Electric motorcycle online fault diagnosis method based on real-time current and voltage signals

Abstract

The invention provides an electric motorcycle online fault diagnosis method based on real-time current and voltage signals. The electric motorcycle online fault diagnosis method comprises the steps that a current sensor and a voltage sensor are used for collecting real-time signals of an incoming line on a motor driving board when a target electric motorcycle runs; the collected real-time signals are transmitted to a remote server in real time through the wireless communication module; in the remote server, performing denoising preprocessing on the real-time signal to obtain a first preprocessed signal; performing noise reduction preprocessing on the first preprocessing signal by adopting an SSA-SGMD algorithm to obtain a second preprocessing signal; processing the second preprocessing signal by using synchronous extrusion wavelet transform to obtain a two-dimensional feature image; and performing feature extraction and classification on the two-dimensional feature image by using the trained fault diagnosis model to distinguish and diagnose various fault types to obtain a fault diagnosis result. By collecting the real-time current and voltage signals during operation of the electric motorcycle, the working state of the motorcycle can be monitored in real time, and potential faults can be found in time, so that the safety is improved.

Description

Technical Field

[0001] The invention relates to the technical field of fault diagnosis, and in particular to an online fault diagnosis method for an electric motorcycle based on real-time current and voltage signals. Background Art

[0002] As an important part of modern urban transportation, the stability and safety of electric motorcycles are of vital importance. However, traditional fault diagnosis methods mainly rely on manual or simple data monitoring, which cannot effectively identify potential problems under complex working conditions. Therefore, it is particularly important to develop a method that can accurately and real-timely perform fault diagnosis.

[0003] The prior art discloses a method for diagnosing a motorcycle electronic control system fault based on the Internet of Things, comprising the following steps: performing parameter extraction and analysis when the motorcycle electronic control system fails: generating a test cycle and an operating parameter i, obtaining the parameter value of the motorcycle electronic control system and marking it as a fault value when the motorcycle electronic control system fails during the test cycle, obtaining the fault range i through the fault value, and forming a fault set by all fault ranges i; the prior art analyzes the fault set of the diagnosis object by performing parameter extraction and analysis when the motorcycle electronic control system fails, and feeds back the correlation between the fault type and the fault parameter through the fault set, so as to directly screen the fault type according to the fault set when performing fault diagnosis. It can be seen from the above records that the prior art forms a fault set by obtaining the fault value and constructing the fault range i, but this method cannot fully cover all potential fault types, especially under complex working conditions, and may miss some hidden faults. Moreover, the parameter extraction and analysis is performed after the fault occurs, which will cause a delay in fault diagnosis. Summary of the invention

[0004] In order to overcome the shortcomings of the prior art, the purpose of the present invention is to provide an online fault diagnosis method for an electric motorcycle based on real-time current and voltage signals. By collecting real-time current and voltage signals when the electric motorcycle is running, the working status of the vehicle can be monitored in real time, potential faults can be discovered in time, thereby improving safety.

[0005] To achieve the above object, the present invention provides the following solutions:

[0006] An online fault diagnosis method for an electric motorcycle based on real-time current and voltage signals, comprising:

[0007] Use current sensors and voltage sensors to collect real-time signals of incoming wires on the motor drive board when the target electric motorcycle is running;

[0008] The collected real-time signals are transmitted to the remote server in real time through the wireless communication module;

[0009] In the remote server, performing denoising preprocessing on the real-time signal to obtain a first preprocessed signal;

[0010] Performing noise reduction preprocessing on the first preprocessed signal using an SSA-SGMD algorithm to obtain a second preprocessed signal;

[0011] Processing the second preprocessed signal using synchronous squeezing wavelet transform to obtain the two-dimensional feature image;

[0012] The trained fault diagnosis model is used to extract and classify the two-dimensional feature image to distinguish and diagnose various fault types and obtain a fault diagnosis result.

