Fault arc data processing method, electronic equipment and storage medium
By acquiring multi-channel current signals, oversampling, and fixed-length sliding storage, combined with zero-crossing voltage signals and wavelet transform, the resolution and real-time issues of fault arc detection in existing technologies have been solved, achieving high-accuracy and fast-response fault arc identification.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-28
- Publication Date
- 2026-03-10
AI Technical Summary
Existing fault arc detection methods suffer from insufficient current sampling resolution, limited computational power, and poor real-time processing capabilities.
Current signals are acquired through multiple channels, oversampling and fixed-length sliding storage are performed, and feature extraction and wavelet transform are performed by combining the zero-crossing signals of the voltage signals. The existence of fault arcs is then determined using a classifier.
This improves the accuracy and response speed of fault arc detection, and ensures the stability and real-time performance of the calculation results.
Smart Images
Figure CN121637040A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of electrical engineering signal detection technology, and more specifically, to a fault arc data processing method, electronic device, and storage medium. Background Technology
[0002] A fault arc refers to the gas ionization discharge phenomenon caused by air breakdown due to factors such as insulation aging and damage, loose electrical connections, humid air, or a sharp increase in voltage and current in electrical circuits or equipment. Fault arcs can easily pose a fire hazard to the surrounding environment; therefore, their detection is crucial. Detection of fault arcs requires analyzing and processing the current signal in the circuit to identify the fault arc.
[0003] In existing technologies, fault arcs can be identified by acquiring a single-channel current signal and using wavelet transform to analyze the current signal in the circuit.
[0004] However, existing methods for acquiring and processing current signals suffer from problems such as insufficient current sampling resolution, limited computing power, and poor real-time processing. Summary of the Invention
[0005] The purpose of this application is to address the shortcomings of the prior art by providing a fault arc data processing method, electronic device, and storage medium to improve the accuracy of fault arc data processing.
[0006] To achieve the above objectives, the technical solutions adopted in the embodiments of this application are as follows:
[0007] In a first aspect, embodiments of this application provide a fault arc data processing method applied to a microcontroller, the method comprising:
[0008] The current signals in the circuit to be analyzed are collected by multiple channels to obtain multiple current sampling values corresponding to each current signal. The acquisition time of each channel is different.
[0009] The target current sampling value of each current signal is obtained based on multiple current sampling values corresponding to each current signal.
[0010] The target current sampling values are stored in a fixed-length sliding storage method, and the current data to be analyzed is obtained based on the stored target current sampling values and the zero-crossing signal in the voltage signal of the circuit to be analyzed.
[0011] Feature extraction is performed on the current data to be analyzed to obtain multiple feature quantities, and the presence of fault arc is determined based on the multiple feature quantities.
[0012] Optionally, obtaining the target current sample value of each current signal based on multiple current sample values corresponding to each current signal includes:
[0013] Each of the multiple current sample values is oversampled to obtain oversampled current sample values.
[0014] The average value of each oversampled current sample is taken as the target current sample value of the current signal.
[0015] Optionally, the step of oversampling each of the plurality of current sample values to obtain oversampled current sample values includes:
[0016] The current sample value is sampled using a preset maximum sampling rate to obtain a new current sample value;
[0017] The new current sample value is processed by signal-to-noise ratio and decimation to obtain the oversampled current sample value.
[0018] Optionally, storing each of the target current sample values in a fixed-length sliding storage manner, and obtaining the current data to be analyzed based on the stored target current sample values and the zero-crossing signal in the voltage signal of the circuit to be analyzed, includes:
[0019] The data length of the current data to be analyzed is determined based on the sampling rate and sampling period of the current signal.
[0020] Each of the target current sample values is slidably stored into an array of the specified data length;
[0021] The voltage signal in the circuit to be analyzed is monitored in real time. If a zero-crossing signal is detected in the voltage signal, the target current sample value currently stored in the array is used as the current data to be analyzed.
[0022] Optionally, determining the data length of the current data to be analyzed based on the sampling rate and sampling period of the current signal includes:
[0023] The product of the sampling rate and the sampling period is used as the sampling point;
[0024] The data length is obtained by dividing the sampling point by the preset number of cycles of the processed signal.
