Method and device for detecting arc fault in electrical loop, and computer program product

By building and training a fault detection model based on backpropagation neural network, the problems of low arc fault detection efficiency and difficulty in positioning in low voltage systems are solved, and efficient and accurate arc fault detection and positioning are achieved.

CN118965991APending Publication Date: 2024-11-15GUANGDONG POWER GRID CO LTD +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411052164.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-01
Publication Date
2024-11-15

Smart Images

  • Figure CN118965991A_ABST
    Figure CN118965991A_ABST
Patent Text Reader

Abstract

The invention discloses a method and a device for detecting an arc fault in an electrical loop, and a computer program product. The method comprises the steps that a simulation test data set of a simulation electrical loop is collected, the simulation test data set comprises first type data and second type data, the first type data is simulation test data when an arc fault occurs in the simulation electrical loop, and the second type data is simulation test data when the arc fault occurs in the simulation electrical loop; the second type of data is simulation test data in which no arc fault occurs in the simulation electrical loop; constructing an initial fault detection model based on a back propagation neural network; training and verifying the initial fault detection model by using the simulation test data set to obtain a target fault detection model; and performing fault detection on the actual electrical loop by using the target fault detection model to obtain a detection result. According to the invention, the technical problems that a series arc fault detection method in a low-voltage system provided by the prior art is low in detection efficiency and the position of a fault arc is difficult to accurately locate are solved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of power system fault detection, and in particular to a method, device and computer program product for detecting arc faults in an electrical circuit. Background Art

[0002] Fault arc is one of the important factors causing power accidents. Usually, problems such as insulation aging, poor contact, chemical corrosion, etc. in the circuit are likely to cause fault arc. Therefore, in order to reduce the frequency of power accidents, reduce the economic losses of electricity users, and ensure personal safety during electricity use, it is necessary to detect whether arc faults occur in electrical circuits.

[0003] At present, the existing technology usually uses circuit breakers or other relay protection devices to perform overcurrent protection, leakage protection, etc. on electrical circuits. However, circuit breakers or other relay protection devices cannot accurately detect the location of arc faults in electrical circuits, resulting in the inability to detect fault arcs in a timely and effective manner.

[0004] From the above analysis, it can be seen that, with respect to the problems that the detection efficiency of the series arc fault detection method in the low-voltage system provided by the above-mentioned prior art is low and it is difficult to accurately locate the position of the fault arc, no effective solution has been proposed yet. Summary of the invention

[0005] The embodiments of the present invention provide a method, device, and computer program product for detecting arc faults in an electrical circuit, so as to at least solve the technical problems that the series arc fault detection method in a low-voltage system provided by the prior art has low detection efficiency and difficulty in accurately locating the position of the fault arc.

[0006] According to one aspect of an embodiment of the present invention, there is provided a method for detecting an arc fault in an electrical circuit, comprising:

[0007] Collect a simulation test data set of a simulated electrical circuit, wherein the simulation test data set includes a first type of data and a second type of data, the first type of data is simulation test data in which an arc fault occurs in the simulated electrical circuit, and the second type of data is simulation test data in which no arc fault occurs in the simulated electrical circuit; construct an initial fault detection model based on a back propagation neural network; use the simulation test data set to train and verify the initial fault detection model to obtain a target fault detection model; use the target fault detection model to perform fault detection on the actual electrical circuit to obtain a detection result, wherein the detection result is at least used to determine whether an arc fault occurs in the actual electrical circuit.

[0008] Optionally, the arc fault detection method in the electrical circuit further includes: constructing an arc simulation model for simulating the electrical circuit; and performing simulation tests on the arc simulation model under different simulation test conditions to obtain a simulation test data set.

[0009] Optionally, different simulation test conditions satisfy one of the following: different types of loads configured in the simulated electrical circuit; different numbers of loads configured in the simulated electrical circuit; and different connection methods of the loads configured in the simulated electrical circuit.

[0010] Optionally, frequency domain characteristics of simulation test data of different types of loads before an arc fault occurs are different from frequency domain characteristics of simulation test data of different types of loads after an arc fault occurs, and the frequency domain characteristics are determined by waveform data of different types of loads.

[0011] Optionally, the initial fault detection model based on the back propagation neural network includes: an input layer, a hidden layer, and an output layer, wherein the input layer is used to transmit the simulation test data to the hidden layer, the hidden layer is used to extract the frequency domain characteristics of the simulation test data, and the output layer is used to determine whether an arc fault occurs in the simulated electrical circuit based on the frequency domain characteristics.

[0012] Optionally, the arc fault detection method in the above electrical circuit also includes: dividing the simulation test data set according to a preset division ratio to obtain a first data set and a second data set, wherein the first data set is used for model training and the second data set is used for model verification.

[0013] Optionally, the initial fault detection model is trained and verified using a simulation test data set to obtain a target fault detection model, including: iteratively training the initial fault detection model using a first data set to obtain a training result, wherein the training result is used to determine the training error of the initial fault detection model; in response to the training error being less than or equal to a preset error value, verifying the initial fault detection model using a second data set to obtain a verification result, wherein the verification result is used to determine the detection accuracy of the initial fault detection model for arc faults; in response to any one detection accuracy being greater than or equal to a target value, determining the initial fault detection model as a target fault detection model.

[0014] Optionally, the above-mentioned arc fault detection method in the electrical circuit also includes: in response to a training error being greater than a preset error value, adjusting the model parameters of the initial fault detection model according to the training error to obtain an updated fault detection model; and using the first data set to repeatedly train and verify the updated fault detection model to determine whether the updated fault detection model is the target fault detection model.

[0015] Optionally, the target fault detection model is used to perform fault detection on the actual electrical circuit to obtain the detection result, including: collecting multiple waveform data to be detected in the actual electrical circuit, wherein different waveform data to be detected correspond to different loads; and using the target fault detection model to classify and identify the multiple waveform data to be detected to obtain the detection result.

