Circuit breaker fault diagnosis method and system, and readable computer storage medium
By combining multiple characteristic quantities of vibration, sound and current signals, a circuit breaker operating status database is established, which solves the problem of low fault identification accuracy caused by single signal characteristic quantities and realizes rapid and accurate diagnosis of high-voltage circuit breakers.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- FUXIN POWER SUPPLY COMPANY STATE GRID LIAONING ELECTRIC POWER
- Filing Date
- 2022-12-02
- Publication Date
- 2026-06-23
AI Technical Summary
Existing fault diagnosis methods for high-voltage circuit breakers mainly rely on single signal characteristics, which lack comprehensiveness, resulting in low fault identification accuracy and large errors, and making comprehensive evaluation impossible.
By combining vibration, sound, and current signals, and fusing multiple signal features through a sparse autoencoder network, a circuit breaker operating status database is constructed, and the Minkowski distance is used for fault diagnosis.
It improves the accuracy and reliability of circuit breaker fault diagnosis, increases the fault tolerance rate, meets the accuracy and speed requirements of on-site diagnosis, and achieves rapid and accurate fault identification.
Smart Images
Figure CN115859053B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of electrical equipment fault detection, and in particular to a fault diagnosis method, system, and readable computer storage medium for circuit breakers. Background Technology
[0002] High-voltage circuit breakers are the most important operational safety control devices in power systems, and the losses caused by their failures far exceed the value of the circuit breaker itself. The opening and closing operation of a circuit breaker involves the conversion and transmission of electrical signals into mechanical actions, making it susceptible to both electrical and mechanical faults.
[0003] With the development of fault diagnosis technology, circuit breaker fault diagnosis methods are mainly based on research into opening and closing coil current signals, mechanical vibration signals, and sound signals. The vibration signal accompanying the circuit breaker operation is an excellent carrier of energy changes in the time and frequency domains, exhibiting low attenuation and easy capture. Using appropriate signal processing methods, state-related information of some key components in the high-voltage circuit breaker operating mechanism can be extracted from it. Acquiring the sound signal generated by the vibration is a non-contact measurement method, convenient to install, and has a wider measurement bandwidth compared to vibration signals. The operation of the high-voltage circuit breaker operating mechanism originates from the normal current signal appearing in the opening and closing coils. The coil current signal can reflect the state information of the secondary circuit, coil resistance, core mechanism, and control power supply, and can directly reflect electrical faults in the circuit breaker.
[0004] Currently, most online monitoring and fault identification technologies for the mechanical characteristics of high-voltage circuit breakers are based on the monitoring results of a single characteristic quantity. They rarely use multiple signals or different characteristic quantities of a single signal for comparison. The status information provided by a single type of signal is limited and lacks comprehensiveness. The number of faults that can be reflected and the accuracy of the fault response are limited, which may lead to large errors. Therefore, it is impossible to comprehensively evaluate the status of high-voltage circuit breakers. Summary of the Invention
[0005] To address the aforementioned technical problems, this invention provides a circuit breaker fault diagnosis method, characterized by comprising the following steps:
[0006] (1) Collect vibration, sound and current signals of the circuit breaker under normal conditions, and calculate the power spectrum of the vibration, sound and current signals.
[0007] (2) The power spectra of vibration signal, sound signal and current signal are used as input values and input into sparse autoencoder network. The output values of sparse autoencoder network are used as feature vectors of vibration signal, sound signal and current signal.
[0008] (3) The feature vectors of the vibration signal and the sound signal are fused together, and the fused feature vector is fused with the feature vector of the current signal to obtain the fused feature vector;
[0009] (4) Make the circuit breaker operate under different fault conditions, collect the corresponding vibration signal, sound signal and current signal, and calculate the power spectrum of the vibration signal, sound signal and current signal to construct the fusion feature vector under different fault conditions.
[0010] (5) Establish a circuit breaker operating status database based on the fused feature vectors obtained under normal and fault operating conditions;
[0011] (6) Collect vibration, sound and current signals of the circuit breaker under test during operation, and calculate the power spectrum of the vibration, sound and current signals to obtain the fused feature vector;
[0012] (7) Compare the distance between the fusion feature vector obtained in step (6) and the fusion feature vector in the circuit breaker operation status database. The operation status corresponding to the fusion feature vector with the shortest distance to the circuit breaker operation status database is the current operation status of the circuit breaker.
