Fault diagnosis method and device for disconnecting switch, equipment and storage medium
The power signal of the isolating switch is decomposed through the weighted set empirical modal decomposition method, which solves the modal aliasing problem and improves the accuracy and effect of fault diagnosis.
Patent Information
- Application Number
- CN202411740029.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-29
- Publication Date
- 2025-03-21
AI Technical Summary
In the fault diagnosis of the isolating switch, the empirical modal decomposition method has a modal aliasing problem, which leads to inaccurate decomposition results and affects the effect of fault diagnosis.
The power signal of the driving motor is decomposed by weighted ensemble empirical modal decomposition method. By adding Gaussian white noise of different amplitudes and performing weighted averages, modal aliasing is reduced and the accuracy of the decomposition results is improved.
Through the weighted set empirical modal decomposition method, the components of different frequency signals can be effectively distinguished, the chance of modal aliasing is reduced, the ability to extract fault characteristics is enhanced, and the fault diagnosis effect of the isolating switch is improved.
Smart Images
Figure CN119667462A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of fault diagnosis, and in particular to a fault diagnosis method and device, equipment and storage medium for an isolating switch. Background Art
[0002] The disconnector in the gas insulated switchgear (GIS) is a key device in the power system. Its main function is to achieve electrical isolation and ensure the safe operation of the power system. With the continuous development and complexity of the power system, the demand for fault diagnosis of disconnectors is also increasing.
[0003] Disconnectors are usually controlled by drive motors, and the reliability of their operation directly affects the stability of the power system. During the operation of the drive motor, its power signal contains a wealth of equipment operation information. The power signal itself has nonlinear and non-stationary characteristics. Empirical mode decomposition (EMD), as an adaptive signal decomposition method, can decompose complex signals into a series of intrinsic mode functions, thereby extracting the essential characteristics of the signal. However, the EMD method has the problem of modal aliasing during application, that is, different modes interfere with each other, resulting in inaccurate decomposition results, which in turn affects the effect of fault diagnosis. Summary of the invention
[0004] Based on this, it is necessary to propose a fault diagnosis method, device, equipment and storage medium for disconnectors to address the above problems, so as to improve the accuracy of the decomposition results and thus improve the accuracy of the fault diagnosis results.
[0005] To achieve the above-mentioned object, the present application provides a first aspect of a fault diagnosis method for an isolating switch, the method comprising:
[0006] Collecting a power signal of a drive motor, wherein the drive motor is used to control the operation of the isolating switch;
[0007] Decomposing the power signal based on weighted set empirical mode decomposition to obtain intrinsic mode functions of different levels of the power signal;
[0008] Extracting fault features from intrinsic mode functions of different levels of the power signal;
[0009] A fault classification model is constructed, the fault feature is input into the fault classification model, and the fault classification model outputs a fault identification result, wherein the fault identification result is the fault type and fault probability of the disconnector.
[0010] Furthermore, the power signal is decomposed based on weighted set empirical mode decomposition to obtain intrinsic mode functions of different levels of the power signal, specifically including:
[0011] Adding N Gaussian white noises of different amplitudes to the power signal to obtain N different signals to be decomposed after adding Gaussian white noises;
[0012] Performing modal decomposition on each of the signals to be decomposed respectively to obtain initial intrinsic mode functions of different levels corresponding to each of the signals to be decomposed;
[0013] A weighted average operation is performed on the initial intrinsic mode functions of the same level in all the signals to be decomposed to obtain the intrinsic mode functions of different levels of the power signal.
[0014] Further, performing a weighted average operation on the initial intrinsic mode functions of the same level in all the signals to be decomposed to obtain the intrinsic mode functions of different levels of the power signal specifically includes:
[0015] Determining a target weight corresponding to a target initial intrinsic mode function based on time domain and frequency domain characteristics of the target initial intrinsic mode function, wherein the target initial intrinsic mode function is an initial intrinsic mode function of any level in any signal to be decomposed;
[0016] A weighted average operation is performed according to the initial intrinsic mode functions of the same level in all the signals to be decomposed and the target weights corresponding to the initial intrinsic mode functions of the same level to obtain the intrinsic mode functions of different levels of the power signal.