[0016] Optionally, the detection result is one of a first detection result and a second detection result, the first detection result includes that an arc fault does not occur in the actual electrical circuit, and the second detection result includes that an arc fault occurs in the actual electrical circuit and a target load that causes the arc fault. The target fault detection model is used to classify and identify multiple waveform data to be detected, and the detection result includes: using the target fault detection model to classify any waveform data to be detected to obtain a classification result, wherein the classification result is used to determine the category of the load corresponding to any waveform data to be detected; comparing any waveform data to be detected with the corresponding normal waveform data to obtain a comparison result, wherein the normal waveform data is used to characterize that no arc fault occurs in the load corresponding to any waveform data to be detected; in response to the comparison result, it is determined that multiple waveform data to be detected are consistent with the corresponding normal waveform data, and a first detection result is generated according to the comparison result; in response to the comparison result, it is determined that at least one waveform data to be detected is inconsistent with the corresponding normal waveform data, and a second detection result is generated according to the comparison result and the classification result, wherein the load corresponding to at least one waveform data to be detected is the target load.

[0017] According to another aspect of an embodiment of the present invention, there is also provided an arc fault detection device in an electrical circuit, comprising:

[0018] An acquisition module is used to acquire a simulation test data set of a simulated electrical circuit, wherein the simulation test data set includes a first type of data and a second type of data, wherein the first type of data is simulation test data of an arc fault occurring in the simulated electrical circuit, and the second type of data is simulation test data of no arc fault occurring in the simulated electrical circuit; a first construction module is used to construct an initial fault detection model based on a back propagation neural network; a training module is used to train and verify the initial fault detection model using the simulation test data set to obtain a target fault detection model; a detection module is used to perform fault detection on an actual electrical circuit using the target fault detection model to obtain a detection result, wherein the detection result is at least used to determine whether an arc fault occurs in the actual electrical circuit.

[0019] Optionally, the arc fault detection device in the above-mentioned electrical circuit also includes: a second construction module, used to construct an arc simulation model of the simulated electrical circuit; under different simulation test conditions, the arc simulation model is simulated and tested to obtain a simulation test data set.

[0020] Optionally, the arc fault detection device in the above-mentioned electrical circuit includes: different simulation test conditions satisfy one of the following: different types of loads configured in the simulated electrical circuit; different numbers of loads configured in the simulated electrical circuit; different connection methods of the loads configured in the simulated electrical circuit.

[0021] Optionally, the arc fault detection device in the above-mentioned electrical circuit includes: the frequency domain characteristics of simulation test data of different types of loads before the arc fault occurs are different from the frequency domain characteristics of simulation test data of different types of loads after the arc fault occurs, and the frequency domain characteristics are determined by the waveform data of different types of loads.

[0022] Optionally, the arc fault detection device in the above-mentioned electrical circuit includes: an initial fault detection model based on a back-propagation neural network includes: an input layer, a hidden layer, and an output layer, the input layer is used to transmit the simulation test data to the hidden layer, the hidden layer is used to extract the frequency domain characteristics of the simulation test data, and the output layer is used to determine whether an arc fault occurs in the simulated electrical circuit based on the frequency domain characteristics.

[0023] Optionally, the arc fault detection device in the above-mentioned electrical circuit also includes: a division module, used to divide the simulation test data set according to a preset division ratio to obtain a first data set and a second data set, wherein the first data set is used for model training and the second data set is used for model verification.

[0024] Optionally, the above-mentioned training module is also used to: use the first data set to iteratively train the initial fault detection model to obtain a training result, wherein the training result is used to determine the training error of the initial fault detection model; in response to the training error being less than or equal to a preset error value, use the second data set to verify the initial fault detection model to obtain a verification result, wherein the verification result is used to determine the detection accuracy of the initial fault detection model for arc faults; in response to any one detection accuracy being greater than or equal to a target value, determine the initial fault detection model as a target fault detection model.

[0025] Optionally, the arc fault detection device in the above-mentioned electrical circuit also includes: an updating module, which is used to adjust the model parameters of the initial fault detection model according to the training error in response to the training error being greater than a preset error value, so as to obtain an updated fault detection model; and use the first data set to repeatedly train and verify the updated fault detection model to determine whether the updated fault detection model is the target fault detection model.

[0026] Optionally, the detection module is also used to: collect multiple waveform data to be detected in an actual electrical circuit, wherein different waveform data to be detected correspond to different loads; and classify and identify the multiple waveform data to be detected using a target fault detection model to obtain a detection result.

[0027] Optionally, the detection result is one of a first detection result and a second detection result, the first detection result includes that an arc fault does not occur in the actual electrical circuit, and the second detection result includes that an arc fault occurs in the actual electrical circuit and a target load that causes the arc fault. The above-mentioned detection module is also used to: classify any waveform data to be detected using a target fault detection model to obtain a classification result, wherein the classification result is used to determine the category of the load corresponding to any waveform data to be detected; compare any waveform data to be detected with the corresponding normal waveform data to obtain a comparison result, wherein the normal waveform data is used to characterize that no arc fault occurs in the load corresponding to any waveform data to be detected; in response to the comparison result, it is determined that multiple waveform data to be detected are consistent with the corresponding normal waveform data, and a first detection result is generated according to the comparison result; in response to the comparison result, it is determined that at least one waveform data to be detected is inconsistent with the corresponding normal waveform data, and a second detection result is generated according to the comparison result and the classification result, wherein the load corresponding to at least one waveform data to be detected is the target load.

[0028] According to another aspect of an embodiment of the present invention, there is further provided a computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the computer program implements any one of the aforementioned methods for detecting arc faults in an electrical circuit.

[0029] According to another aspect of an embodiment of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium including a stored executable program, wherein when the executable program is running, the device where the computer-readable storage medium is located is controlled to execute any one of the aforementioned methods for detecting arc faults in an electrical circuit.

[0030] According to another aspect of an embodiment of the present invention, there is also provided an electronic device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to execute any one of the aforementioned methods for detecting arc faults in an electrical circuit.