[0013] Furthermore, the method for calculating the power spectrum in step (1) is as follows:
[0014]
[0015] Where p(ω) is the power spectrum of the vibration signal, sound signal, and current signal, X T (jω) is the Fourier transform function for vibration signal, sound signal and current signal.
[0016] Furthermore, in step (2), the sparse autoencoder network has 3 hidden layers, each with a learning rate of 0.1, 600 iterations, and a weight decay parameter of 3 × 10⁻⁶. -5 The sparsity parameter is set to 0.1, the weight of the sparsity penalty term is 3, and the weight matrix and bias term of the autoencoder are randomly generated.
[0017] Furthermore, in step (3), the method for fusing the feature vectors of the vibration signal and the sound signal is parallel fusion, the algorithm used is the MKCCA algorithm, and the kernel function is the Gaussian radial basis function.
[0018] Furthermore, in step (3), the method of fusing the feature vector after fusing the feature vectors of the vibration signal and the sound signal with the feature vector of the current signal is to input the feature vector after fusing the feature vector of the current signal with the feature vectors of the vibration signal and the sound signal into the neural network, and the output value obtained is the fused feature vector.
[0019] Furthermore, in step (7), the fused feature vector is compared with the fused feature vector in the circuit breaker operation status database. The distance used is the Minkowski distance, and the p value in the Minkowski distance is 7.
[0020] In addition, the present invention also provides a circuit breaker operation status diagnostic system, including a measurement module and a data processing unit, characterized in that the data processing unit further includes:
[0021] The power spectrum calculation module is used to calculate the power spectrum of vibration signals, sound signals, and current signals.
[0022] Sparse autoencoder network is used to extract feature vectors of the power spectrum of vibration signals, sound signals, and current signals;
[0023] The first fusion module is used to fuse the feature vectors of vibration signals and sound signals;
[0024] The second fusion module is used to fuse the feature vector obtained by fusing the vibration signal and the sound signal with the feature vector of the current signal to obtain a fused feature vector;
[0025] A circuit breaker operating status database is used to store the fused feature vectors of circuit breakers under different operating states;
[0026] The distance calculation module is used to calculate the distance between different fused feature vectors;
[0027] The data comparison module is used to compare the distance values between different fused feature vectors.
[0028] Furthermore, the measurement module includes a circuit breaker, a vibration sensor, a sound sensor, and a current sensor. The vibration sensor is used to collect the vibration signal of the circuit breaker, the sound sensor is used to collect the sound signal of the circuit breaker, and the current sensor is used to collect the current signal of the circuit breaker.
[0029] In addition, the present invention also provides a readable computer storage medium storing a computer program, characterized in that the program, when executed by a processor, implements the circuit breaker operation status diagnosis method described above.
[0030] The circuit breaker fault diagnosis method provided by this invention is based on the joint analysis of sound, vibration, and current signals. Vibration, sound, and current signals all contain important information about the circuit breaker operation process. By combining information obtained from different sensors, the method can more accurately reflect the signal source's condition, effectively solving the problem that a single signal cannot fully distinguish the different states of the circuit breaker, thus improving the reliability of fault diagnosis. Calculating the power spectrum of the circuit breaker's sound, vibration, and current signals preserves the amplitude information of the spectrum. After training with a sparse autoencoder network, the most representative feature values of the signal can be extracted. By fusing the feature vectors of the original signals using a fusion neural network, optimal feature fusion can be achieved, making the feature values more representative, improving the accuracy of judgment, increasing the fault tolerance rate of circuit breaker fault diagnosis, and meeting the accuracy and speed requirements of on-site fault diagnosis of operating circuit breakers, enabling rapid and accurate identification. Attached Figure Description
[0031] The accompanying drawings, which form part of this application, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.
[0032] Figure 1 This is a flowchart of the circuit breaker fault diagnosis method of the present invention;
[0033] Figure 2 This is a schematic diagram of the circuit breaker fault diagnosis system of the present invention. Detailed Implementation
[0034] To enhance understanding of the present invention, the invention will be further described in detail below with reference to embodiments and accompanying drawings.
