A fault arc experiment platform and a fault arc detection method

Abstract

The application discloses a kind of fault arc experimental platform and fault arc detection method, the experimental platform includes power supply, electrical fault simulator, circuit breaker, fault arc generation module and current acquisition analysis module, power supply is connected with electrical fault simulator, electrical fault simulator, circuit breaker and fault arc generation module are connected in series between, current acquisition analysis module is arranged between circuit breaker and fault arc generation module, fault arc generation module includes fault arc generation device and experimental load, fault arc generation device and experimental load are connected in series or in parallel connection.The experimental platform of fault arc of the application can be realized experimental load and the function of fault arc device series connection or parallel connection, and different types of load combination can be used for load replacement, ensure the reliability of fault arc detection, can obtain the current data of multiple fault arcs, provide sufficient and effective data support for fault arc detection method.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of fault arc experiment, and in particular to a fault arc experiment platform and a fault arc detection method. BACKGROUND

[0002] Arc, commonly known as "electric spark", has the characteristics of high temperature, small current and short duration, and is a kind of gas ionization discharge phenomenon formed by excessive electric field strength of gas breakdown. Once breakdown occurs, it will occur frequently. Because a large amount of heat is generated during arc discharge, it can ignite the surrounding flammable and explosive materials, causing fire and even explosion, so it is necessary to conduct in-depth research on fault arc detection.

[0003] Fault arc is mainly divided into three types, including series fault arc, parallel fault arc and ground fault arc. The existing alternating current fault arc experiment platform is mostly a series circuit, and the experimental load is a single load, without considering the complexity of the electrical circuit in the power supply and distribution line and the diversity of electrical loads, lacking fault reproduction of series and parallel connection modes and different load conditions. And there is no electrical fault device in the existing alternating current fault arc experiment platform, lacking arc data under different electrical fault conditions. SUMMARY

[0004] The purpose of the present application is to provide a fault arc experiment platform and a fault arc detection method, which can realize the functions of series or parallel connection of experimental load and fault arc device through the built fault arc experiment platform, and can replace the load by using different types of load combination, ensuring the reliability of fault arc detection.

[0005] In order to achieve the above-mentioned purpose of the application, the technical scheme adopted by the present application is as follows:

[0006] According to one aspect of the present application, a fault arc experiment platform and a fault arc detection method are provided, comprising a power supply, an electrical fault simulator, a circuit breaker, a fault arc generating module and a current collection and analysis module.

[0007] The power supply is connected with the electrical fault simulator.

[0008] The electrical fault simulator, the circuit breaker and the fault arc generating module are connected in series.

[0009] The circuit breaker and the fault arc generating module are provided with a current sampling point, and the current collection and analysis module is connected with the current sampling point.

[0010] The arc fault occurrence module comprises an arc fault occurrence device and an experimental load, and the arc fault occurrence device is connected in series or parallel with the experimental load.

[0011] Preferably, the electrical fault simulator comprises a lightning surge generator, an impulse group generator and an electrostatic discharge generator.

[0012] Preferably, the arc fault occurrence device comprises a moving electrode and a stationary electrode, the moving electrode is fixedly arranged on a moving mechanism, the moving mechanism is used for moving the moving electrode, the stationary electrode is fixedly arranged on an insulating base, and the moving electrode can be close to or away from the stationary electrode.

[0013] Preferably, the experimental load comprises an electric kettle, a hair dryer, an induction cooker, an electric fan, an energy-saving lamp, a notebook computer and a microwave oven.

[0014] Preferably, the current collection and analysis module comprises a current clamp and an oscilloscope, the current clamp is sleeved on the current sampling point, and the current clamp is electrically connected with the oscilloscope.

[0015] Preferably, the method comprises the following steps:

[0016] S1, sequentially connecting a power supply, an electrical fault simulator, a circuit breaker, an arc fault occurrence device and an experimental load, wherein the arc fault occurrence device and the experimental load are connected in series or parallel, a current clamp is connected to a circuit and connected with an oscilloscope;

[0017] S2, turning on the electrical fault simulator, and controlling the electrical fault simulator to generate an electrical fault input to the arc fault occurrence device;

[0018] S3, when the simulated electrical fault occurs, adjusting the distance between the moving electrode and the stationary electrode to generate a stable burning arc;

[0019] S4, collecting and storing current signals in normal and arc fault states by the oscilloscope;