[0025] Optionally, the step of extracting features from the current data to be analyzed to obtain multiple feature quantities, and determining whether a fault arc exists based on the multiple feature quantities, includes:
[0026] The current data to be analyzed is subjected to wavelet transform using a four-term wavelet to obtain multiple wavelet transform coefficients. Each wavelet transform coefficient is used to indicate the corresponding high-frequency signal component in the current data to be analyzed.
[0027] Based on the wavelet transform coefficients, multiple feature quantities corresponding to the current data to be analyzed are determined. These multiple feature quantities include: kurtosis feature quantity, energy feature quantity, energy entropy feature quantity, mean feature quantity, variance feature quantity, skewness feature quantity, and standard deviation feature quantity.
[0028] The presence or absence of a fault arc is determined based on the kurtosis feature, the energy feature, the energy entropy feature, the mean feature, the variance feature, the skewness feature, and the standard deviation feature.
[0029] Optionally, determining whether a fault arc exists based on the kurtosis feature, the energy feature, the energy entropy feature, the mean feature, the variance feature, the skewness feature, and the standard deviation feature includes:
[0030] Threshold discrimination is performed on the kurtosis feature, the energy feature, the energy entropy feature, the mean feature, the variance feature, the skewness feature, and the standard deviation feature to obtain a first discrimination result, which includes the presence of a fault arc and the absence of a fault arc.
[0031] The pre-trained classifier is used to classify and discriminate the kurtosis feature, the energy feature, the energy entropy feature, the mean feature, the variance feature, the skewness feature, and the standard deviation feature to obtain a second discrimination result, which includes whether a fault arc exists or not.
[0032] Based on the first and second discrimination results, it is determined whether a fault arc exists.
[0033] Optionally, the step of performing threshold discrimination on the kurtosis feature, the energy feature, the energy entropy feature, the mean feature, the variance feature, the skewness feature, and the standard deviation feature to obtain a first discrimination result includes:
[0034] If the kurtosis feature is greater than the kurtosis threshold, the energy feature is greater than the energy threshold, the energy entropy feature is greater than the energy entropy threshold, the mean feature is greater than the mean threshold, the variance feature is greater than the variance threshold, the skewness feature is greater than the skewness threshold, and the standard deviation feature is greater than the standard deviation threshold, then the first discrimination result is determined to be the presence of a faulty arc; otherwise, the first discrimination result is determined to be the absence of a faulty arc.
[0035] Secondly, embodiments of this application also provide an electronic device, including: a processor, a storage medium, and a bus. The storage medium stores program instructions executable by the processor. When the application runs, the processor communicates with the storage medium via the bus, and the processor executes the program instructions to perform the steps of the fault arc data processing method described in the first aspect.
[0036] Fourthly, embodiments of this application also provide a computer-readable storage medium storing a computer program, which is read and executes the steps of the fault arc data processing method described in the first aspect.
[0037] The beneficial effects of this application are:
[0038] This application provides a fault arc data processing method, electronic device, and storage medium. It acquires current signals from the circuit under analysis through multiple channels, obtaining multiple current sample values corresponding to each current signal. Based on these multiple current sample values, it obtains the target current sample value for each current signal. Compared to existing technologies that only acquire a single current sample value for analysis, this method achieves higher sampling resolution for the target current sample value. Simultaneously, it stores each target current sample value using a fixed-length sliding storage method. Based on the stored target current sample values and the zero-crossing signal in the voltage signal of the circuit under analysis, it obtains the current data to be analyzed. This ensures the consistency and continuity of the current data in phase for each calculation and analysis, guaranteeing the stability of the calculation results and improving the real-time performance of data analysis as well as the accuracy and response speed of fault arc detection. Furthermore, it extracts features from the current data to obtain multiple feature quantities, and determines the presence of a fault arc based on these feature quantities, ensuring the accuracy of fault arc identification. Attached Figure Description
[0039] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.