[0017] Furthermore, the intrinsic mode functions of different levels of the power signal are calculated by the following formula:
[0018]
[0019] In the formula, is the intrinsic mode function of the power signal at level i, IMF i,n (t) is the initial intrinsic mode function of the i-th level of the n-th signal to be decomposed, 1≤n≤N, and n is an integer, w i，n is the target weight corresponding to the initial intrinsic mode function of the i-th level of the n-th signal to be decomposed.
[0020] Further, the determining the target weight corresponding to the target initial intrinsic mode function based on the time domain and frequency domain characteristics of the target initial intrinsic mode function specifically includes:
[0021] Obtaining instantaneous energy, instantaneous frequency and instantaneous phase of the target initial intrinsic mode function;
[0022] A weight calculation is performed according to the instantaneous energy, instantaneous frequency and instantaneous phase of the target initial intrinsic mode function, as well as a preset nonlinear adjustment factor and an adjustment coefficient, to obtain a target weight corresponding to the target initial intrinsic mode function.
[0023] Furthermore, the target weight corresponding to the target initial intrinsic mode function is calculated by the following formula:
[0024]
[0025] In the formula, w i，n (t) is the target weight corresponding to the target initial intrinsic mode function, 1≤n≤N, and n is an integer, i represents the i-th time of adding Gaussian white noise, E i，n (t) is the instantaneous energy of the target initial eigenmode function, f i，n (t) is the instantaneous frequency of the target initial eigenmode function, θ i，n (t) is the instantaneous phase of the target initial intrinsic mode function, α, β, γ are preset adjustment coefficients, and p, q, r are preset nonlinear adjustment factors.
[0026] Furthermore, the extracting fault features from the intrinsic mode functions of different levels of the power signal specifically includes:
[0027] extracting initial fault features from the intrinsic mode functions at different levels of the power signal;
[0028] The initial fault features are screened using a principal component analysis method to obtain target fault features in the intrinsic mode functions of different levels of the power signal.
[0029] To achieve the above-mentioned object, the second aspect of the present application provides a fault diagnosis device for an isolating switch, the device comprising a data acquisition unit, a data analysis unit and a fault identification unit;
[0030] The data acquisition unit is used to collect a power signal of a drive motor, and the drive motor is used to control the operation of the isolating switch;
[0031] The data analysis unit is used to decompose the power signal based on weighted set empirical mode decomposition to obtain intrinsic mode functions of different levels of the power signal;
[0032] Extracting fault features from intrinsic mode functions of different levels of the power signal;
[0033] The fault identification unit is used to construct a fault classification model, input the fault feature into the fault classification model, and the fault classification model outputs a fault identification result, which is the fault type and fault probability of the disconnector.
[0034] To achieve the above-mentioned purpose, the third aspect of the present application provides a computer-readable storage medium storing a computer program, characterized in that when the computer program is executed by a processor, the processor executes the steps of the method described in the first aspect.
[0035] To achieve the above-mentioned purpose, the fourth aspect of the present application provides a computer device, including a memory and a processor, characterized in that the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the method described in the first aspect.
