information of the circuit breaker operating mechanism as the mechanical state information of the circuit breaker operating mechanism.

[0134] The working principle and beneficial effects of the above technical solution are as follows: First, the action profile of the circuit breaker operating mechanism is extracted based on the Hall sensor. In the process of optimizing the action profile based on the optimization model, step 1, the initial position v(0) of the action profile is given based on the initial estimation or prior knowledge; step 2, iterative calculation: select a larger time step Δt, use the iterative formula to calculate, and obtain the iteration result v(nΔt) of each step. This step is to gradually approach the real action profile through iteration. Step 3, convergence judgment: judge whether the current profile converges to a given threshold. If so, stop the iteration; otherwise, proceed to the next step. Step 4, direction adjustment and step optimization: if it is found that the movement direction of the profile is reversed (i.e., jitter or discontinuity occurs) during the iteration process, the profile is retracted to the previous step (such as the n-1th step), and the time step Δt is halved (to Δt / 2). Then the iteration result v(nΔt) of the current step is recalculated. This step is to avoid the profile crossing the equilibrium point due to excessive step length. Continuous iteration and optimization: repeat steps 3 to 4 until the profile converges to a given threshold. In this way, the time step Δt can be dynamically adjusted so that the algorithm can converge to the equilibrium point faster and reduce the number of unnecessary iterations. It is convenient to realize fast optimization processing of the motion profile based on the optimization model. Accurately calculate the optimization signal of the motion profile obtained after N iterations in the optimization model, select any pixel point other than the central pixel point on the motion profile after optimization processing, the purpose is to realize unequal segmentation of the motion profile, divide the motion profile into a first sub-motion profile and a second sub-motion profile based on the segmentation pixel point, determine the center point of the first sub-motion profile and the second sub-motion profile respectively, as the first feature point and the second feature point; obtain the first posture angle corresponding to the first sub-motion profile according to the first feature point and the segmentation pixel point, obtain the second posture angle corresponding to the second sub-motion profile according to the second feature point and the segmentation pixel point, query the preset posture data table according to the first posture angle corresponding to the first sub-motion profile and the second posture angle corresponding to the second sub-motion profile, and the preset posture data table is a comparison data table of the first posture angle-the second posture angle-motion posture information. When determining the motion posture information of the circuit breaker operating mechanism, the motion posture information of the circuit breaker operating mechanism is accurately determined based on the motion profile and the angle information, and then the mechanical state information of the circuit breaker operating mechanism is accurately determined.

[0135] In summary, the method for diagnosing the mechanical state fault of the circuit breaker operating mechanism of the present invention uses multi-scale entropy to extract the features of the mechanical state of the circuit breaker operating mechanism, thereby ensuring the richness and criticality of the feature information of the mechanical state feature extraction of the circuit breaker operating mechanism, and also avoiding the redundancy of the feature decomposition when extracting the features of the mechanical state of the circuit breaker operating mechanism. According to different types of parameters of the mechanical characteristics of the circuit breaker, an evaluation model that can evaluate the operating state of the circuit breaker is constructed, and the concept of intuition is added on the basis of the fuzzy hierarchical analysis method. The mechanical state of the circuit breaker operating mechanism is evaluated by using the intuitive hierarchical analysis method, thereby ensuring the accuracy of fault diagnosis.

[0136] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0137] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for diagnosing mechanical state faults of a circuit breaker operating mechanism, characterized in that: The following steps are involved: Step 1: Obtain historical mechanical characteristics of the circuit breaker operating mechanism, analyze the mechanical characteristics of the circuit breaker operating mechanism, and obtain historical mechanical performance parameters of the circuit breaker operating mechanism under normal working conditions; Step 2: Establish a health model of the circuit breaker operating mechanism according to the historical mechanical performance parameters of the circuit breaker operating mechanism under normal working conditions; Step 3: The operating data of the circuit breaker operating mechanism is collected in real time through sensors, and then the collected data is preprocessed, which includes cleaning, transformation and abnormality detection of the collected data; Step 4: extracting characteristic parameters related to the mechanical properties of the circuit breaker operating mechanism from the preprocessed operating data, and establishing a feature extraction algorithm based on the extracted characteristic parameters related to the mechanical properties of the circuit breaker operating mechanism; Step 5: Quantitatively analyze the mechanical characteristics according to the established feature extraction algorithm, and obtain the real-time mechanical performance parameters of the circuit breaker operating mechanism through simulation; Step six: Analyze the spectrum and envelope of the real-time mechanical performance parameters of the circuit breaker operating mechanism, extract the real-time characteristic parameters of the circuit breaker operating mechanism, input the extracted real-time characteristic parameters into the health model of the circuit breaker operating mechanism, calculate the health score of the circuit breaker operating mechanism, and diagnose the circuit breaker operating mechanism fault according to the health score of the circuit breaker operating mechanism.