[0035] like Figure 2As shown, the circuit breaker fault diagnosis system of this embodiment includes a measurement module and a data processing unit. The measurement module includes a circuit breaker, a vibration sensor, a sound sensor, and a current sensor. The vibration sensor is used to collect the vibration signal of the circuit breaker, the sound sensor is used to collect the sound signal of the circuit breaker, and the current sensor is used to collect the current signal of the circuit breaker. The data processing unit includes a power spectrum calculation module, a sparse autoencoder network, a first fusion module, a second fusion module, a circuit breaker operating status database, a distance calculation module, and a data comparison module. The power spectrum calculation module is used to calculate the power spectra of the vibration signal, sound signal, and current signal. The sparse autoencoder network is used to extract feature vectors from the power spectra of the vibration signal, sound signal, and current signal. The first fusion module is used to fuse the feature vectors of the vibration signal and sound signal. The second fusion module is used to fuse the feature vector obtained after fusing the vibration signal and sound signal with the feature vector of the current signal to obtain a fused feature vector. The circuit breaker operating status database is used to store the fused feature vectors of the circuit breaker under different operating states. The distance calculation module is used to calculate the distance between different fused feature vectors. The data comparison module is used to compare the distance values between different fused feature vectors.
[0036] like Figure 1 As shown, the method for diagnosing circuit breaker faults using the circuit breaker fault diagnosis system of this embodiment includes the following steps:
[0037] (1) Make the circuit breaker operate under normal conditions, collect the vibration signal, sound signal and current signal of the circuit breaker under normal conditions, and calculate the power spectrum of the vibration signal, sound signal and current signal.
[0038] (2) The power spectra of vibration signal, sound signal and current signal are used as input values and input into sparse autoencoder network. The output values of sparse autoencoder network are used as feature vectors of vibration signal, sound signal and current signal.
[0039] (3) The feature vectors of the vibration signal and the sound signal are fused together, and the fused feature vector is fused with the feature vector of the current signal to obtain the fused feature vector;
[0040] (4) Make the circuit breaker operate under different fault conditions, use vibration sensors, sound sensors and current sensors to collect vibration signals, sound signals and current signals, and calculate the power spectrum of vibration signals, sound signals and current signals to construct a fusion feature vector under different fault conditions;
[0041] (5) Use the fused feature vectors obtained under normal and fault operating conditions as elements to establish a circuit breaker operating status database;
[0042] (6) Collect vibration, sound and current signals of the circuit breaker under test during operation, and calculate the power spectrum of the vibration, sound and current signals to obtain the fused feature vector;
[0043] (7) Compare the distance between the fusion feature vector obtained in step (6) and the fusion feature vector in the circuit breaker operation status database. The operation status corresponding to the fusion feature vector with the shortest distance to the circuit breaker operation status database is the current operation status of the circuit breaker.
[0044] The method for calculating the power spectrum in step (1) is as follows:
[0045]
[0046] Where p(ω) is the power spectrum of the vibration signal, sound signal, and current signal, X T (jω) is the Fourier transform function for vibration signal, sound signal and current signal.
[0047] In step (2), the sparse autoencoder network has 3 hidden layers, each with a learning rate of 0.1 and 600 iterations. The weight decay parameter of the autoencoder is 3 × 10⁻⁶. -5 The sparsity parameter is set to 0.1, the weight of the sparsity penalty term is 3, and the weight matrix and bias term of the autoencoder are randomly generated.
[0048] The method for fusing the feature vectors of vibration signal and sound signal in step (3) is parallel fusion, the algorithm used is MKCCA algorithm, and the kernel function is Gaussian radial basis function.
[0049] The method for fusing the feature vector of the vibration signal and the sound signal after fusing them with the feature vector of the current signal in step (3) is to input the feature vector of the current signal after fusing it with the feature vector of the vibration signal and the sound signal into the neural network, and the output value obtained is the fused feature vector.
[0050] In step (7), the fused feature vector is compared with the fused feature vector in the circuit breaker operation status database. The distance used is the Minkowski distance, and the p value in the Minkowski distance is 7.
[0051] Using the method of this embodiment to diagnose faults in circuit breakers under different operating conditions, the accuracy rate can reach 98.97% in the case of failure to close, and the accuracy rates for normal operation, linkage detachment and base loosening are 98.66%, 97.86% and 96.8% respectively, all of which are relatively high.
[0052] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code. The solutions in the embodiments of this application can be implemented in various computer languages, such as the object-oriented programming language Java and the interpreted scripting language JavaScript.