[0040] Figure 1 This is a schematic diagram of the structure of a fault arc detection device provided in an embodiment of this application;
[0041] Figure 2 A flowchart illustrating the first fault arc data processing method provided in this application embodiment;
[0042] Figure 3 A flowchart illustrating the second fault arc data processing method provided in this application embodiment;
[0043] Figure 4 A flowchart illustrating the third fault arc data processing method provided in this application embodiment;
[0044] Figure 5 A flowchart illustrating the fourth fault arc data processing method provided in this application embodiment;
[0045] Figure 6 A flowchart illustrating the fifth fault arc data processing method provided in this application embodiment;
[0046] Figure 7 A flowchart illustrating the sixth fault arc data processing method provided in this application embodiment;
[0047] Figure 8 This is a structural block diagram of an electronic device provided in an embodiment of this application. Detailed Implementation
[0048] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. It should be understood that the accompanying drawings in this application are for illustrative and descriptive purposes only and are not intended to limit the scope of protection of this application. Furthermore, it should be understood that the schematic drawings are not drawn to scale. The flowcharts used in this application illustrate operations implemented according to some embodiments of this application. It should be understood that the operations in the flowcharts may not be implemented in sequence, and steps without logical contextual relationships may be reversed or implemented simultaneously. In addition, those skilled in the art, guided by the content of this application, may add one or more other operations to the flowcharts, or remove one or more operations from the flowcharts.
[0049] Furthermore, the described embodiments are merely some, not all, of the embodiments of this application. The components of the embodiments of this application described and illustrated herein can typically be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of the application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.
[0050] It should be noted that the term "comprising" will be used in the embodiments of this application to indicate the presence of the features declared thereafter, but does not exclude the addition of other features.
[0051] Optionally, the fault arc data processing method provided in this application embodiment is applied to a microcontroller (MCU). The microcontroller (MCU) can be a software platform in an electronic device, such as a mobile phone, tablet computer, laptop computer, PDA, desktop computer, or other terminal device with computing and display capabilities, or it can be a server. Specifically, it can be applied to applications in terminal devices, such as mobile phone apps (APPs) or computer application systems.
[0052] The following section will explain in detail the specific implementation process of fault arc data processing provided in the embodiments of this application.
[0053] Figure 1 This is a schematic diagram of the structure of a fault arc detection device provided in an embodiment of this application, as shown below. Figure 1 As shown, the fault arc detection device may include: a voltage signal processing circuit, a current signal processing circuit, a communication module, and the aforementioned microcontroller. The microcontroller can be communicatively connected to both the voltage signal processing circuit and the current signal processing circuit. The voltage signal processing circuit can detect the voltage signal input to the main circuit and input the detected voltage signal to the microcontroller. The current signal processing circuit can receive the current signal detected by the current sensor in the main circuit and send the received current signal to the microcontroller. Based on the received voltage and current signals, the microcontroller uses the fault arc processing method provided in this embodiment to determine whether a fault arc exists in the main circuit.
[0054] Meanwhile, the microcontroller can also communicate with the host computer via a communication module, allowing the host computer to update the classifier training parameters.
[0055] Figure 2 This is a flowchart illustrating a first method for processing fault arc data provided in an embodiment of this application. The execution entity of this method is the microcontroller described above. Figure 2 As shown, the method includes:
[0056] S101. Collect each current signal in the circuit to be analyzed through multiple channels to obtain multiple current sampling values corresponding to each current signal.
[0057] Here, "multiple channels" refers to multiple channels in the microcontroller, such as channel 1, channel 2, and channel 3. Each channel can acquire the same current signal using a time difference. Since the current signal is a time-domain waveform signal, different channels can acquire the waveform signal at different times, thus obtaining multiple current sampling values corresponding to the current signal.
[0058] For example, channel 1 can acquire the current signal 1 at time t1, channel 2 can acquire the current signal 1 at time t2, and channel 3 can acquire the current signal 1 at time t3, thus obtaining three sampled values corresponding to the current signal 1, such as 207, 211, and 213 respectively.
[0059] Optionally, other current signals can be sampled in the same way as current signal 1 mentioned above, so that multiple current sample values corresponding to each current signal can be obtained.
[0060] S102. Obtain the target current sampling value of each current signal based on the multiple current sampling values corresponding to each current signal.
[0061] Optionally, the target current sampling value of each current signal can be obtained by using a preset method based on multiple current sampling values corresponding to each current signal. Then, the multiple current sampling values corresponding to each current signal are merged, that is, the current sampling values collected from multiple channels are merged to obtain the target current sampling value of each current signal. The obtained target current sampling value is a current sampling value.
[0062] For example, for the aforementioned current signal 1, the target current sampling value of the current signal 1 can be determined by using a preset method based on the three sampling values 207, 211 and 213 corresponding to the obtained current signal 1.
[0063] S103. Store each target current sample value in a fixed-length sliding storage mode, and obtain the current data to be analyzed based on the stored target current sample values and the zero-crossing signal in the voltage signal of the circuit to be analyzed.