[0036] The embodiments of the present invention have the following beneficial effects:
[0037] The embodiment of the present invention proposes a fault diagnosis method for an isolating switch, the method comprising: collecting a power signal of a driving motor, the driving motor is used to control the operation of the isolating switch; decomposing the power signal based on weighted set empirical mode decomposition to obtain the intrinsic mode functions of different levels of the power signal; extracting fault features in the intrinsic mode functions of different levels of the power signal; constructing a fault classification model, inputting the fault features into the fault classification model, the fault classification model outputs a fault identification result, and the fault identification result is the fault type and fault probability of the isolating switch. The present invention decomposes the power signal by a weighted set empirical mode decomposition method, assigns different weights to the decomposition results under different noise addition conditions, and more effectively distinguishes the components of different frequency signals to reduce the probability of modal aliasing, enhance the ability to extract fault features, and thus improve the fault diagnosis effect of the isolating switch. BRIEF DESCRIPTION OF THE DRAWINGS
[0038] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0039] in:
[0040] Figure 1 Schematic diagram of a flow chart of a fault diagnosis method for an isolating switch in an embodiment of the present invention;
[0041] Figure 2 is a structural block diagram of a fault diagnosis device for an isolating switch in an embodiment of the present invention;
[0042] Figure 3 1 is a diagram showing the internal structure of a computer device in an embodiment of the present invention. DETAILED DESCRIPTION
[0043] 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.
[0044] In the embodiment of the present invention, a fault diagnosis method for an isolating switch is proposed, which can be referred to Figure 1 , Figure 1 The figure is a flow chart of a fault diagnosis method for an isolating switch in an embodiment of the present invention, and the method comprises:
[0045] Step 110: Collect the power signal of the drive motor, where the drive motor is used to control the operation of the isolating switch.
[0046] The power signal of the drive motor can reflect the mechanical state of the disconnector. During the operation of the disconnector, the drive motor needs to consume a certain amount of power to complete the opening and closing of the disconnector. If the disconnector has mechanical defects, such as jamming, wear of transmission parts, etc., the power required by the drive motor during operation will change. Therefore, by monitoring the power signal of the drive motor, these mechanical defects can be discovered in time to avoid accidents.
[0047] In the embodiment of the present invention, in order to obtain the accurate power signal of the drive motor, a high-precision and fast-response data acquisition device can be configured. For example, a current sensor and a voltage sensor. The current sensor can accurately measure the current change of the drive motor, and the voltage sensor can accurately measure the voltage change of the drive motor. At the same time, the data acquisition device should have a high sampling rate and large-capacity storage to ensure the integrity and accuracy of the data.
[0048] Specifically, the data sampling frequency should be set within a range that can capture all key information during the operation of the drive motor. For example, a sampling frequency of more than 10kHz can be selected. The data acquisition time window should cover the entire process from the start to the stop of the isolation switch to ensure that complete data on the motor operation is collected.
[0049] Step 120 : Decompose the power signal based on weighted set empirical mode decomposition to obtain intrinsic mode functions of different levels of the power signal.
[0050] In the embodiment of the present invention, before decomposing the power signal, the power signal needs to be preprocessed to filter out the noise in the power signal. In the process of power signal acquisition, high-frequency noise and various interference signals often exist. In order to extract useful fault features from the power signal, the power signal needs to be denoised first.
[0051] Specifically, denoising methods such as wavelet transform, mean filtering and adaptive filtering can be used. Wavelet transform can effectively remove high-frequency noise in the power signal while retaining the mutation characteristics of the power signal; mean filtering eliminates random noise in the power signal through smoothing; adaptive filtering dynamically adjusts the filtering parameters according to the characteristics of the power signal to filter the power signal, which can achieve better denoising effect.
[0052] In addition, in order to facilitate subsequent signal analysis operations, the power signal needs to be organized within a unified scale range. The preprocessed power signal can be normalized to eliminate the influence of different dimensions on fault feature extraction and fault diagnosis. The normalization method can be minimum-maximum normalization or Z-score normalization. Minimum-maximum normalization scales the power signal data to the [0,1] interval, while Z-score normalization adjusts the data to a normal distribution with a mean of 0 and a standard deviation of 1.
[0053] After denoising and normalizing the power signal, the power signal is decomposed using weighted set empirical mode decomposition to obtain intrinsic mode functions of different levels.