[0031] In an embodiment of the present invention, a simulation test data set of a simulated electrical circuit is first collected, wherein the simulation test data set includes a first type of data and a second type of data, the first type of data is simulation test data of arc faults occurring in the simulated electrical circuit, and the second type of data is simulation test data of no arc faults occurring in the simulated electrical circuit, and an initial fault detection model based on a back propagation neural network is constructed, and then the initial fault detection model is trained and verified using the simulation test data set to obtain a target fault detection model, and finally the target fault detection model is used to perform fault detection on the actual electrical circuit to obtain a detection result, wherein the detection result is at least used to determine whether an arc fault occurs in the actual electrical circuit.

[0032] It is easy to understand that the above-mentioned embodiment of the present invention pre-constructs a simulated electrical circuit including an arc simulation model, performs simulation tests on the arc simulation model under the condition that the simulated electrical circuit is configured with different loads, obtains a simulation data set, and then uses the simulation data set to detect the fault detection model based on the back propagation neural network, so as to continuously adjust the model parameters, so as to achieve the purpose of determining the target fault detection model for efficiently and accurately detecting and locating arc faults in the electrical circuit, thereby achieving the technical effect of improving the detection efficiency of series fault arcs in low-voltage systems and accurately locating fault arcs, and further solves the technical problems of low detection efficiency and difficulty in accurately locating the position of fault arcs in the series arc fault detection method in low-voltage systems provided by the prior art. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0034] Figure 1 is a hardware structure block diagram of a mobile terminal for an optional method for detecting arc faults in an electrical circuit according to an embodiment of the present invention;

[0035] Figure 2 is a flow chart of a method for detecting arc faults in an electrical circuit according to an embodiment of the present invention;

[0036] Figure 3 is a schematic diagram of an arc simulation model according to an embodiment of the present invention;

[0037] Figure 4 is a schematic diagram of a simulated electrical circuit including a resistive load according to an embodiment of the present invention;

[0038] Figure 5 is a schematic diagram of a simulated electrical circuit including an inductive load according to an embodiment of the present invention;

[0039] Figure 6 is a schematic diagram of a simulated electrical circuit including a resistive load and an inductive load according to an embodiment of the present invention;

[0040] Figure 7 is a schematic diagram of an arc fault detection process in an electrical circuit according to an embodiment of the present invention;

[0041] Figure 8 is a schematic diagram of a fault detection model training result according to an embodiment of the present invention;

[0042] Fig. 9 is a current waveform diagram of an optional electrical circuit according to an embodiment of the present invention;

[0043] Fig.10 is a current waveform diagram of another optional electrical circuit according to an embodiment of the present invention;

[0044] Fig.11 is a current waveform diagram of another optional electrical circuit according to an embodiment of the present invention;

[0045] Fig.12 is a current waveform diagram of another optional electrical circuit according to an embodiment of the present invention;

[0046] Fig.13 4 is a structural block diagram of an arc fault detection device in an electrical circuit according to an embodiment of the present invention. DETAILED DESCRIPTION

[0047] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings 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 should fall within the scope of protection of the present invention.

[0048] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0049] According to an embodiment of the present invention, a method embodiment of an arc fault detection method in an electrical circuit is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0050] Figure 1 is a hardware structure block diagram of a mobile terminal for an optional method for detecting arc faults in an electrical circuit according to an embodiment of the present invention, such as Figure 1As shown, the mobile terminal (or mobile device) may include one or more processors 102 (the processor 102 may include but is not limited to a processing device such as a microprocessor (Microcontroller Unit, MCU) or a programmable logic device (Field Programmable Gate Array, FPGA)), a memory 104 for storing data, and a transmission device 106 for communication functions. In addition, it may also include: a display device 110, an input and output device 108 (i.e., an IO device), a Universal Serial Bus (USB) port (which may be included as one of the ports of a computer bus, not shown in the figure), a network interface (not shown in the figure), a power supply (not shown in the figure) and / or a camera (not shown in the figure). It can be understood by those skilled in the art that Figure 1 The structure shown is for illustration only and does not limit the structure of the mobile terminal. Figure 1 More or fewer components as shown, or with Figure 1 Different configurations are shown.

[0051] It should be noted that the one or more processors 102 and / or other data processing circuits may be embodied in whole or in part as software, hardware, firmware or any other combination thereof. In addition, the data processing circuit may be a single independent processing module, or may be fully or partially integrated into any of the other components in the mobile terminal (or mobile device).

[0052] The memory 104 can be used to store software programs and modules of application software, such as program instructions / data storage devices corresponding to the arc fault detection method in the electrical circuit in the embodiment of the present invention. The processor 102 executes various functional applications and data processing by running the software programs and modules stored in the memory 104, that is, the arc fault detection method in the electrical circuit described above is realized. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories may be connected to the mobile terminal via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0053] The transmission device 106 is used to receive or send data via a network. The specific example of the above network may include a wireless network provided by a communication provider of the mobile terminal. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 can be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0054] Under the above operating environment, the embodiment of the present invention provides the following Figure 2 The arc fault detection method in the electrical circuit shown in FIG. Figure 2 is a flow chart of a method for detecting arc faults in an electrical circuit according to an embodiment of the present invention. Figure 2 As shown, the method includes the following implementation steps:

[0055] Step S201, collecting a simulation test data set of a simulated electrical circuit, wherein the simulation test data set includes a first type of data and a second type of data, the first type of data is simulation test data of an arc fault occurring in the simulated electrical circuit, and the second type of data is simulation test data of no arc fault occurring in the simulated electrical circuit;

[0056] Step S202, constructing an initial fault detection model based on a back propagation neural network;

[0057] Step S203, using the simulation test data set to train and verify the initial fault detection model to obtain a target fault detection model;

[0058] Step S204: perform fault detection on the actual electrical circuit using the target fault detection model to obtain a detection result, wherein the detection result is at least used to determine whether an arc fault occurs in the actual electrical circuit.