[0064] Here, each target current sample value refers to the target current sample value of each current signal. The microprocessor uses the acquisition method in S101 to acquire different current signals in real time, and uses the method in S102 mentioned above to obtain the target current sample value of each current signal. The obtained target current sample value of each current signal is then stored in a fixed-length sliding storage method. The fixed-length sliding storage method refers to sliding the target current sample value into an array of a certain data length.
[0065] Optionally, the current data to be analyzed refers to data used to analyze whether a fault arc exists in the main circuit. Specifically, the current data to be analyzed can be determined using a preset method based on the stored target current sampling value and the zero-crossing signal in the voltage signal of the circuit to be analyzed. The current data to be analyzed may include at least one target current sampling value, and the current data to be analyzed is the current waveform data to be analyzed.
[0066] S104. Perform feature extraction on the current data to be analyzed to obtain multiple feature quantities, and determine whether there is a fault arc based on the multiple feature quantities.
[0067] Optionally, a preset method can be used to extract features from the current data to be analyzed. This extraction can yield multiple feature quantities, and the preset method can be used to determine whether a fault arc exists in the main circuit based on these feature quantities. These multiple feature quantities may include, for example, kurtosis, energy, energy entropy, mean, variance, skewness, and standard deviation.
[0068] In this embodiment, multiple channels are used to collect current signals from the circuit under analysis, resulting in multiple current sample values corresponding to each current signal. Based on these multiple current sample values, the target current sample value for each current signal is obtained. Compared to existing technologies that only collect a single current sample value as analysis data, this method achieves higher sampling resolution for the target current sample value. Simultaneously, each target current sample value is stored using a fixed-length sliding storage method. Based on the stored target current sample values and the zero-crossing signal in the voltage signal of the circuit under analysis, the current data to be analyzed is obtained. This ensures the consistency and continuity of the current data in phase for each calculation and analysis, guaranteeing the stability of the calculation results. This improves the real-time performance of data analysis and the accuracy and response speed of fault arc detection. Feature extraction is performed on the current data to be analyzed to obtain multiple feature quantities. Based on these feature quantities, the presence of a fault arc is determined, ensuring the accuracy of fault arc identification.
[0069] Figure 3 A flowchart illustrating the second fault arc data processing method provided in this application embodiment is shown below. Figure 3 As shown, the process of obtaining the target current sample value for each current signal in S102 based on multiple current sample values corresponding to each current signal may include:
[0070] S201. Perform oversampling processing on each current sample value among multiple current sample values to obtain oversampled current sample values.
[0071] Optionally, each current sample value of a current signal can be oversampled to obtain oversampled current sample values. By oversampling each current sample value, the sampling resolution can be improved, resulting in higher sampling resolution of the obtained current sample values.
[0072] For example, the three sampled values 207, 211 and 213 corresponding to the current signal 1 in the above-mentioned case are oversampled to obtain three oversampled sampled values 210, 213 and 214 respectively.
[0073] S202. The average value of each current sampled value after oversampling is taken as the target current sampled value of the current signal.
[0074] For example, the average of the three sampled values 210, 213 and 214 after sampling is calculated to obtain 212.333, then the target current sampled value of current signal 1 is 212.333.
[0075] In this embodiment, by first oversampling each current sample value, and then using the average value of the oversampled current sample values as the target current sample value, the resolution of the obtained target current sample value can be significantly improved compared with the sample value of a single channel.
[0076] Figure 4 A flowchart illustrating the third fault arc data processing method provided in this application embodiment is shown below. Figure 4 As shown, the process of oversampling each current sample value among multiple current sample values in S201 above to obtain oversampled current sample values may include:
[0077] S301. Sample the current sample value using the preset highest sampling rate to obtain a new current sample value.
[0078] Optionally, if the preset maximum sampling rate is higher than the sampling rate of the current sample value, then the higher sampling rate is used to resample the current sample value to obtain a new current sample value. The sampling rate of the new current sample value is higher than the sampling rate of the previous current sample value. For example, if the sampling rate of the current sample value acquired through the channel is 1 sampling point per second, and the current sample value is resampled using the preset maximum sampling rate, the sampling rate of the new current sample value is 100 sampling points per second.
[0079] S302. Perform signal-to-noise ratio processing and decimation processing on the new current sample value to obtain the sampled current sample value.