[0054] Weighted Ensemble Empirical Mode Decomposition (WEEMD) is an improved method based on ensemble empirical mode decomposition (EEMD). EEMD solves the modal aliasing problem of empirical mode decomposition (EMD) by adding Gaussian white noise. EEMD adds Gaussian white noise of different amplitudes to the power signal to increase the local disturbance of the signal, and then performs multiple EMD decompositions on the noisy power signal. By averaging the multiple decomposition results, the influence of noise is eliminated and a stable intrinsic mode function is obtained. The improved weighted ensemble empirical mode decomposition is to assign a weight to each decomposition result when averaging the decomposition results to reduce the interference of noise on the power signal and enhance the ability to extract the power signal features. The weight can be calculated by different adaptive methods.
[0055] Step 130: extract fault features from the intrinsic mode functions of different levels of the power signal.
[0056] In the embodiment of the present invention, among the different levels of intrinsic mode functions obtained by decomposition, only the intrinsic mode functions of some levels are related to the fault characteristics. Therefore, the intrinsic mode functions of all levels can be screened first, and only the main modes related to the fault characteristics are retained. Specifically, the eigenmode functions of the main layers can be determined by analyzing the energy, frequency distribution and instantaneous characteristics of the eigenmode functions.
[0057] Key features are extracted from the intrinsic mode functions of the main levels of the power signal for fault diagnosis. The fault features can be instantaneous frequency, energy, and kurtosis, among which the instantaneous frequency reflects the frequency change of the intrinsic mode function; the energy reflects the intensity of the intrinsic mode function; and the kurtosis is used to describe the sharpness of the intrinsic mode function.
[0058] Step 140: construct a fault classification model, input the fault characteristics into the fault classification model, and the fault classification model outputs a fault identification result, which is the fault type and fault probability of the disconnector.
[0059] In an embodiment of the present invention, the power data of the drive motor under normal and fault conditions are collected to construct a training data set. The robustness of the fault classification model is increased by collecting power data under different working conditions and operating conditions. Each training sample in the training data set should include power data and the corresponding fault type. The training data set should contain data of multiple fault types to ensure the generalization ability of the fault classification model.
[0060] By selecting an appropriate classification algorithm, a fault classification model is constructed. The classification algorithm can be a support vector machine (SVM), a random forest, or a neural network, etc. The model is trained using a training data set to obtain a fault classification model.
[0061] After training and generating the fault classification model, the fault classification model is verified by methods such as cross-validation. Cross-validation divides the training data set into a training set and a validation set, repeatedly trains and validates the model, and evaluates the accuracy of the fault classification method through performance indicators. The performance indicators for evaluating the fault classification model can be accuracy, precision, recall rate, and F1-score. Only when the performance indicators of the fault classification model meet the preset standards can the fault classification work be put into use. If not, the fault classification model is adjusted to ensure the reliability and accuracy of the fault classification model in practical applications.
[0062] The fault features extracted in steps 110 - 130 are input into the fault classification model to perform fault diagnosis, thereby determining the possible fault type and fault probability of the drive motor at this time, so that maintenance measures can be taken in a timely manner based on the diagnosis results.
[0063] The embodiment of the present invention decomposes the power signal through the weighted set empirical mode decomposition method, assigns different weights to the decomposition results under different noise addition conditions, and more effectively distinguishes the components of signals of different frequencies, so as to reduce the probability of modal aliasing, enhance the ability to extract fault features, and thereby improve the fault diagnosis effect of the disconnector.
[0064] In one embodiment of the present invention, step 120, decomposing the power signal based on weighted set empirical mode decomposition to obtain intrinsic mode functions of different levels of the power signal, specifically includes:
[0065] Step 210, adding N Gaussian white noises of different amplitudes to the power signal to obtain N different signals to be decomposed after adding Gaussian white noises.