[0013] Preferably, it also includes:

[0014] Using the remote server to provide the fault diagnosis result to the user and maintenance personnel of the target electric motorcycle through a user interface;

[0015] Store operating data and diagnostic results in a cloud database for future data analysis and pattern recognition.

[0016] Preferably, the method for constructing the fault diagnosis model includes:

[0017] Get the preset annotation dataset;

[0018] Standardizing and data enhancing the labeled data set to obtain an enhanced data set;

[0019] Using a deep convolutional generative adversarial network to perform image synthesis on the enhanced dataset to obtain a training dataset;

[0020] The training data set is input into the improved VGG network for training to obtain the fault diagnosis model; the VGG network includes: a convolution layer, a pooling layer, a fully connected layer and an output layer; the pooling layer is connected after each of the convolution layers; the output of the pooling layer is connected to the fully connected layer; the output of the fully connected layer is connected to the output layer.

[0021] Preferably, in the remote server, performing denoising preprocessing on the real-time signal to obtain a first preprocessed signal includes:

[0022] Performing multi-level decomposition on the real-time signal to obtain wavelet coefficients;

[0023] Performing threshold processing on the wavelet coefficients to obtain processed wavelet coefficients;

[0024] Signal reconstruction is performed on the processed wavelet coefficients to obtain the first preprocessed signal.

[0025] Preferably, thresholding is performed on the wavelet coefficients to obtain processed wavelet coefficients, including:

[0026] When the value of the wavelet coefficient is greater than or equal to the preset processing threshold, the wavelet coefficient is processed by using the first coefficient processing function to obtain the processed wavelet coefficient; wherein the first coefficient processing function is:

[0027]

[0028] Where u is the first adjustment coefficient, sgn() is the sign function, and w j,l is the lth wavelet coefficient under the jth decomposition level, λ is the processing threshold, α is the second adjustment coefficient, β is the third adjustment coefficient, γ is the fourth adjustment coefficient, p is the wavelet decomposition level, is the wavelet coefficient after processing;

[0029] When the value of the wavelet coefficient is less than the processing threshold, the wavelet coefficient is processed by a second coefficient processing function to obtain a processed wavelet coefficient; wherein the second coefficient processing function is:

[0030]

[0031] Preferably, the processing threshold calculation formula is:

[0032]

[0033] Wherein, λ is the processing threshold, δ is the standard deviation of Gaussian noise, M is the length of the real-time signal, and z is the preset coefficient.

[0034] Preferably, the first preprocessed signal is subjected to noise reduction preprocessing using an SSA-SGMD algorithm to obtain a second preprocessed signal, comprising:

[0035] The optimization parameters of symplectic geometric mode decomposition are optimized using the sparrow search algorithm to obtain the SSA-SGMD algorithm;

[0036] The first preprocessed signal is subjected to noise reduction preprocessing by using the SSA-SGMD algorithm to obtain a second preprocessed signal.

[0037] Preferably, after the first preprocessed signal is subjected to noise reduction preprocessing by adopting the SSA-SGMD algorithm to obtain the second preprocessed signal, the method further comprises:

[0038] The second preprocessed signal is segmented, the length of each segment is determined, and overlapping sampling is used to obtain the second preprocessed signal converted into a two-dimensional image.

[0039] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0040] The present invention provides an online fault diagnosis method for electric motorcycles based on real-time current and voltage signals, comprising: using a current sensor and a voltage sensor to collect real-time signals of incoming lines on a motor drive board of a target electric motorcycle when the target electric motorcycle is running; transmitting the collected real-time signals to a remote server in real time through a wireless communication module; in the remote server, performing denoising preprocessing on the real-time signals to obtain a first preprocessing signal; using an SSA-SGMD algorithm to perform denoising preprocessing on the first preprocessing signal to obtain a second preprocessing signal; using a synchronous squeezed wavelet transform to process the second preprocessing signal to obtain the two-dimensional feature image; using a trained fault diagnosis model to perform feature extraction and classification on the two-dimensional feature image to distinguish and diagnose various fault types and obtain a fault diagnosis result. By collecting real-time current and voltage signals when the electric motorcycle is running, the present invention can monitor the working status of the vehicle in real time, discover potential faults in time, and thus improve safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0042] Figure 1 A flow chart of a method provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0043] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0044] The purpose of the present invention is to provide an online fault diagnosis method for an electric motorcycle based on real-time current and voltage signals. By collecting real-time current and voltage signals when the electric motorcycle is running, the working state of the vehicle can be monitored in real time, potential faults can be discovered in time, thereby improving safety.