[0080] Signal-to-noise ratio (SNR) processing utilizes digital filters to improve the SNR of the current sample values. Furthermore, decimation processing extracts the current sample values after sampling; for example, it can extract the current sample values from the aforementioned 100 points. Thus, by improving the SNR of the current sample values through SNR processing, every 3dB increase in SNR is equivalent to a 0.5-bit increase in the resolution of the analog-to-digital converter, thereby improving measurement accuracy without replacing hardware. Decimation processing reduces the sampling rate while retaining crucial information from the current signal.
[0081] Optionally, the aforementioned steps S301 to S302 are oversampling processes for microprocessors that do not have oversampling capabilities. For microprocessors with oversampling capabilities, oversampling can be implemented by configuring relevant registers.
[0082] Figure 5 A flowchart illustrating the fourth fault arc data processing method provided in this application embodiment is shown below. Figure 5 As shown, in S103, the sampled values of each target current are stored in a fixed-length sliding storage method. Based on the stored sampled values of the target current and the zero-crossing signal in the voltage signal of the circuit to be analyzed, the current data to be analyzed is obtained, which may include:
[0083] S401. Determine the data length of the current data to be analyzed based on the sampling rate and sampling period of the current signal.
[0084] The sampling rate of the current signal refers to the sampling rate at which the microprocessor acquires the current signal through each channel, and the sampling rate is the same for all channels. For example, the sampling rate is 25.6 kHz.
[0085] S402. Slide the sampled values of each target current into an array of data length.
[0086] The length of the array is the data length calculated in S401 above. For example, if it is 256, then the array is an array with a data length of 256.
[0087] Optionally, the array can continuously store each target current sample value in real time. For example, it can slide to store each target current sample value from right to left. When the array is full and a new target current sample value needs to be stored, it performs an operation to shift one position to the left, removes the earliest target current sample value stored in the array, and adds a new target current sample value to the rightmost position of the array, thereby realizing continuous sliding storage of data.
[0088] S403. Monitor the voltage signal in the circuit to be analyzed in real time to determine whether there is a zero-crossing signal in the voltage signal.
[0089] If a zero-crossing signal exists in the voltage signal, execute S404 below; if no zero-crossing signal exists in the voltage signal, execute S405 below.
[0090] Specifically, the voltage signal input to the circuit to be analyzed can be monitored in real time by a voltage signal processing circuit. When the voltage signal is received, the microprocessor can perform zero-crossing detection on the voltage signal to determine whether there is a zero-crossing signal in the voltage signal.
[0091] When a zero-crossing signal appears, it indicates that the starting phases of the upper and lower half-cycle waveforms of the voltage signal are the same. When a zero-crossing signal appears in the voltage signal, the process returns to step S402 and continues to slide and store the obtained target current sample value into the array.
[0092] S404. Use the target current sample value currently stored in the array as the current data to be analyzed.
[0093] Optionally, when the voltage signal shows a zero-crossing signal, the target current sample value currently stored in the array can be used as the current data to be analyzed.
[0094] It is worth noting that when a zero-point signal occurs, the target current sample value in the array may not be fully stored. In this case, the target current sample value currently stored in the array is not used as the current data to be analyzed, that is, no data analysis processing is performed.
[0095] In this embodiment, by sliding the target current sampling values into an array of data length, continuous sliding storage of data can be achieved. When a zero-crossing signal occurs, the target current sampling values stored in the array are used as the data to be analyzed. This ensures that the current data analyzed each time are consistent in phase, which helps to reduce the fluctuation range of the analysis results during analysis and processing, making the judgment of the results more accurate and improving the accuracy of fault arc identification.
[0096] Optionally, determining the data length of the current data to be analyzed based on the sampling rate and sampling period of the current signal in S401 above may include:
[0097] Optionally, the product of the sampling rate and the sampling period can be used as the sampling point. For example, if the sampling rate is 25.6 kHz and the sampling period is 20 s, then the sampling point is 512. The data length is then obtained by dividing the sampling point by a preset number of cycles of the processed signal. This preset number of cycles of the processed signal can be, for example, 2. For instance, the value of 512 sampling points divided by 2 is 256, so the data length is 256.
[0098] Figure 6 A flowchart illustrating the fifth fault arc data processing method provided in this application embodiment is shown below. Figure 6 As shown, in step S104 above, feature extraction of the current data to be analyzed yields multiple feature quantities, and the determination of whether a fault arc exists based on these multiple feature quantities may include:
[0099] S501. Perform wavelet transform on the current data to be analyzed using a four-term wavelet to obtain multiple wavelet transform coefficients.