[0020] S5, replacing different loads, and repeating steps S2-S4;

[0021] S6, replacing the connection mode of the arc fault occurrence device and the experimental load, and repeating steps S2-S5;

[0022] S7, replacing the electrical fault simulator, and repeating steps S2-S6;

[0023] S8, pre-process the collected arc current data, analyze and extract the time domain features of the current waveform, extract the frequency domain features of different time scales, input the time domain features and the frequency domain features into a double-channel convolutional neural network respectively for training, obtain a fusion feature vector, train a fault arc detection model using the fusion feature vector, and detect the fault arc using the trained fault arc detection model to output the fault arc type.

[0024] Preferably, in step S8, the double-channel convolutional neural network comprises a one-dimensional convolutional neural network and a two-dimensional convolutional neural network, the frequency domain features are input into the one-dimensional convolutional neural network for training, and the time domain features are input into the two-dimensional convolutional neural network for training.

[0025] Preferably, the frequency domain features are extracted using wavelet transform, and the formula of the wavelet transform is:

[0026]

[0027] wherein f(t) is a current time domain signal, ψ is a wavelet base function, a is a scale controlling the stretching of the wavelet function; τ is a translation amount controlling the translation of the wavelet function, and t is time.

[0028] Preferably, in step S8, the loss function of the fault arc detection model is:

[0029]

[0030] wherein N represents the total number of data samples; y i represents the label of the data sample, represents the predicted label of the sample.

[0031] In summary, due to the adoption of the above technical solutions, the present application has the following beneficial effects:

[0032] 1. The fault arc experimental platform built by the present application can realize the functions of series connection or parallel connection of the experimental load and the fault arc device, and can replace the load by using different types of loads, thereby ensuring the reliability of fault arc detection, and can simulate fault arc reproduction experiments under different line connection modes and different load conditions, thereby obtaining current data of various fault arcs and providing sufficient and effective data support for fault arc detection.

[0033] 2. The present application adds an electrical fault device to simulate different electrical fault generators for simulating fault arc experimental research under different electrical faults. BRIEF DESCRIPTION OF DRAWINGS

[0034] Figure 1 is a structural schematic diagram of a series connection experimental platform of the present application;

[0035] Figure 2 Figure 1 is a structural schematic diagram of a parallel experiment platform of the present application;

[0036] Figure 3 Figure 3 is a flow chart of a fault arc detection method of the present application;

[0037] Figure 4 Figure 4 is a schematic diagram of a dual-channel convolutional neural network architecture of the present application. DETAILED DESCRIPTION

[0038] In order to make the objectives, technical solutions and advantages of the present application clearer, the present application is further described in detail below with reference to the drawings and preferred embodiments. However, it should be noted that many details listed in the description are only for the reader to have a thorough understanding of one or more aspects of the present application, and the aspects of the present application can be realized even without these specific details.

[0039] Referring to Figures 1 to 4 The present application provides a fault arc experiment platform and a fault arc detection method, and the technical solutions are as follows:

[0040] A fault arc experiment platform, comprising a power supply, an electrical fault simulator, a circuit breaker, a fault arc generation module and a current collection and analysis module. The power supply is a 220V / 50Hz alternating current power supply. The electrical fault simulator is one of a lightning surge generator, a pulse group generator and an electrostatic discharge generator. The lightning surge generator is used to simulate high-energy transient interference caused by lightning strikes from switches and the natural world; the pulse group generator is used to simulate electrical fast transient pulse group interference; and the electrostatic discharge generator is used to simulate electrostatic discharge interference directly from operators and adjacent objects. The alternating current power supply is connected with the electrical fault simulator. The electrical fault simulator, the circuit breaker and the fault arc generation module are connected in series. The fault arc generation module comprises a fault arc generation device and an experimental load. The fault arc generation device comprises a moving electrode and a stationary electrode. The moving electrode is a 6.0mm copper rod with a conical front end, and the moving electrode can be controlled to move forward or backward. The moving electrode is fixedly arranged on a moving mechanism, and the moving mechanism is used to move the moving electrode. It should be noted that the moving mechanism can be an electrically adjustable mechanism, such as a pneumatic cylinder driving the moving electrode to move, or an electric telescopic rod driving the moving electrode to move. Alternatively, the moving mechanism can be a manually adjustable mechanism, such as a sliding groove with a sliding block arranged therein, and the moving electrode is arranged on the sliding block and can be manually controlled to move. As long as the mechanism can move the moving electrode, the moving operation of the moving electrode can be realized, and the present application will not be described in more detail. The stationary electrode is a 6.0mm carbon rod with a planar front end, and is fixedly arranged on an insulating base. When simulating an electrical fault, the distance between the moving electrode and the stationary electrode is adjusted to generate a stable burning arc. For example, the moving electrode is controlled to move towards the stationary electrode to generate an arc, or the moving electrode is controlled to move away from the stationary electrode to extinguish the arc. The experimental load is a 220V / 50Hz load, and the load is connected in series with the fault arc generation device. Figures 1-2As shown, the fault arc generating device is connected with the experimental load in series or in parallel. When the fault arc generating device is connected with the experimental load in series, the moving electrode is connected in series with the circuit breaker, the stationary electrode is connected in series with the experimental load, or the stationary electrode is connected in series with the circuit breaker, and the moving electrode is connected in series with the experimental load. When the fault arc generating device is connected with the experimental load in parallel, the fault arc generating device is first connected in series with a fixed load and then connected in parallel with the experimental load. In this embodiment, the experimental load is divided into series loads and parallel loads, as shown in the following table:

[0041] Table 1: Series experimental load

[0042]

[0043]

[0044] Table 2: Parallel experimental load

[0045] Serial number Load name Serial number Load name 1 Electric kettle 9 IH cooker + electric fan 2 Hair dryer 10 Electric kettle + electric kettle 3 IH cooker 11 Electric kettle + energy saving lamp 4 Electric fan 12 Hair dryer + electric fan 5 Energy saving lamp 13 IH cooker + energy saving lamp 6 Notebook computer 14 Hair dryer + energy saving lamp 7 Microwave oven 15 Hair dryer + notebook computer 8 Electric kettle + electric fan 16 Energy saving lamp + microwave oven

[0046] As shown in Table 1, the series experimental load includes an electric kettle, a hair dryer, an induction cooker, an electric fan, an energy-saving lamp, a notebook computer, and a microwave oven. When the fault arc experiment of the experimental load is carried out in series connection, different experimental loads can be replaced to obtain fault arc data under different loads.

[0047] As shown in Table 2, the parallel experimental load includes an electric kettle, a hair dryer, an induction cooker, an electric fan, an energy-saving lamp, a notebook computer, and a microwave oven, or two of them are combined as a parallel load. When the fault arc experiment of the experimental load is carried out in parallel connection, different experimental loads can be replaced to obtain fault arc data under different loads.

[0048] In order to collect the current data generated when the fault arc occurs, a current sampling point is arranged between the wires connected in series between the circuit breaker and the fault arc generating module. The current collection and analysis module is connected with the current sampling point, and the current collection and analysis module includes a current clamp and an oscilloscope. The current clamp is sleeved on the wire of the current sampling point, and the current clamp is electrically connected with the oscilloscope. The oscilloscope is used to collect and store the current signals in normal and fault arc states.

[0049] Through the fault arc experimental platform built, the function of connecting the experimental load and the fault arc device in series or in parallel can be realized. Since the arc current presents complex nonlinear characteristics, typical household loads are selected and different types of load combinations are fully considered to ensure the reliability of fault arc detection. Fault arc reproduction experiments under different line connection modes and different load conditions can be simulated.

[0050] Currently, in the fault arc detection method, many studies focus on the time-frequency domain feature analysis of the arc current signal, and use discrete wavelet transform and empirical mode decomposition to extract the time-frequency characteristic components in the current signal. However, whether the method based on time domain, frequency domain or time-frequency domain fusion needs to manually calculate and adjust the threshold, which is difficult to adapt to complex situations. There are also methods that combine machine learning to identify fault arcs, such as using least squares support vector machine (LS-SVM) to identify current feature vectors to detect fault arcs; using a BP neural network based on wavelet transform components to identify series fault arcs of common loads; collecting high-frequency coupled signals of typical loads under normal and fault arc conditions for band-pass filtering, and using TDV-CNN algorithm to extract high-dimensional features of gray images; using the parallel deep convolutional neural network of AlexNet and parallel AlexNet for fault arc detection. However, the current fault arc detection combined with neural networks often extracts single feature values, ignoring the relationship between time domain and frequency domain, and cannot fully integrate the features of fault arc current signals.