[0066] Specifically, Gaussian white noise of different amplitudes is added to the power signal multiple times to obtain the signal to be decomposed x′(t):
[0067] x n ′(t)=x(t)+∈ n (t)
[0068] Where x(t) is the power signal, ∈ n (t) is the Gaussian white noise added for the nth time, 0≤n≤N.
[0069] Step 220, perform modal decomposition on each signal to be decomposed, and obtain initial intrinsic mode functions of different levels corresponding to each signal to be decomposed.
[0070] Specifically, each signal to be decomposed is decomposed multiple times to obtain the initial intrinsic mode functions of different levels of each signal to be decomposed. The process can be expressed by the following formula:
[0071]
[0072] In the formula, x n ′(t) is the nth signal to be decomposed after adding Gaussian white noise for the nth time, IMF i，n (t) is the initial intrinsic mode function of the i-th level corresponding to the n-th signal to be decomposed, and r(t) is the residual signal.
[0073] Based on this, the initial intrinsic mode functions of m levels corresponding to each signal to be decomposed can be obtained.
[0074] Step 230, performing a weighted average operation on the initial intrinsic mode functions of the same level in all the signals to be decomposed, to obtain the intrinsic mode functions of different levels of the power signal.
[0075] Specifically, the initial intrinsic mode functions of the same level in each signal to be decomposed are weighted and calculated to obtain the intrinsic mode functions of the level corresponding to the power signal.
[0076] In one embodiment of the present invention, Step 230, performing a weighted average operation on the initial intrinsic mode functions of the same level in all the signals to be decomposed to obtain the intrinsic mode functions of different levels of the power signal, specifically includes:
[0077] Step 231, determine the target weight corresponding to the target initial intrinsic mode function based on the time domain and frequency domain characteristics of the target initial intrinsic mode function, wherein the target initial intrinsic mode function is the initial intrinsic mode function of any level in any signal to be decomposed.
[0078] In an embodiment of the present invention, the time domain and frequency domain features may be the instantaneous energy, instantaneous frequency, and instantaneous phase of the intrinsic mode function. Based on this, Step 231, based on the time domain and frequency domain features of the target initial intrinsic mode function, determines the target weight corresponding to the target initial intrinsic mode function, specifically including: obtaining the instantaneous energy, instantaneous frequency, and instantaneous phase of the target initial intrinsic mode function; performing weight calculation according to the instantaneous energy, instantaneous frequency, and instantaneous phase of the target initial intrinsic mode function, as well as the preset nonlinear adjustment factor and adjustment coefficient, to obtain the target weight corresponding to the target initial intrinsic mode function.
[0079] Specifically, the adjustment coefficient controls the contribution of each feature to the total weight, and the nonlinear adjustment factor controls the nonlinear contribution of each feature, which is used to enhance the adaptability of nonlinear signals.
[0080] In one embodiment of the present invention, the target weight corresponding to the target initial intrinsic mode function is calculated by the following formula:
[0081]
[0082] In the formula, w i，n (t) is the target weight corresponding to the target initial intrinsic mode function, 1≤n≤N, and n is an integer, i represents the i-th time Gaussian white noise is added, E o，n (t) is the instantaneous energy of the target initial eigenmode function, f i，n (t) is the instantaneous frequency of the target initial eigenmode function, θ i，n (t) is the instantaneous phase of the target initial intrinsic mode function, α, β, γ are preset adjustment coefficients, and p, q, r are preset nonlinear adjustment factors.
[0083] Instantaneous energy is a quantification of the power signal strength, which can capture the amplitude change of the signal over time. When the energy of the signal is large, its contribution to the fault characteristics may be greater. By adjusting the nonlinear adjustment factor p corresponding to the instantaneous energy, the influence of the instantaneous energy can be controlled. If p>1, the signal with large energy will be suppressed; if p<1, the signal with large energy will be amplified. Therefore, adaptively adjusting the nonlinear adjustment factor p based on actual conditions can avoid excessive influence of energy maximum or minimum on feature extraction.