[0100] Among them, each wavelet transform coefficient can be used to indicate the corresponding high-frequency signal component in the current data to be analyzed.
[0101] The four-term wavelet is a type of wavelet basis function (Daubechies). The DB4 wavelet function has four non-zero rectangular waveforms, which can provide good time and frequency resolution. Traditional DB4 wavelet transform requires four levels of signal decomposition. In order to adapt to microprocessors with low computing power, this embodiment selects the first level of decomposition (D1) to analyze the current data to be analyzed. After signal analysis, multiple wavelet transform coefficients can be obtained.
[0102] S502. Based on the wavelet transform coefficients, determine the multiple characteristic quantities corresponding to the current data to be analyzed.
[0103] Among them, multiple features may include kurtosis features, energy features, energy entropy features, mean features, variance features, skewness features, and standard deviation features.
[0104] Specifically, the kurtosis feature can be obtained by the following formula (I).
[0105]
[0106] Where xi is the i-th wavelet transform coefficient. σ is the mean of the wavelet transform coefficients, N is the wavelet transform length, which is the total number of wavelet transform coefficients, and σ is the standard deviation of the wavelet transform coefficients.
[0107] Optionally, kurtosis can be used to characterize the non-Gaussianity of a signal. When K = 3, the signal distribution is Gaussian; when K > 3, the signal distribution is more kurtotic than a Gaussian distribution; when K < 3, the signal distribution is flatter than a Gaussian distribution. The larger the absolute value of K-3, the greater the difference from a Gaussian distribution. In this embodiment, the kurtosis of the wavelet coefficients of the loop current signal is calculated, and the difference in the magnitude of the kurtosis of the wavelet coefficients before and after the occurrence of a fault arc is used to determine whether a fault arc exists in the loop.
[0108] Optionally, the energy characteristic is obtained by calculating the sum of squares of the coefficients of each frequency band after wavelet transform, which helps in signal classification and anomaly detection.
[0109] Optionally, the energy entropy characteristic is calculated using the following formula (ii).
[0110]
[0111] Where H(x) is the energy entropy characteristic quantity, P(x) i) represents the probability of the i-th event in the current signal, which is the proportion of the energy of each wavelet coefficient in the current signal to the total energy; N is the total number of wavelet transform coefficients, and b is the base when calculating entropy, usually taking the value of e or 2.
[0112] Optionally, the mean characteristic is obtained by summing the wavelet transform coefficients and dividing the sum by the total number of wavelet transform coefficients.
[0113] Optionally, the variance characteristic can be obtained by the following formula (iii), which can reflect the degree of dispersion between current data.
[0114]
[0115] Where F1 is the variance eigenvalue, and xi is the i-th wavelet transform coefficient. Let be the mean of the wavelet transform coefficients, and N be the total number of wavelet transform coefficients.
[0116] Alternatively, the standard deviation characteristic can be obtained by the following formula (iv).
[0117]
[0118] Where F2 is the standard deviation characteristic, and xi is the i-th wavelet transform coefficient. Let be the mean of the wavelet transform coefficients, and N be the total number of wavelet transform coefficients.
[0119] Alternatively, the skewness characteristic can be obtained by the following formula (V).
[0120]
[0121] Where F3 is the skewness characteristic, xi is the i-th wavelet transform coefficient, and N is the total number of wavelet transform coefficients.
[0122] S503. Determine whether a fault arc exists based on kurtosis, energy, energy entropy, mean, variance, skewness, and standard deviation characteristics.
[0123] Alternatively, the presence of a fault arc in the circuit can be determined based on multiple characteristic quantities.
[0124] In this embodiment, using multiple feature quantities to identify fault arcs can improve the accuracy of fault arc identification.
[0125] Figure 7 A flowchart illustrating the sixth fault arc data processing method provided in this application embodiment is shown below. Figure 7As shown, S503 determines the presence of a fault arc based on kurtosis, energy, energy entropy, mean, variance, skewness, and standard deviation characteristics, which may include:
[0126] S601. Threshold discrimination is performed on the kurtosis feature, energy feature, energy entropy feature, mean feature, variance feature, skewness feature, and standard deviation feature respectively to obtain the first discrimination result.
[0127] The first discrimination result includes whether a fault arc exists or not.