[0051] Therefore, in order to solve the above problems, as shown in Figure 3 In this embodiment, based on the above fault arc experiment platform, a fault arc detection method is disclosed, which comprises the following steps:

[0052] S1, electrical circuit connection: sequentially connecting power supply, electrical fault simulator, circuit breaker, fault arc generating device and experimental load, wherein the fault arc generating device and the experimental load are connected in series or parallel, the series circuit is as shown in Figure 1 The parallel circuit is as shown in Figure 2 Finally, the current clamp is connected to the circuit and connected to the oscilloscope.

[0053] S2, simulate electrical fault: turn on the electrical fault simulator and control the electrical fault simulator to generate electrical fault input to the fault arc generating device.

[0054] S3, pull the arc: when the simulated electrical fault occurs, slowly retreat the moving electrode to a certain electrode spacing by manually adjusting the thousandth knob of the fault arc generating device to generate a stable burning arc.

[0055] S4, data acquisition: the oscilloscope acquires and stores the current signal under normal and fault arc conditions, and sets the sampling frequency of the oscilloscope to 62.5 kHz and the sampling time to 4 s.

[0056] S5, repeat the experiment: replace the different load and repeat steps S2-S4;

[0057] S6, repeat the experiment: replace the connection mode of the fault arc generating device and the experimental load, and repeat steps S2-S5;

[0058] S7, repeating the experiment: replacing the electrical fault simulator, repeating steps S2-S6.

[0059] By replacing different loads, replacing the fault arc generating device, and replacing the series-parallel connection mode of the experimental load, and replacing the electrical fault simulator, various types of fault arc types can be simulated, and current data under various conditions can be obtained for subsequent detection and analysis of the current data.

[0060] S8, fault arc detection: preprocessing the arc current data collected above, analyzing and extracting the time domain features of the current waveform, and extracting the frequency domain features of different time scales, inputting the time domain features and the frequency domain features into a dual-channel convolutional neural network for training, obtaining a fusion feature vector, training a fault arc detection model using the fusion feature vector, and detecting the fault arc using the trained fault arc detection model, and outputting the fault arc type.

[0061] Specifically, the dual-channel convolutional neural network DCNN includes a one-dimensional convolutional neural network 1D-CNN and a two-dimensional convolutional neural network 2D-CNN. As shown in Figure 4 The dual-channel convolutional neural network DCNN inputs the time-frequency domain features extracted by wavelet transform into the 1D-CNN training, where the formula of the wavelet transform is as follows:

[0062]

[0063] Where f(t) is the current time domain signal, ψ is the wavelet base function, a is the scale that controls the stretching of the wavelet function, and the scale corresponds to the frequency; τ is the translation that controls the translation of the wavelet function, and the translation corresponds to the time, t is the time.

[0064] The 1D-CNN mainly includes 3 convolutional layers, 3 max pooling layers, and 1 fully connected layer, and the output plane of the last pooling layer is flattened. To avoid the problems of gradient explosion and gradient disappearance, a nonlinear activation function ReLU is added after each convolutional layer, and the ReLU function is as follows:

[0065] ReLU(x) = Max(0, x)

[0066] A sigmoid function is added after the fully connected layer, and the sigmoid function is as follows:

[0067]

[0068] The collected current time domain signal is input to 2D-CNN training, and 2D-CNN mainly includes 5 convolution layers, 3 maximum pooling layers and 3 fully connected layers, and ReLU activation function is applied before each maximum pooling layer.

[0069] The features extracted by the two channels are input to a fully connected layer for label classification, each neuron in the fully connected layer is connected to all neurons in the previous layer with learnable weights, the two channel extracted feature vectors are spliced for output discrimination, and a fusion feature vector is obtained. The fusion feature vector is used to train a fault arc detection model, the fault arc detection model is an artificial neural network model, and the trained fault arc detection model is used to detect fault arcs and identify fault arc types.

[0070] In order to make the fault arc detection model have better generalization ability and improve the recognition ability of the convolutional neural network to the features, and make the convergence speed faster. The fault arc detection model sets the number of iterations to 120, the learning rate to 0.0001, and the loss function to binary cross-entropy, and the formula of the loss function is as follows:

[0071]

[0072] Where, N represents the total number of data samples; y i represents the label of the data sample, represents the predicted label of the sample. In the model optimization process, the Adam optimization algorithm is adopted, the Dropout is introduced and the Dropout rate is set to 0.5.