[0059] The above-mentioned simulated electrical circuit can be a simulation model of a series circuit in a low-voltage system of an electric power system. The above-mentioned back propagation neural network refers to a BP neural network (Back Propagation Neural Network), which can minimize the error between the predicted output and the actual output by continuously adjusting the weights and biases in the network.

[0060] In the present invention, by conducting simulation tests on the simulated electrical circuit under different simulation test conditions, a large amount of multiple types of simulation test data can be obtained, and then the simulation test data is used to train and verify the constructed initial fault detection model based on the BP neural network, so that a target fault detection model with optimal model parameters can be obtained. The target fault detection model can then be used to efficiently and accurately locate the position of the fault arc in the electrical circuit, so as to promptly remind the power fault personnel to take fault handling measures and avoid power safety accidents.

[0061] In an embodiment of the present invention, a simulation test data set of a simulated electrical circuit is first collected, wherein the simulation test data set includes a first type of data and a second type of data, the first type of data is simulation test data of arc faults occurring in the simulated electrical circuit, and the second type of data is simulation test data of no arc faults occurring in the simulated electrical circuit, and an initial fault detection model based on a back propagation neural network is constructed, and then the initial fault detection model is trained and verified using the simulation test data set to obtain a target fault detection model, and finally the target fault detection model is used to perform fault detection on the actual electrical circuit to obtain a detection result, wherein the detection result is at least used to determine whether an arc fault occurs in the actual electrical circuit.

[0062] It is easy to understand that the above-mentioned embodiment of the present invention pre-constructs a simulated electrical circuit including an arc simulation model, performs simulation tests on the arc simulation model under the condition that the simulated electrical circuit is configured with different loads, obtains a simulation data set, and then uses the simulation data set to detect the fault detection model based on the back propagation neural network, so as to continuously adjust the model parameters, so as to achieve the purpose of determining the target fault detection model for efficiently and accurately detecting and locating arc faults in the electrical circuit, thereby achieving the technical effect of improving the detection efficiency of series fault arcs in low-voltage systems and accurately locating fault arcs, and further solves the technical problems of low detection efficiency and difficulty in accurately locating the position of fault arcs in the series arc fault detection method in low-voltage systems provided by the prior art.

[0063] The above method of the embodiment of the present invention is further introduced below.

[0064] In an optional embodiment, the arc fault detection method in the electrical circuit further includes:

[0065] Step S251, constructing an arc simulation model for simulating an electrical circuit;

[0066] Step S252: performing simulation tests on the arc simulation model under different simulation test conditions to obtain a simulation test data set.

[0067] It should be noted that different simulation test conditions meet one of the following: different types of loads configured in the simulated electrical circuit; different numbers of loads configured in the simulated electrical circuit; different connection methods of loads configured in the simulated electrical circuit. It should be further noted that the frequency domain characteristics of simulation test data of different types of loads before an arc fault occurs are different from the frequency domain characteristics of simulation test data of different types of loads after an arc fault occurs, and the frequency domain characteristics are determined by the waveform data of different types of loads.

[0068] As an optional embodiment, the arc simulation model used in the present invention may be a Cassie arc model, such as Figure 3 As shown, the Cassie arc model may include: Differential Equation Editor (DEE), Voltage Measurement Unit, Step Signal, Controlled Current Source, and Hit Crossing. It should be noted that the staff can set the parameters of the differential equation editor, arc time constant (such as 0.000003), arc voltage constant (such as 50), initial value of arc conductance (such as 1.e4), and time of circuit breaker disconnection (such as 0.1) according to the requirements of the simulation test.

[0069] It is understandable that the Cassie arc model has a good simulation effect when the electrical circuit contains low resistance, and is more suitable for low-resistance and low-voltage systems in distribution networks.

[0070] In order to improve the richness of simulation test data in the simulation test data set, the present invention can perform simulation tests on the simulated electrical circuit under various load conditions, such as Figure 4 , Figure 5 , Figure 6 As shown, based on Figure 4 It can simulate the electrical circuit including the resistance load. Figure 5 It can simulate the electrical circuit containing inductive load (connected in parallel with small capacitor) based on Figure 6 A simulation test can be performed on a simulated electrical circuit including a resistive load and an inductive load, so that during the simulation test of different simulated electrical circuits, a simulation test data set can be obtained by collecting the current waveform data and the voltage waveform data displayed by the oscilloscope. The simulation test data set can be stored in the workspace so that the simulation test data set can be obtained from the workspace when needed.

[0071] In an optional embodiment, the initial fault detection model based on the back propagation neural network includes: an input layer, a hidden layer, and an output layer. The input layer is used to transmit the simulation test data to the hidden layer, the hidden layer is used to extract the frequency domain characteristics of the simulation test data, and the output layer is used to determine whether an arc fault occurs in the simulated electrical circuit based on the frequency domain characteristics.

[0072] In the present invention, the number of hidden layer neurons in the BP neural network (denoted as L) can be calculated based on the following formula (1):

[0073]

[0074] In the above formula (1), m is the number of input neurons, n is the number of output neurons, and α is a random constant. As an optional implementation, m can be 5, n can be 1, so that the number of hidden layer neurons L is a constant between 3 and 13. Further, after calculating the error of the BP neural network based on different values ​​of L, it can be determined that the neural network error is the smallest when L is 10, so that the number of neurons in the hidden layer of the BP neural network can be determined to be 10.

[0075] For the BP neural network, the function types that need to be used include: activation function, training function, performance function, and transfer function between the hidden layer and the output layer. In the present invention, the training function can be one of tansig and logsig, the transfer function can be at least one of tansig and purelin, and the performance function can be a mean square error loss function (Mean Squared Error, MSE).

[0076] As another optional implementation, the values ​​of various model parameters in the constructed fault detection model based on BP neural network can be as shown in the following Table 1:

[0077] Table 1

[0078]

[0079]

[0080] In an optional embodiment, the arc fault detection method in the electrical circuit further includes:

[0081] Step S206, dividing the simulation test data set according to a preset division ratio to obtain a first data set and a second data set, wherein the first data set is used for model training and the second data set is used for model verification.