[0053] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.
[0054] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.
[0055] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.
[0056] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.
[0057] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.
Claims
1. A method for diagnosing circuit breaker faults, characterized in that, Includes the following steps: S1 collects the vibration, sound, and current signals of the circuit breaker under normal conditions and calculates the power spectrum of the vibration, sound, and current signals. S2 takes the power spectra of vibration signal, sound signal and current signal as input values and inputs them into a sparse autoencoder network. The output values of the sparse autoencoder network are used as the feature vectors of vibration signal, sound signal and current signal. S3 fuses the feature vectors of the vibration signal and the sound signal, and then fuses the fused feature vector with the feature vector of the current signal to obtain a fused feature vector. The method for fusing the feature vectors of vibration signals and sound signals is parallel fusion, using the MKCCA algorithm and the Gaussian radial basis function as the kernel function. S4 enables the circuit breaker to operate under different fault conditions, collects corresponding vibration, sound and current signals, calculates the power spectrum of the vibration, sound and current signals, and constructs a fused feature vector under different fault conditions. S5 uses the fused feature vectors obtained under normal and fault operating conditions as elements to establish a circuit breaker operating status database. S6 collects vibration, sound, and current signals of the circuit breaker under test during operation, and calculates the power spectrum of the vibration, sound, and current signals to obtain a fused feature vector. S7 compares the distance between the fused feature vector obtained in step S6 and the fused feature vector in the circuit breaker operating status database. The operating status corresponding to the fused feature vector with the shortest distance to the circuit breaker operating status database is the current operating status of the circuit breaker.
2. The circuit breaker fault diagnosis method according to claim 1, characterized in that, The power spectrum calculation method for step S1 is as follows: in: The power spectrum of vibration signals, sound signals, and current signals; Let T be the Fourier transform function of the vibration signal, sound signal, and current signal, and let T be the signal duration.
3. The circuit breaker fault diagnosis method as described in claim 1, characterized in that, In step S2, the sparse autoencoder network has 3 hidden layers, each with a learning rate of 0.1 and 600 iterations. The autoencoder's weight decay parameter is 3×10, the sparsity parameter is 0.1, the sparsity penalty term has a weight of 3, and the autoencoder weight matrix and bias term are randomly generated.
4. The circuit breaker fault diagnosis method as described in claim 1, characterized in that, The method for fusing the feature vector obtained by fusing the feature vectors of vibration signal and sound signal with the feature vector of current signal in step S3 is to input the feature vector obtained by fusing the feature vector of current signal with the feature vectors of vibration signal and sound signal into the neural network, and the output value obtained is the fused feature vector.
5. The circuit breaker fault diagnosis method as described in claim 1, characterized in that, In step S7, the fused feature vector is compared with the fused feature vector in the circuit breaker operation status database. The distance used is the Minkowski distance, and the p value in the Minkowski distance is 7.
6. A circuit breaker operation status diagnostic system, comprising a measurement module and a data processing unit, characterized in that, The data processing unit includes: The power spectrum calculation module is used to calculate the power spectrum of vibration signals, sound signals, and current signals. Sparse autoencoder network is used to extract feature vectors of the power spectrum of vibration signals, sound signals, and current signals; The first fusion module is used to fuse the feature vectors of vibration signals and sound signals. The fusion method is parallel fusion, the algorithm used is MKCCA, and the kernel function is Gaussian radial basis function. The second fusion module is used to fuse the feature vector obtained by fusing the vibration signal and the sound signal with the feature vector of the current signal to obtain a fused feature vector; A circuit breaker operating status database is used to store the fused feature vectors of circuit breakers under different operating states; The distance calculation module is used to calculate the distance between different fused feature vectors; The data comparison module is used to compare the distance values between different fused feature vectors.
7. The circuit breaker operation status diagnostic system according to claim 6, characterized in that, The measurement module includes a circuit breaker, a vibration sensor, a sound sensor, and a current sensor. The vibration sensor is used to collect the vibration signal of the circuit breaker, the sound sensor is used to collect the sound signal of the circuit breaker, and the current sensor is used to collect the current signal of the circuit breaker.
8. A readable computer storage medium storing a computer program, characterized in that, When the program is executed by the processor, it implements the method as described in any one of claims 1-5.