[0073] By inputting different kinds of loads of series circuits and parallel circuits for training, the accuracy of fault arcs can reach 96.6%, which verifies the effectiveness of the method in fault arc detection. The artificial neural network model is trained using the fusion feature vector, which can realize fault arc detection of different simulated appliance faults, different load conditions and different connection modes. Wavelet analysis is used for multi-scale decomposition to extract the time-frequency domain features of the current, and the connection between the time-frequency domain of the fault arc signal is improved.

[0074] The above only describes the preferred embodiments of the present application, and it should be noted that for those skilled in the art, without departing from the principles of the present application, a number of improvements and refinements can be made, and these improvements and refinements should also be considered as the protection scope of the present application.

Claims

1. A fault arc testing method for a fault arc test platform, characterized by: The following steps are involved: S1. Connect the power supply, electrical fault simulator, circuit breaker, arc fault generator, and experimental load in series or parallel. Connect the current clamp to the circuit and connect it to an oscilloscope. S2. Turn on the electrical fault simulator and control the electrical fault simulator to generate an electrical fault and input it into the fault arc generating device; S3. When a simulated electrical fault occurs, the distance between the moving electrode and the stationary electrode is adjusted to generate a stable burning arc; S4. Collecting and storing data of current signals in normal and fault arc states through an oscilloscope; S5. Change to a different load and repeat steps S2-S4; S6. Change the connection mode between the fault arc generating device and the experimental load, and repeat steps S2-S5; S7, replace the electrical fault simulator and repeat steps S2-S6; S8. Preprocessing the collected arc current data, analyzing and extracting the time domain features of the current waveform, and simultaneously extracting the frequency domain features of different time scales, respectively inputting the time domain features and the frequency domain features into a dual-channel convolutional neural network for training to obtain a fused feature vector, using the fused feature vector to train a fault arc detection model, and using the trained fault arc detection model to detect the fault arc and output the fault arc type; Among them, the dual-channel convolutional neural network includes a one-dimensional convolutional neural network and a two-dimensional convolutional neural network, the frequency domain features are input into the one-dimensional convolutional neural network for training, and the time domain features are input into the two-dimensional convolutional neural network for training.

2. The arc fault detection method of the arc fault experimental platform according to claim 1, characterized in that: The frequency domain features are extracted using wavelet transform, and the formula of the wavelet transform is: ， is the current time domain signal, is the wavelet basis function, To control the scale of the wavelet function; To control the translation amount of the wavelet function, For time.

3. The arc fault detection method of the arc fault experimental platform according to claim 1, characterized in that: In step S8, the loss function of the arc fault detection model is: ， in, Indicates the total number of data samples; represents the label of the data sample, Indicates the label of the sample prediction.

4. A fault arc experimental platform, characterized in that: It includes power supply, electrical fault simulator, circuit breaker, arc fault generation module and current acquisition and analysis module; The power supply is connected to the electrical fault simulator; The electrical fault simulator, the circuit breaker and the arc fault generating module are connected in series, wherein the electrical fault simulator includes a lightning surge generator, a pulse group generator and an electrostatic discharge generator; A current sampling point is provided between the circuit breaker and the arc fault generation module, and the current acquisition and analysis module is connected to the current sampling point; The fault arc generating module includes a fault arc generating device and an experimental load, and the fault arc generating device is connected to the experimental load in series or in parallel.

5. The arc fault experimental platform according to claim 4, characterized in that: The fault arc generating device includes a movable electrode and a stationary electrode. The movable electrode is fixedly arranged on a movable mechanism. The movable mechanism is used to move the movable electrode. The stationary electrode is fixedly arranged on an insulating base. The movable electrode can be moved closer to or farther away from the movable electrode.

6. The arc fault experimental platform according to claim 4, characterized in that: The experimental loads include an electric kettle, a hair dryer, an induction cooker, an electric fan, an energy-saving lamp, a laptop computer and a microwave oven.

7. The arc fault experimental platform according to claim 4, characterized in that: The current acquisition and analysis module includes a current clamp and an oscilloscope. The current clamp is mounted on the current sampling point, and the current clamp is electrically connected to the oscilloscope.

Images