[0082] The following combination Figure 7 The above method is further explained.

[0083] like Figure 7 As shown, first, a simulation test platform is built to simulate the low-voltage distribution network line. Then, different load types are selected and simulation test data such as voltage sampling windows and current sampling windows are determined. Further, a low-voltage series arc simulation test is carried out to collect the current waveform data and voltage waveform data of the load when an arc fault occurs in the simulation test circuit under the test condition that the simulation test circuit contains different loads, and the current waveform data and voltage waveform data of the load when no arc fault occurs in the simulation test circuit are collected. When the total number of collected simulation test data (denoted as A) reaches 2000 or more, the simulation test data can be randomly divided to use 90% of the simulation test data as the training data set A1 for model training, and use 10% of the simulation test data as the verification data set A2 for model verification.

[0084] In an optional embodiment, in step S203, the initial fault detection model is trained and verified using the simulation test data set to obtain the target fault detection model including:

[0085] Step S231, iteratively training the initial fault detection model using the first data set to obtain a training result, wherein the training result is used to determine a training error of the initial fault detection model;

[0086] Step S232, in response to the training error being less than or equal to the preset error value, verifying the initial fault detection model using the second data set to obtain a verification result, wherein the verification result is used to determine the detection accuracy of the initial fault detection model for arc faults;

[0087] Step S233 , in response to any one detection accuracy being greater than or equal to the target value, determining the initial fault detection model as the target fault detection model.

[0088] The following combination Figure 7 , Figure 8 The above method is further explained.

[0089] like Figure 7 , Figure 8 As shown in FIG. 1 , as an optional implementation, after the model parameters of the initial fault detection model are determined, the initial fault detection model is trained using the training data set A1, as shown in FIG. Figure 8 As shown in the training curve, as the number of iterations increases, the mean square error loss of the model continues to decrease. After 9 iterations of training, the mean square error loss of the model is reduced to the preset error value (10 -9 ) below, and according to Figure 8 As shown in the validation curve, when the trained model is validated using the validation data set A2, after five iterations, the model error reaches the target error value of 10 -3, the model detection accuracy can reach more than 99% (the above target value). At this time, the verified fault detection model can be used as the target fault detection model.

[0090] As another optional implementation, a certain number of samples may be randomly selected from the verification data set A2 to verify the trained fault detection model. An optional verification result may be shown in the following Table 2:

[0091] Table 2

[0092] Load Type Sample size Number of identification results Recognition success rate resistance 10 10 100% inductance 10 10 100% Resistors and inductors in parallel 10 9 90%

[0093] In an optional embodiment, the arc fault detection method in the electrical circuit further includes:

[0094] Step S271, in response to the training error being greater than a preset error value, adjusting the model parameters of the initial fault detection model according to the training error to obtain an updated fault detection model;

[0095] Step S272: repeatedly training and verifying the updated fault detection model using the first data set to determine whether the updated fault detection model is a target fault detection model.

[0096] Still Figure 7 As shown, when the detection accuracy of the verified model is lower than the target value of 99%, based on the back propagation stage of the BP neural network, the network parameters of the neural network are adjusted to update the fault detection model, and then the simulation test data set is used again to iteratively train and verify the updated fault detection model.

[0097] In an optional embodiment, in step S204, the target fault detection model is used to perform fault detection on the actual electrical circuit, and the detection result obtained includes:

[0098] Step S241, collecting a plurality of waveform data to be detected in an actual electrical circuit, wherein different waveform data to be detected correspond to different loads;

[0099] Step S242: using the target fault detection model to classify and identify the multiple waveform data to be detected to obtain a detection result.

[0100] And, in step S242, the detection result is one of a first detection result and a second detection result, the first detection result includes that an arc fault does not occur in the actual electrical circuit, and the second detection result includes that an arc fault occurs in the actual electrical circuit and a target load that causes the arc fault, and a target fault detection model is used to classify and identify a plurality of waveform data to be detected, and the detection results obtained include:

[0101] Step S2421, using the target fault detection model to classify any waveform data to be detected to obtain a classification result, wherein the classification result is used to determine the category of the load corresponding to any waveform data to be detected;

[0102] Step S2422, comparing any one of the waveform data to be detected with the corresponding normal waveform data to obtain a comparison result, wherein the normal waveform data is used to indicate that no arc fault occurs in the load corresponding to any one of the waveform data to be detected;

[0103] Step S2423, in response to the comparison result, determining that the plurality of waveform data to be detected are consistent with the corresponding normal waveform data, and generating a first detection result according to the comparison result;

[0104] Step S2424, in response to the comparison result determining that at least one waveform data to be detected is inconsistent with the corresponding normal waveform data, a second detection result is generated according to the comparison result and the classification result, wherein the load corresponding to the at least one waveform data to be detected is the target load.

[0105] The target fault detection model obtained after training and verification can detect and identify the waveform data of the load in the electrical circuit containing different loads, accurately determine whether there is a fault arc in the electrical circuit, and locate the load type of the target load that causes the fault arc.

[0106] Fig. 9 is a current waveform diagram of an optional electrical circuit according to an embodiment of the present invention, such as Fig. 9 As shown, the electrical circuit corresponding to the current waveform diagram contains a resistive load with a resistance value of 100Ω. The horizontal axis of the waveform diagram represents the detection time in seconds (s), and the vertical axis represents the current value in amperes (A). After analysis, it is determined that no arc fault occurs in the electrical circuit before 0.1s (the waveform diagram is a sinusoidal current waveform diagram at this time, Fig. 9 ), an arc fault occurred 0.1s later (the timing was restarted to obtain Fig. 9 The waveform shown in Fig. 9 As shown, the phenomenon of an arc fault is that the current value remains at 0 for a period of time. The reason for this phenomenon is that the arc current in the AC circuit is briefly extinguished for a period of time after passing through the zero point once, and then reappears. It can be observed that the wavelength division waveform curve is not smooth, so it can be determined that there are high-order harmonics and noise signals in the current, that is, it can be determined that an arc fault has occurred in the electrical circuit.