2. The circuit breaker operating mechanism mechanical state fault diagnosis method according to claim 1, characterized in that: The establishing of the circuit breaker operating mechanism health model according to the historical mechanical performance parameters of the circuit breaker operating mechanism under normal working conditions specifically includes: Obtain the historical mechanical performance parameters of the circuit breaker operating mechanism under normal working conditions, and train the model through the historical mechanical performance parameters. During the training process, the weights and parameters of the model are adjusted by the optimization algorithm by minimizing the loss function. The historical mechanical performance parameters are the transmission ratio, transmission efficiency, and motion stability of the circuit breaker operating mechanism under normal working conditions. The historical mechanical performance parameters are divided into a training set and a validation set. The validation set is used to evaluate the performance indicators of the model. According to the loss function of the training set, adjust the weights and parameters of the model so that the model gradually converges and fits the historical data; During the training process, the performance indicators on the validation set are used to evaluate the effect of the model, and the hyperparameters and parameters of the model are adjusted based on the evaluation results; Repeat parameter adjustment and verification until the optimal parameter setting is found and the circuit breaker operating mechanism health model is obtained.

3. The circuit breaker operating mechanism mechanical state fault diagnosis method according to claim 2, characterized in that: The method of repeatedly adjusting and verifying parameters until the best parameter setting is found also includes: adjusting the hyperparameters and parameters of the model according to the performance of the verification set, and finding the best parameter setting by experimenting with different parameter combinations and using grid search or random search.

4. The circuit breaker operating mechanism mechanical state fault diagnosis method according to claim 1, characterized in that: The analysis of the spectrum and envelope of the real-time mechanical performance parameters of the circuit breaker operating mechanism to extract the real-time characteristic parameters of the circuit breaker operating mechanism specifically includes: Obtain the historical types of mechanical faults of the circuit breaker operating mechanism, and obtain the characteristic parameters of the multi-dimensional sensing signal of the operating mechanism under the fault; Establish a characteristic parameter database corresponding to different fault types based on the characteristic parameters of the multi-dimensional sensing signal, and compare the real-time mechanical performance parameters of the circuit breaker operating mechanism collected in real time with the data in the characteristic parameter database corresponding to different fault types; The data matching the data in the characteristic parameter database corresponding to different fault types are extracted to obtain the real-time characteristic parameters of the circuit breaker operating mechanism.

5. The circuit breaker operating mechanism mechanical state fault diagnosis method according to claim 4, characterized in that: The obtaining of the characteristic parameters of the multi-dimensional sensing signal of the operating mechanism under fault specifically includes: Extract the energy storage current data, and judge whether the energy storage motor of the circuit breaker operating mechanism is idling according to the change of the energy storage current; Extract the frequency spectrum amplitude of the sound and vibration signal of the circuit breaker operating mechanism, the frequency spectrum waveform similarity or the speed in the giant magnetoresistance sensor monitoring signal, and judge whether the circuit breaker operating mechanism is not in place according to the frequency spectrum amplitude of the sound and vibration signal of the circuit breaker operating mechanism, the frequency spectrum waveform similarity or the speed change curve in the giant magnetoresistance sensor monitoring signal; Extract the speed change curve or the duration of the spectrum waveform of the sound and vibration signal in the giant magnetoresistance sensor monitoring signal, and judge whether the opening and closing speed of the circuit breaker operating mechanism is too low according to the speed change curve or the duration of the spectrum waveform of the sound and vibration signal in the giant magnetoresistance sensor monitoring signal; Extract the speed change curve or the spectrum waveform data of the sound and vibration signals in the giant magnetoresistance sensor monitoring signal, and judge whether the circuit breaker transmission mechanism of the circuit breaker operating mechanism is stuck, the circuit breaker spring energy storage is insufficient, or the circuit breaker operating mechanism is abnormally mechanically worn according to the speed change curve or the spectrum waveform similarity of the sound and vibration signals in the giant magnetoresistance sensor monitoring signal; According to the results of the discrimination, the characteristic parameters of the multi-dimensional sensing signal of the operating mechanism under fault are obtained.

6. The circuit breaker operating mechanism mechanical state fault diagnosis method according to claim 1, characterized in that: The feature extraction algorithm is established according to the extracted data, specifically comprising: The frequency domain characteristics of the time series in the operation data of the circuit breaker operating mechanism are extracted, and the energy entropy, singular entropy, approximate entropy, sample entropy and fuzzy entropy of the vibration signal are obtained as characteristic entropy; The global features in the time domain and the frequency domain in the operation data of the circuit breaker operating mechanism are extracted to form a feature matrix, and the global features of the vibration signal are selected as the skewness coefficient, the kurtosis coefficient, and the peak coefficient. Use feature entropy dimensionality reduction to perform normalization processing, and subtract the average value of the corresponding dimension from each feature of the feature matrix; Calculate the covariance matrix, where the diagonal and off-diagonal elements represent the autocovariance coefficients and cross-covariance coefficients, respectively; Calculate the eigenvectors and eigenvalues ​​of the covariance matrix and sort them from large to small, and use the eigenvectors corresponding to the first n largest eigenvalues ​​as the new orthogonal eigenmatrix values.