[0084] The instantaneous frequency is the frequency characteristic of the power signal, which is used to capture the changes of the signal in different frequency components. Generally speaking, the frequency characteristic is an important basis for detecting high-frequency faults (such as mechanical vibration, fatigue, etc.). By adjusting the nonlinear adjustment factor q corresponding to the instantaneous frequency, the sensitivity to different frequency components can be enhanced. If q>1, the low-frequency component will get more weight, and the influence of the high-frequency part will be weakened; if q<1, the high-frequency characteristics will be enhanced. Therefore, selecting a suitable nonlinear adjustment factor q helps to process multi-frequency signals, especially when the fault signal is hidden in the high-frequency or low-frequency part, the fault identification ability can be improved by the nonlinear adjustment factor q.
[0085] The instantaneous phase reflects the phase change of the power signal. The phase feature is very important in both vibration and rotating mechanical systems, especially in periodic fault signals. By adjusting the nonlinear adjustment factor r corresponding to the instantaneous phase, the contribution of the phase in the weight can be controlled. If r>1, the influence of the signal with a large phase change is weakened, and if r<1, the signal with a large phase change is enhanced. Adaptively adjusting the nonlinear adjustment factor r can improve the recognition ability of fault modes with obvious phase changes.
[0086] The present invention can dynamically adjust the weight distribution according to the different characteristics of the actual power signal by introducing a nonlinear adjustment factor, so that the weighted set empirical mode decomposition algorithm can flexibly deal with nonlinear and non-stationary signals. This nonlinear adjustment mechanism can improve the processing ability of complex signals, especially when the fault signal is masked or hidden by noise, the adjustment factor can help the algorithm better focus on key features, thereby improving the accuracy and robustness of fault diagnosis.
[0087] Step 232: Perform a weighted average operation based on the initial intrinsic mode functions of the same level in all signals to be decomposed and the target weights corresponding to the initial intrinsic mode functions of the same level to obtain the intrinsic mode functions of different levels of the power signal.
[0088] Specifically, after determining the target weights corresponding to the initial intrinsic mode functions of different levels in different signals to be decomposed, the weighted average of the initial intrinsic mode functions of the same level in different signals to be decomposed is calculated based on the target weights, and the weighted average is used as the intrinsic mode function of the power signal.
[0089] In one embodiment of the present invention, the intrinsic mode functions of different levels of the power signal are calculated by the following formula:
[0090]
[0091] In the formula, is the intrinsic mode function of the power signal at level i, IMF i,n(t) is the initial intrinsic mode function of the i-th level of the n-th signal to be decomposed, 1≤n≤N, and n is an integer, w i，n is the target weight corresponding to the initial intrinsic mode function of the i-th level of the n-th signal to be decomposed.
[0092] The embodiment of the present invention introduces a weight optimization strategy of nonlinear adjustment factors, which can adaptively adjust the weight distribution of instantaneous energy, frequency and phase of the power signal, and flexibly extract features for different power fluctuations, thereby more accurately capturing potential fault signals.
[0093] In one embodiment of the present invention, step 130, extracting fault features in the intrinsic mode functions of different levels of the power signal, specifically includes:
[0094] Step 310, extracting initial fault features from the intrinsic mode functions of different levels of the power signal.
[0095] Step 320, use the principal component analysis method to screen the initial fault features and obtain the target fault features in the intrinsic mode functions of different levels of the power signal.
[0096] Fault features are extracted from the intrinsic mode functions obtained based on the power signal decomposition. The fault features may include instantaneous frequency, energy, and kurtosis. After the fault features are extracted, a high-dimensional feature vector is usually obtained. In order to reduce the computational complexity, dimensionality reduction methods such as principal component analysis (PCA) can be used to reduce the dimensionality of high-dimensional features. For example, principal component analysis PCA is used to calculate the eigenvalues and eigenvectors of the fault feature covariance matrix, and the eigenvectors corresponding to the first k largest eigenvalues are selected as the target fault features to reduce redundant information.