[0107] When the resistance value of the resistor load is increased to 200Ω, around 0.1s, as the resistance value increases, the distortion of the current waveform becomes more and more obvious. Starting from 0.1s as the starting point of the timing test, it can be detected that Fig.10As shown in the waveform diagram, due to the increase in the resistance value, the current in the electrical circuit changes. Compared with the current waveform curve detected when the electrical circuit operates normally, the change amplitude of the current waveform curve when an arc fault occurs in the electrical circuit decreases (that is, the change range of the current amplitude decreases).

[0108] When the load in the electrical circuit is inductor, the detected current waveform can be as follows Fig.11 As shown, in Fig.11 In the figure, the horizontal axis represents the detection time in seconds (s), and the vertical axis represents the current value in milliamperes (mA). The same as the electrical circuit current waveform corresponding to the resistive load is that a curve that is maintained for a period of time when the current value is 0 (hereinafter referred to as the "target curve") appears. However, the target curve is not prominent in the electrical circuit current waveform corresponding to the inductive load. This is because the inductor has an energy storage characteristic. When the current passes through the zero point, the arc that was originally to be extinguished is in turn provided with energy by the inductor, so that the arc can be further ignited, making the target curve unclear. In addition, it should be noted that when there is an inductive load in the electrical circuit, under the influence of the energy storage characteristic of the voltage not changing suddenly, higher harmonics will be generated in the current waveform, causing the voltage to remain constant for a short time near the zero point. The special phenomenon occurs, which causes the amplitude of the normal operating voltage to exceed the working voltage when the fault arc appears. The maintenance time of the target curve is shorter than that in the case of current load, and the arc is not easy to extinguish. It should be further explained that when the system load is too high, the energy storage characteristics of the inductive load will cause the arc to continue to burn, making it more likely to cause an electrical fire.

[0109] When the load in the electrical circuit is a resistor and an inductor in parallel, the detected current waveform can be as follows Fig.12 As shown in the figure, consistent with the case of pure resistance load, when an arc fault occurs in the electrical circuit, obvious distortion occurs in the current waveform, and Fig.12 The target curve is also maintained for a shorter time under the load conditions shown, but compared to Fig.11 The target curve for the load conditions shown, Fig.12 The target curve is maintained longer under the load conditions shown.

[0110] Based on the above analysis, when the target fault detection model is used to judge whether an arc fault occurs in an actual electrical circuit, it can be analyzed and determined whether an arc fault occurs in the actual electrical circuit according to the characteristics of the target curve in the waveform diagram corresponding to the actual electrical circuit, and the load type that causes the fault arc when an arc fault occurs can be analyzed and determined.

[0111] In an embodiment of the present invention, a simulation test data set of a simulated electrical circuit is first collected, wherein the simulation test data set includes a first type of data and a second type of data, the first type of data is simulation test data of arc faults occurring in the simulated electrical circuit, and the second type of data is simulation test data of no arc faults occurring in the simulated electrical circuit, and an initial fault detection model based on a back propagation neural network is constructed, and then the initial fault detection model is trained and verified using the simulation test data set to obtain a target fault detection model, and finally the target fault detection model is used to perform fault detection on the actual electrical circuit to obtain a detection result, wherein the detection result is at least used to determine whether an arc fault occurs in the actual electrical circuit.

[0112] It is easy to understand that the above-mentioned embodiment of the present invention pre-constructs a simulated electrical circuit including an arc simulation model, performs simulation tests on the arc simulation model under the condition that the simulated electrical circuit is configured with different loads, obtains a simulation data set, and then uses the simulation data set to detect the fault detection model based on the back propagation neural network, so as to continuously adjust the model parameters, so as to achieve the purpose of determining the target fault detection model for efficiently and accurately detecting and locating arc faults in the electrical circuit, thereby achieving the technical effect of improving the detection efficiency of series fault arcs in low-voltage systems and accurately locating fault arcs, and further solves the technical problems of low detection efficiency and difficulty in accurately locating the position of fault arcs in the series arc fault detection method in low-voltage systems provided by the prior art.

[0113] In this embodiment, an arc fault detection device in an electrical circuit is also provided, which is used to implement the above-mentioned embodiments and preferred implementation modes, and the descriptions that have been made are not repeated here. As used below, a "module" is a combination of software and / or hardware that can implement a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0114] Fig.13 is a structural block diagram of an arc fault detection device in an electrical circuit according to an embodiment of the present invention. Fig.13 As shown, the device comprises:

[0115] The acquisition module 1301 is used to acquire a simulation test data set of a simulated electrical circuit, wherein the simulation test data set includes a first type of data and a second type of data, the first type of data is the simulation test data of an arc fault occurring in the simulated electrical circuit, and the second type of data is the simulation test data of no arc fault occurring in the simulated electrical circuit;

[0116] A first building module 1302 is used to build an initial fault detection model based on a back propagation neural network;

[0117] The training module 1303 is used to train and verify the initial fault detection model using the simulation test data set to obtain the target fault detection model;

[0118] The detection module 1304 is used to perform fault detection on the actual electrical circuit using the target fault detection model to obtain a detection result, wherein the detection result is at least used to determine whether an arc fault occurs in the actual electrical circuit.

[0119] Optionally, the arc fault detection device in the above-mentioned electrical circuit also includes: a second construction module 1305 (not shown in the figure), which is used to construct an arc simulation model of the simulated electrical circuit; under different simulation test conditions, the arc simulation model is simulated to obtain a simulation test data set.

[0120] Optionally, the arc fault detection device in the above-mentioned electrical circuit includes: different simulation test conditions satisfy one of the following: different types of loads configured in the simulated electrical circuit; different numbers of loads configured in the simulated electrical circuit; different connection methods of the loads configured in the simulated electrical circuit.