[0045] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0046] Figure 1 A flow chart of a method provided by an embodiment of the present invention, such as Figure 1 As shown, the present invention provides an online fault diagnosis method for an electric motorcycle based on real-time current and voltage signals, comprising:

[0047] Step 100: using a current sensor and a voltage sensor to collect a real-time signal of an incoming line on a motor drive board when the target electric motorcycle is running;

[0048] Step 200: Transmitting the collected real-time signal to a remote server in real time through a wireless communication module;

[0049] Step 300: In the remote server, the real-time signal is subjected to denoising preprocessing to obtain a first preprocessed signal;

[0050] Step 400: using the SSA-SGMD algorithm to perform noise reduction preprocessing on the first preprocessed signal to obtain a second preprocessed signal;

[0051] Step 500: Processing the second preprocessed signal using synchronous squeezing wavelet transform to obtain the two-dimensional feature image;

[0052] Step 600: Use the trained fault diagnosis model to perform feature extraction and classification on the two-dimensional feature image to distinguish and diagnose various fault types and obtain a fault diagnosis result.

[0053] Specifically, in step 100, the present embodiment selects a sensor suitable for the current range of the electric motorcycle, such as a Hall effect current sensor. The sensor can measure the current without directly contacting the circuit. And a sensor suitable for the voltage range of the electric motorcycle is selected, such as a voltage divider circuit or a dedicated voltage sensor, to ensure that the battery voltage and the motor voltage can be accurately measured.

[0054] Furthermore, in this embodiment, the current sensor is connected in series in the power line of the electric motorcycle to ensure that the current can pass through the sensor. Ensure that the power connection of the sensor is normal, usually it is necessary to provide an appropriate power supply voltage. And the voltage sensor is connected in parallel to the power supply terminal of the battery or motor to ensure that the voltage can be accurately measured. If a voltage divider circuit is used, ensure that the resistance value is reasonably selected to avoid burdening the circuit. In this embodiment, the output signals of the current sensor and the voltage sensor are connected to the analog input port of the microcontroller. Ensure that the microcontroller can read the output signal of the sensor, which is usually an analog voltage signal.

[0055] As an optional implementation, the hardware selection and connection of this embodiment are as follows:

[0056] Select a suitable current sensor, such as a Hall effect current sensor, to ensure that the current range of the electric motorcycle is met, and select a suitable voltage sensor, such as a voltage divider circuit or a dedicated voltage sensor (such as the ACS series) to ensure that the input voltage range meets the requirements. Use a microcontroller (such as Arduino, RaspberryPi, etc.) for signal acquisition and processing.

[0057] Connect the input of the current sensor to the incoming line of the electric motorcycle motor driver board, and the output to the analog input of the microcontroller. And connect the voltage sensor to the power supply of the electric motorcycle motor driver board, and output to another analog input of the microcontroller. Finally, make sure that the microcontroller and sensor get stable power supply.

[0058] The software programming method for data acquisition in this embodiment is as follows:

[0059] 1. Initialize the current and voltage sensors in the microcontroller code and set the sampling frequency and accuracy.

[0060] Read the signal:

[0061] int currentValue=analogRead(currentSensorPin);

[0062] intvoltageValue=analogRead(voltageSensorPin);

[0063] 2. Data processing: basic processing of the read current and voltage signals (such as conversion into actual current and voltage values).

[0064] 3. Data Collection

[0065] The current and voltage data are collected periodically in the main loop, for example, every 100 milliseconds.