7. The circuit breaker operating mechanism mechanical state fault diagnosis method according to claim 1, characterized in that: The sensor specifically comprises: Hall sensor, used to extract mechanical state information of the circuit breaker operating mechanism and changes in load characteristics; Voiceprint and vibration sensors are used to perform spectrum and envelope analysis on the sound and vibration signals of the circuit breaker operating mechanism; Image sensor, used to paste marks on the moving parts of the circuit breaker, use a high-speed camera to record the displacement of the operating rod of the circuit breaker operating mechanism during the breaking process, and obtain its stroke and time characteristics; The giant magnetoresistance sensor is used to detect changes in angle and acceleration by sensing changes in the magnetic field based on the movement of the transmission shaft of the circuit breaker operating mechanism during opening and closing.

8. The circuit breaker operating mechanism mechanical state fault diagnosis method according to claim 7, characterized in that: The workflow of the voiceprint and vibration sensor specifically includes: Perform feature analysis on the vibration signal of the circuit breaker operating mechanism, compare the time domain waveforms of different fault types, and analyze the spectrum of the circuit breaker operating mechanism signal after filtering; Determine whether the signal spectrum of the circuit breaker operating mechanism meets the fault characteristics. If yes, determine the fault type according to the fault characteristics. The features of all fault types are combined to generate a feature matrix, and the feature matrix is ​​compressed using principal component reduction (PCA). Diagnose circuit breaker operating mechanism faults based on the compressed special diagnosis matrix.

9. The circuit breaker operating mechanism mechanical state fault diagnosis method according to claim 7, characterized in that: Also includes: Extract the action profile of the circuit breaker operating mechanism based on the Hall sensor; Optimizing the action profile based on the optimization model; Where W(k) is the discrete signal of the motion profile before optimization processing; W N (k) is the optimized signal of the motion profile obtained after N iterations in the optimization model; S is the motion profile of the circuit breaker operating mechanism before optimization; M is the number of pixel points on the motion profile of the circuit breaker operating mechanism; T N-1 (k+i) is the weight at the i-th pixel at the N-1th iteration in the optimization model, T N-1 (k+i)>0;W N-1 (k+i) is the optimization signal at the i-th pixel at the N-1th iteration in the optimization model; Selecting any pixel point other than the central pixel point on the optimized motion contour as a segmentation pixel point, segmenting the motion contour into a first sub-motion contour and a second sub-motion contour based on the segmentation pixel point, and determining the center points of the first sub-motion contour and the second sub-motion contour as the first feature point and the second feature point respectively; Obtaining a first posture angle corresponding to the first sub-motion profile according to the first feature point and the segmented pixel point; Among them, θ a (j) is the first posture angle corresponding to the first sub-motion profile; y(j) is the ordinate of the segmented pixel point j; x(j) is the abscissa of the segmented pixel point j; y a (j) is the ordinate of the first feature point; x a (j) is the horizontal coordinate of the first feature point; Obtaining a second posture angle corresponding to the second sub-motion profile according to the second feature point and the segmented pixel point; Among them, θ b (j) is the second posture angle corresponding to the second sub-motion profile; y b (j) is the ordinate of the second feature point; x b (j) is the horizontal coordinate of the second feature point; The preset posture data table is queried according to the first posture angle corresponding to the first sub-motion profile and the second posture angle corresponding to the second sub-motion profile to determine the motion posture information of the circuit breaker operating mechanism as the mechanical state information of the circuit breaker operating mechanism.

10. The circuit breaker operating mechanism mechanical state fault diagnosis method according to claim 7, characterized in that: The step of inputting the extracted real-time characteristic parameters into the health model of the circuit breaker operating mechanism to calculate the health score of the circuit breaker operating mechanism specifically includes: When the health status score is 100, the assessment level is no maintenance; When the health status score is 90≤X<100, the assessment level is D-level maintenance; When the health status score is 80≤X<90, the assessment level is C-level maintenance; When the health status score is 60≤X<80, the assessment level is B-level maintenance; When the health status score is 0≤X<60, the assessment level is A-level maintenance.

Citation Information

Patent Citations

  • Circuit breaker fault diagnosis method based on vibration signal feature entropy fusion

    CN113408458A

  • Road condition identification method based on multi-source heterogeneous data fusion

    CN116738211A

  • Air compressor fault diagnosis method and device based on vibration detection

    CN117272032A

  • Intelligent wheelchair control method based on multi-sensing information fusion

    CN118105251A

Cited By

  • Fault diagnosis method and system for valve actuating mechanism of steam turbine

    CN119756842A

  • Circuit breaker fault diagnosis system and method

    CN120629918A