[0121] Optionally, the arc fault detection device in the above-mentioned electrical circuit includes: the frequency domain characteristics of simulation test data of different types of loads before the arc fault occurs are different from the frequency domain characteristics of simulation test data of different types of loads after the arc fault occurs, and the frequency domain characteristics are determined by the waveform data of different types of loads.

[0122] Optionally, the arc fault detection device in the above-mentioned electrical circuit includes: an initial fault detection model based on a back-propagation neural network includes: an input layer, a hidden layer, and an output layer, the input layer is used to transmit the simulation test data to the hidden layer, the hidden layer is used to extract the frequency domain characteristics of the simulation test data, and the output layer is used to determine whether an arc fault occurs in the simulated electrical circuit based on the frequency domain characteristics.

[0123] Optionally, the arc fault detection device in the above-mentioned electrical circuit also includes: a division module 1306 (not shown in the figure), which is used to divide the simulation test data set according to a preset division ratio to obtain a first data set and a second data set, wherein the first data set is used for model training and the second data set is used for model verification.

[0124] Optionally, the above-mentioned training module 1303 is also used to: use the first data set to iteratively train the initial fault detection model to obtain a training result, wherein the training result is used to determine the training error of the initial fault detection model; in response to the training error being less than or equal to a preset error value, use the second data set to verify the initial fault detection model to obtain a verification result, wherein the verification result is used to determine the detection accuracy of the initial fault detection model for arc faults; in response to any one detection accuracy being greater than or equal to a target value, determine the initial fault detection model as a target fault detection model.

[0125] Optionally, the arc fault detection device in the above-mentioned electrical circuit also includes: an updating module 1307 (not shown in the figure), which is used to adjust the model parameters of the initial fault detection model according to the training error in response to the training error being greater than a preset error value, so as to obtain an updated fault detection model; and repeatedly train and verify the updated fault detection model using the first data set to determine whether the updated fault detection model is the target fault detection model.

[0126] Optionally, the detection module 1304 is further used to: collect multiple waveform data to be detected in an actual electrical circuit, wherein different waveform data to be detected correspond to different loads; and classify and identify the multiple waveform data to be detected using a target fault detection model to obtain a detection result.

[0127] Optionally, the detection result is one of a first detection result and a second detection result, the first detection result includes that an arc fault does not occur in the actual electrical circuit, and the second detection result includes that an arc fault occurs in the actual electrical circuit and a target load that causes the arc fault. The above-mentioned detection module 1304 is also used to: classify any waveform data to be detected using a target fault detection model to obtain a classification result, wherein the classification result is used to determine the category of the load corresponding to any waveform data to be detected; compare any waveform data to be detected with the corresponding normal waveform data to obtain a comparison result, wherein the normal waveform data is used to characterize that no arc fault occurs in the load corresponding to any waveform data to be detected; in response to the comparison result, it is determined that multiple waveform data to be detected are consistent with the corresponding normal waveform data, and a first detection result is generated according to the comparison result; in response to the comparison result, it is determined that at least one waveform data to be detected is inconsistent with the corresponding normal waveform data, and a second detection result is generated according to the comparison result and the classification result, wherein the load corresponding to at least one waveform data to be detected is the target load.

[0128] It should be noted that the above modules can be implemented by software or hardware. For the latter, it can be implemented in the following ways, but not limited to: the above modules are all located in the same processor; or the above modules are located in different processors in any combination.

[0129] According to another aspect of an embodiment of the present invention, there is further provided a computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the computer program implements any one of the aforementioned methods for detecting arc faults in an electrical circuit.

[0130] According to another aspect of an embodiment of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium including a stored executable program, wherein when the executable program is running, the device where the computer-readable storage medium is located is controlled to execute any one of the aforementioned methods for detecting arc faults in an electrical circuit.

[0131] Optionally, in this embodiment, the storage medium may be configured to store a computer program for performing the following steps:

[0132] Step S1, collecting a simulation test data set of a simulated electrical circuit, wherein the simulation test data set includes a first type of data and a second type of data, the first type of data is simulation test data of an arc fault occurring in the simulated electrical circuit, and the second type of data is simulation test data of no arc fault occurring in the simulated electrical circuit;

[0133] Step S2, constructing an initial fault detection model based on a back propagation neural network;

[0134] Step S3, using the simulation test data set to train and verify the initial fault detection model to obtain the target fault detection model;

[0135] Step S4, performing fault detection on the actual electrical circuit using the target fault detection model to obtain a detection result, wherein the detection result is at least used to determine whether an arc fault occurs in the actual electrical circuit.

[0136] Optionally, in this embodiment, the above-mentioned storage medium may include but is not limited to: a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk, and other media that can store computer programs.

[0137] According to another aspect of an embodiment of the present invention, there is also provided an electronic device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to execute any one of the aforementioned methods for detecting arc faults in an electrical circuit.

[0138] Optionally, in this embodiment, the processor may be configured to perform the following steps through a computer program:

[0139] Step S1, collecting a simulation test data set of a simulated electrical circuit, wherein the simulation test data set includes a first type of data and a second type of data, the first type of data is simulation test data of an arc fault occurring in the simulated electrical circuit, and the second type of data is simulation test data of no arc fault occurring in the simulated electrical circuit;

[0140] Step S2, constructing an initial fault detection model based on a back propagation neural network;

[0141] Step S3, using the simulation test data set to train and verify the initial fault detection model to obtain the target fault detection model;

[0142] Step S4, performing fault detection on the actual electrical circuit using the target fault detection model to obtain a detection result, wherein the detection result is at least used to determine whether an arc fault occurs in the actual electrical circuit.

[0143] Optionally, the specific examples in this embodiment may refer to the examples described in the above embodiment and its optional implementation modes, and this embodiment will not be described in detail here.

[0144] The serial numbers of the above embodiments of the present invention are for description only and do not represent the advantages or disadvantages of the embodiments.

[0145] In the above embodiments of the present invention, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0146] In the several embodiments provided by the present invention, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the units can be a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0147] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0148] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0149] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM), random access memory (RAM), mobile hard disk, magnetic disk or optical disk, etc., various media that can store program codes.