[0097] In one embodiment of the present invention, the fault identification result output by the fault classification model may include the fault type, fault occurrence time, characteristic information and environmental conditions, etc. It can further compare the characteristic changes under normal and fault conditions, analyze the causes and laws of the fault, automatically generate a fault diagnosis report, describe the fault type, fault characteristics, fault occurrence time and recommended maintenance measures in detail, and send the fault diagnosis report to the server or the mobile terminal of the maintenance personnel so that the maintenance personnel can take maintenance measures in time. The fault report may include charts and text descriptions to facilitate the maintenance personnel to understand and take measures.
[0098] The present invention also provides a fault diagnosis device for an isolating switch, which can be referred to Figure 2 , Figure 2 2 is a structural block diagram of a fault diagnosis device for an isolating switch in an embodiment of the present invention. The device includes a data acquisition unit 201 , a data analysis unit 202 and a fault identification unit 203 .
[0099] The data acquisition unit 201 is used to acquire the power signal of the drive motor, and the drive motor is used to control the operation of the isolating switch.
[0100] The data analysis unit 202 is used to decompose the power signal based on weighted set empirical mode decomposition to obtain intrinsic mode functions of different levels of the power signal.
[0101] Fault features in the intrinsic mode functions of different levels of the power signal are extracted.
[0102] The fault identification unit 203 is used to construct a fault classification model, input the fault feature into the fault classification model, and the fault classification model outputs a fault identification result, which is the fault type and fault probability of the disconnector.
[0103] The fault diagnosis device for the isolating switch proposed in the embodiment of the present invention decomposes the power signal through the weighted set empirical mode decomposition method, assigns different weights to the decomposition results under different noise addition conditions, and more effectively distinguishes the components of signals of different frequencies, so as to reduce the probability of modal aliasing, enhance the ability to extract fault features, and thereby improve the fault diagnosis effect of the isolating switch.
[0104] Figure 3 FIG. 1 shows an internal structure diagram of a computer device in one embodiment of the present invention. The computer device may be a terminal or a system. Figure 3 As shown, the computer device includes a processor, a memory and a network interface connected via a system bus. The memory includes a non-volatile storage medium and an internal memory. The non-volatile storage medium of the computer device stores an operating system and may also store a computer program. When the computer program is executed by the processor, the processor may implement each step in the above method embodiment. The internal memory may also store a computer program. When the computer program is executed by the processor, the processor may implement each step in the above method embodiment. Those skilled in the art will understand that Figure 3 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.
[0105] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes each step in the above method embodiment.
[0106] In one embodiment, a computer-readable storage medium is proposed, which stores a computer program. When the computer program is executed by a processor, the processor executes each step in the above method embodiment. A person of ordinary skill in the art can understand that all or part of the process in the above embodiment method can be completed by instructing the relevant hardware through a computer program. The program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the process of the embodiment of each method as described above. Among them, any reference to memory, storage, database or other medium used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.
[0107] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
[0108] The above embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several modifications and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.
Claims
1. A fault diagnosis method for an isolating switch, characterized in that: The method comprises: Collecting a power signal of a drive motor, wherein the drive motor is used to control the operation of the isolating switch; Decomposing the power signal based on weighted set empirical mode decomposition to obtain intrinsic mode functions of different levels of the power signal; Extracting fault features from intrinsic mode functions of different levels of the power signal; A fault classification model is constructed, the fault feature is input into the fault classification model, and the fault classification model outputs a fault identification result, wherein the fault identification result is the fault type and fault probability of the disconnector.