[0128] S602. Using a pre-trained classifier, classify and discriminate kurtosis features, energy features, energy entropy features, mean features, variance features, skewness features, and standard deviation features to obtain a second discrimination result.
[0129] The second discrimination result includes whether a fault arc exists or not.
[0130] Alternatively, a Naive Bayes classifier can be used for feature discrimination. This Naive Bayes classifier, based on Bayes' theorem, achieves both simplicity and computational efficiency through a simplified feature independence assumption. Other classifiers such as neural networks, random forests, and logistic regression can also be chosen.
[0131] Alternatively, when training the classifier, the model data can be trained locally on the microprocessor, and the classifier training parameters can be updated by a host computer that communicates with the microprocessor.
[0132] Optionally, when the classifier combines kurtosis features, energy features, energy entropy features, mean features, variance features, skewness features, and standard deviation features for classification, a second discrimination result can be obtained.
[0133] S603. Based on the first and second discrimination results, determine whether a fault arc exists.
[0134] Optionally, if the first judgment result is that a fault current exists and the second judgment result is also that a fault current exists, then it is determined that a fault current exists in the circuit; otherwise, no fault current exists.
[0135] In this embodiment, the presence of a fault arc is determined by combining threshold discrimination of feature quantities and classifier discrimination, which can improve the accuracy of fault arc identification.
[0136] Optionally, in step S601 above, threshold discrimination is performed on the kurtosis feature, energy feature, energy entropy feature, mean feature, variance feature, skewness feature, and standard deviation feature to obtain a first discrimination result, which may include:
[0137] Optionally, if the kurtosis feature is greater than the kurtosis threshold, and the energy feature is greater than the energy threshold, and the energy entropy feature is greater than the energy entropy threshold, and the mean feature is greater than the mean threshold, and the variance feature is greater than the variance threshold, and the skewness feature is greater than the skewness threshold, and the standard deviation feature is greater than the standard deviation threshold, then the first discrimination result is determined to be that a faulty arc exists; otherwise, the first discrimination result is determined to be that a faulty arc does not exist.
[0138] Figure 8 This is a structural block diagram of an electronic device provided in an embodiment of this application. This electronic device can be, for example, the fault arc detection device described in the foregoing embodiments. Figure 8 As shown, the electronic device may include: a processor 701 and a memory 702.
[0139] Optionally, a bus 703 may also be included, wherein the memory 702 is used to store machine-readable instructions executable by the processor 701. When the electronic device 700 is running, the processor 701 and the memory 702 communicate via the bus 703. When the machine-readable instructions are executed by the processor 701, the method steps in the above method embodiments are performed.
[0140] This application also provides a computer-readable storage medium storing a computer program, which, when run by a processor, executes the method steps described in the above-described fault arc data processing method embodiments.
[0141] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems and devices described above can be referred to the corresponding processes in the method embodiments, and will not be repeated here. In the several embodiments provided in this application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. Furthermore, multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed mutual coupling or direct coupling or communication connection can be through some communication interfaces; the indirect coupling or communication connection of devices or modules can be electrical, mechanical, or other forms.
[0142] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. If the functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes: USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, optical disks, and other media capable of storing program code.
[0143] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.
Claims
1. A method of processing fault arc data, characterized by, The method applied to a microcontroller comprises: Collecting each current signal in a to-be-analyzed loop through multiple channels respectively to obtain multiple current sampling values corresponding to each current signal, wherein the collection time of each channel is different; Obtaining a target current sampling value of each current signal according to the multiple current sampling values corresponding to each current signal; Storing each target current sampling value in a fixed-length sliding storage manner, and obtaining to-be-analyzed current data according to the stored target current sampling value and a zero-crossing point signal in a voltage signal in the to-be-analyzed loop; Performing feature extraction on the to-be-analyzed current data to obtain multiple feature quantities, and determining whether there is a fault arc according to the multiple feature quantities.
2. The fault arc data processing method of claim 1, wherein, The method of obtaining a target current sampling value of each current signal according to the multiple current sampling values corresponding to each current signal comprises: Performing oversampling processing on each current sampling value in the multiple current sampling values to obtain each current sampling value after oversampling; Taking the average value of each current sampling value after oversampling as the target current sampling value of the current signal.