[0150] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A method for detecting arc faults in an electrical circuit, characterized in that: include: Collecting a simulation test data set of a simulated electrical circuit, wherein the simulation test data set includes a first type of data and a second type of data, the first type of data is simulation test data in which an arc fault occurs in the simulated electrical circuit, and the second type of data is simulation test data in which the arc fault does not occur in the simulated electrical circuit; Construct an initial fault detection model based on back-propagation neural network; Using the simulation test data set to train and verify the initial fault detection model to obtain a target fault detection model; The target fault detection model is used to perform fault detection on an actual electrical circuit to obtain a detection result, wherein the detection result is at least used to determine whether the arc fault occurs in the actual electrical circuit.

2. The method for detecting arc faults in an electrical circuit according to claim 1, characterized in that: The arc fault detection method in the electrical circuit also includes: Constructing an arc simulation model of the simulated electrical circuit; Under different simulation test conditions, a simulation test is performed on the arc simulation model to obtain the simulation test data set.

3. The method for detecting arc faults in an electrical circuit according to claim 2, characterized in that: The different simulation test conditions meet one of the following conditions: The types of loads configured in the simulated electrical circuit are different; The number of loads configured in the simulated electrical circuit is different; The load connection methods configured in the simulated electrical circuit are different.

4. The method for detecting arc faults in an electrical circuit according to claim 3, characterized in that: The frequency domain characteristics of the simulation test data of different types of loads before an arc fault occurs are different from the frequency domain characteristics of the simulation test data of the different types of loads after an arc fault occurs, and the frequency domain characteristics are determined by the waveform data of the different types of loads.

5. The method for detecting arc faults in an electrical circuit according to claim 1, characterized in that: The initial fault detection model based on back propagation neural network includes: an input layer, a hidden layer, and an output layer. The input layer is used to transmit the simulation test data to the hidden layer, the hidden layer is used to extract the frequency domain characteristics of the simulation test data, and the output layer is used to determine whether an arc fault occurs in the simulated electrical circuit based on the frequency domain characteristics.

6. The method for detecting arc faults in an electrical circuit according to claim 1, characterized in that: The arc fault detection method in the electrical circuit also includes: The simulation test data set is divided according to a preset division ratio to obtain a first data set and a second data set, wherein the first data set is used for model training and the second data set is used for model verification.

7. The method for detecting arc faults in an electrical circuit according to claim 6, characterized in that: The initial fault detection model is trained and verified by using the simulation test data set to obtain the target fault detection model, including: Iteratively training the initial fault detection model using the first data set to obtain a training result, wherein the training result is used to determine a training error of the initial fault detection model; In response to the training error being less than or equal to a preset error value, verifying the initial fault detection model using the second data set to obtain a verification result, wherein the verification result is used to determine the detection accuracy of the initial fault detection model for the arc fault; In response to any one detection accuracy being greater than or equal to a target value, the initial fault detection model is determined as the target fault detection model.

8. The method for detecting arc faults in an electrical circuit according to claim 7, characterized in that: The arc fault detection method in the electrical circuit also includes: In response to the training error being greater than the preset error value, adjusting a model parameter of the initial fault detection model according to the training error to obtain an updated fault detection model; The updated fault detection model is repeatedly trained and verified using the first data set to determine whether the updated fault detection model is the target fault detection model.

9. The method for detecting arc faults in an electrical circuit according to claim 1, characterized in that: The target fault detection model is used to perform fault detection on an actual electrical circuit, and the detection result obtained includes: Collecting a plurality of waveform data to be detected in the actual electrical circuit, wherein different waveform data to be detected correspond to different loads; The target fault detection model is used to classify and identify the multiple waveform data to be detected to obtain the detection result.

10. The method for detecting arc faults in an electrical circuit according to claim 9, characterized in that: The detection result is one of a first detection result and a second detection result, wherein the first detection result includes that the arc fault does not occur in the actual electrical circuit, and the second detection result includes that the arc fault occurs in the actual electrical circuit and the target load causing the arc fault. The target fault detection model is used to classify and identify the multiple waveform data to be detected, and the detection result obtained includes: Using the target fault detection model to classify any waveform data to be detected, and obtain a classification result, wherein the classification result is used to determine the category of the load corresponding to the any waveform data to be detected; Compare the any one of the waveform data to be detected with the corresponding normal waveform data to obtain a comparison result, wherein the normal waveform data is used to indicate that the load corresponding to the any one of the waveform data to be detected does not have the arc fault; In response to the comparison result, it is determined that the plurality of waveform data to be detected are all consistent with the corresponding normal waveform data, and the first detection result is generated according to the comparison result; In response to the comparison result, it is determined that at least one waveform data to be detected is inconsistent with the corresponding normal waveform data, and the second detection result is generated according to the comparison result and the classification result, wherein the load corresponding to the at least one waveform data to be detected is the target load.

11. An arc fault detection device in an electrical circuit, characterized in that: include: An acquisition module, used for acquiring a simulation test data set of a simulated electrical circuit, wherein the simulation test data set includes a first type of data and a second type of data, the first type of data is the simulation test data when an arc fault occurs in the simulated electrical circuit, and the second type of data is the simulation test data when the arc fault does not occur in the simulated electrical circuit; A first building module is used to build an initial fault detection model based on a back propagation neural network; A training module, used to train and verify the initial fault detection model using the simulation test data set to obtain a target fault detection model; The detection module is used to perform fault detection on the actual electrical circuit using the target fault detection model to obtain a detection result, wherein the detection result is at least used to determine whether the arc fault occurs in the actual electrical circuit.

12. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the arc fault detection method in an electrical circuit of any one of claims 1 to 10 is implemented.

13. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored executable program, wherein when the executable program is run, the device where the computer-readable storage medium is located is controlled to execute the arc fault detection method in the electrical circuit of any one of claims 1 to 10.

14. An electronic device, characterized in that: The method comprises a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to execute the method for detecting arc faults in an electrical circuit according to any one of claims 1 to 10.