2. The method according to claim 1, characterized in that The step of decomposing the power signal based on weighted set empirical mode decomposition to obtain intrinsic mode functions of different levels of the power signal specifically includes: Adding N Gaussian white noises of different amplitudes to the power signal to obtain N different signals to be decomposed after adding Gaussian white noises; Performing modal decomposition on each of the signals to be decomposed respectively to obtain initial intrinsic mode functions of different levels corresponding to each of the signals to be decomposed; A weighted average operation is performed on the initial intrinsic mode functions of the same level in all the signals to be decomposed to obtain the intrinsic mode functions of different levels of the power signal.
3. The method according to claim 2, characterized in that The step of performing a weighted average operation on the initial intrinsic mode functions of the same level in all the signals to be decomposed to obtain the intrinsic mode functions of different levels of the power signal specifically includes: Determining a target weight corresponding to a target initial intrinsic mode function based on time domain and frequency domain characteristics of the target initial intrinsic mode function, wherein the target initial intrinsic mode function is an initial intrinsic mode function of any level in any signal to be decomposed; A weighted average operation is performed according to the initial intrinsic mode functions of the same level in all the signals to be decomposed and the target weights corresponding to the initial intrinsic mode functions of the same level to obtain the intrinsic mode functions of different levels of the power signal.
4. The method according to claim 3, characterized in that The intrinsic mode functions of different levels of the power signal are calculated by the following formula: In the formula, is the intrinsic mode function of the power signal at level i, IMF i,n (t) is the initial intrinsic mode function of the i-th level of the n-th signal to be decomposed, 1≤n≤N, and n is an integer, w i，n is the target weight corresponding to the initial intrinsic mode function of the i-th level of the n-th signal to be decomposed.
5. The method according to claim 3, characterized in that The determining the target weight corresponding to the target initial intrinsic mode function based on the time domain and frequency domain characteristics of the target initial intrinsic mode function specifically includes: Obtaining instantaneous energy, instantaneous frequency and instantaneous phase of the target initial intrinsic mode function; A weight calculation is performed according to the instantaneous energy, instantaneous frequency and instantaneous phase of the target initial intrinsic mode function, as well as a preset nonlinear adjustment factor and an adjustment coefficient, to obtain a target weight corresponding to the target initial intrinsic mode function.
6. The method according to claim 5, characterized in that The target weight corresponding to the target initial intrinsic mode function is calculated by the following formula: In the formula, w i，n (t) is the target weight corresponding to the target initial intrinsic mode function, 1≤n≤N, and n is an integer, i represents the i-th time of adding Gaussian white noise, E i，n (t) is the instantaneous energy of the target initial eigenmode function, f i，n (t) is the instantaneous frequency of the target initial eigenmode function, θ i，n (t) is the instantaneous phase of the target initial intrinsic mode function, α, β, γ are preset adjustment coefficients, and p, q, r are preset nonlinear adjustment factors.
7. The method according to claim 3, characterized in that The extracting fault features from the intrinsic mode functions of different levels of the power signal specifically includes: extracting initial fault features from the intrinsic mode functions at different levels of the power signal; The initial fault features are screened using a principal component analysis method to obtain target fault features in the intrinsic mode functions of different levels of the power signal.
8. A fault diagnosis device for an isolating switch, characterized in that: The device comprises a data acquisition unit, a data analysis unit and a fault identification unit; The data acquisition unit is used to collect a power signal of a drive motor, and the drive motor is used to control the operation of the isolating switch; The data analysis unit is used to decompose the power signal based on weighted set empirical mode decomposition to obtain intrinsic mode functions of different levels of the power signal; Extracting fault features from intrinsic mode functions of different levels of the power signal; The fault identification unit is used to construct a fault classification model, input the fault feature into the fault classification model, and the fault classification model outputs a fault identification result, which is the fault type and fault probability of the disconnector.
9. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the processor is caused to perform the steps of the method according to any one of claims 1 to 7.
10. A computer device comprising a memory and a processor, characterized in that: The memory stores a computer program, and when the computer program is executed by the processor, the processor is caused to perform the steps of the method according to any one of claims 1 to 7.