3. The fault arc data processing method of claim 2, wherein, The method of performing oversampling processing on each current sampling value in the multiple current sampling values to obtain each current sampling value after oversampling comprises: Sampling each current sampling value using a preset highest sampling rate to obtain a new current sampling value; Performing signal-to-noise ratio processing and extraction processing on the new current sampling value to obtain each current sampling value after oversampling.
4. The fault arc data processing method of claim 1, wherein, The method of storing each target current sampling value in a fixed-length sliding storage manner, and obtaining to-be-analyzed current data according to the stored target current sampling value and a zero-crossing point signal in a voltage signal in the to-be-analyzed loop comprises: Determining the data length of the to-be-analyzed current data storage according to the sampling rate and sampling period of the current signal; Sliding storing each target current sampling value into an array of the data length; Real-time monitoring the voltage signal in the to-be-analyzed loop, and if a zero-crossing point signal is monitored in the voltage signal, taking the target current sampling value currently stored in the array as the to-be-analyzed current data.
5. The fault arc data processing method of claim 4, wherein, The method of determining the data length of the to-be-analyzed current data storage according to the sampling rate and sampling period of the current signal comprises: Taking the product of the sampling rate and the sampling period as a sampling point; Taking the value of the sampling point divided by the period number of a preset processing signal as the data length.
6. The fault arc data processing method of claim 1, wherein, The method of performing feature extraction on the to-be-analyzed current data to obtain multiple feature quantities, and determining whether there is a fault arc according to the multiple feature quantities comprises: Performing wavelet transform on the to-be-analyzed current data using a quartic wavelet to obtain multiple wavelet transform coefficients, each wavelet transform coefficient being used to indicate a corresponding high-frequency signal component in the to-be-analyzed current data; Determining multiple feature quantities corresponding to the to-be-analyzed current data according to each wavelet transform coefficient, the multiple feature quantities comprising a kurtosis feature quantity, an energy feature quantity, an energy entropy feature quantity, a mean value feature quantity, a variance feature quantity, a skewness feature quantity, and a standard deviation feature quantity. Determine whether there is a fault arc according to the kurtosis feature quantity, the energy feature quantity, the energy entropy feature quantity, the mean value feature quantity, the variance feature quantity, the skewness feature quantity and the standard deviation feature quantity.
7. The fault arc data processing method of claim 6, wherein, The step of determining whether there is a fault arc according to the kurtosis feature quantity, the energy feature quantity, the energy entropy feature quantity, the mean value feature quantity, the variance feature quantity, the skewness feature quantity and the standard deviation feature quantity comprises: Respectively perform threshold discrimination on the kurtosis feature quantity, the energy feature quantity, the energy entropy feature quantity, the mean value feature quantity, the variance feature quantity, the skewness feature quantity and the standard deviation feature quantity to obtain a first discrimination result, wherein the first discrimination result comprises existence of a fault arc and non-existence of a fault arc; Perform classification discrimination on the kurtosis feature quantity, the energy feature quantity, the energy entropy feature quantity, the mean value feature quantity, the variance feature quantity, the skewness feature quantity and the standard deviation feature quantity by using a pre-trained classifier to obtain a second discrimination result, wherein the second discrimination result comprises existence of a fault arc and non-existence of a fault arc; Determine whether there is a fault arc according to the first discrimination result and the second discrimination result.
8. The fault arc data processing method of claim 7, wherein, The step of respectively performing threshold discrimination on the kurtosis feature quantity, the energy feature quantity, the energy entropy feature quantity, the mean value feature quantity, the variance feature quantity, the skewness feature quantity and the standard deviation feature quantity to obtain a first discrimination result comprises: If the kurtosis feature quantity is greater than a kurtosis threshold, the energy feature quantity is greater than an energy threshold, the energy entropy feature quantity is greater than an energy entropy threshold, the mean value feature quantity is greater than a mean value threshold, the variance feature quantity is greater than a variance threshold, the skewness feature quantity is greater than a skewness threshold, and the standard deviation feature quantity is greater than a standard deviation threshold, then determine that the first discrimination result is existence of a fault arc; otherwise, determine that the first discrimination result is non-existence of a fault arc.
9. An electronic device, comprising: The computer readable storage medium stores a computer program, and the computer program is run by a processor to execute the steps of the fault arc data processing method according to any one of claims 1-8.
10. A computer-readable storage medium, characterized in that, The computer readable storage medium stores a computer program, and the computer program is run by a processor to execute the steps of the fault arc data processing method according to any one of claims 1-8.