[0066] Format the collected data into a format suitable for transmission, such as JSON.

[0067] 4. Fault detection

[0068] Monitor current and voltage changes in real time, set thresholds to detect abnormalities, and record abnormalities.

[0069] 5. Testing and debugging

[0070] The test is performed while the electric motorcycle is running to check the accuracy and real-time performance of the signal. The sensor parameters and program logic are adjusted according to the test results to ensure stable and accurate data acquisition.

[0071] Through the above steps, this embodiment can realize real-time signal acquisition of current and voltage for further data processing and fault diagnosis.

[0072] Preferably, it also includes:

[0073] Using the remote server to provide the fault diagnosis result to the user and maintenance personnel of the target electric motorcycle through a user interface;

[0074] Store operating data and diagnostic results in a cloud database for future data analysis and pattern recognition.

[0075] Specifically, the operation data of this embodiment includes the following types of information:

[0076] Current data, real-time current value (unit: ampere, A), indicates the current consumption of the electric motorcycle under different working conditions; current fluctuation data reflects the change of current during the operation of the electric motorcycle.

[0077] Voltage data, real-time voltage value (unit: volt, V), including battery voltage and motor voltage. Voltage fluctuation data, reflecting the voltage change of the electric motorcycle under different loads.

[0078] Timestamp: a time stamp for each data collection to facilitate subsequent analysis and comparison.

[0079] Environmental data, environmental parameters such as temperature and humidity, may affect the operating performance of the electric motorcycle.

[0080] Furthermore, the diagnosis result of this embodiment is obtained by analyzing the operation data, including the following information:

[0081] Fault type, the category of fault identified, such as battery failure, motor failure, connection problem, etc.

[0082] Fault severity, an assessment of the severity of the fault, which may be classified as minor, moderate, or severe.

[0083] The time when the fault occurred and the time when the diagnostic result was generated make it easier to track the time when the fault occurred.

[0084] Recommended actions: maintenance suggestions or measures provided for the identified fault type, such as replacing parts, checking connections, etc.

[0085] Furthermore, the present embodiment stores the operation data and diagnostic results in a cloud database for future data analysis and pattern recognition. The data can be stored in time series for historical data comparison and trend analysis.

[0086] The fault diagnosis results and operating data are provided to the users and maintenance personnel of the electric motorcycle through a user interface (such as a mobile application or a web page). Users can view real-time data, historical data and fault diagnosis results, and take corresponding measures based on the recommendations.

[0087] Preferably, the method for constructing the fault diagnosis model includes:

[0088] Get the preset annotation dataset;

[0089] Standardizing and data enhancing the labeled data set to obtain an enhanced data set;

[0090] Using a deep convolutional generative adversarial network to perform image synthesis on the enhanced dataset to obtain a training dataset;

[0091] The training data set is input into the improved VGG network for training to obtain the fault diagnosis model; the VGG network includes: a convolution layer, a pooling layer, a fully connected layer and an output layer; the pooling layer is connected after each of the convolution layers; the output of the pooling layer is connected to the fully connected layer; the output of the fully connected layer is connected to the output layer.

[0092] Specifically, the data enhancement of this embodiment is as follows: flipping, rotating, cropping, color jittering and adding noise to the two-dimensional image, and using the deep convolutional generative adversarial network framework layer based on DCGAN to generate the image to obtain the enhanced data set.

[0093] Preferably, in the remote server, performing denoising preprocessing on the real-time signal to obtain a first preprocessed signal includes:

[0094] Performing multi-level decomposition on the real-time signal to obtain wavelet coefficients;

[0095] Performing threshold processing on the wavelet coefficients to obtain processed wavelet coefficients;

[0096] Signal reconstruction is performed on the processed wavelet coefficients to obtain the first preprocessed signal.

[0097] Preferably, thresholding is performed on the wavelet coefficients to obtain processed wavelet coefficients, including:

[0098] When the value of the wavelet coefficient is greater than or equal to the preset processing threshold, the wavelet coefficient is processed by using the first coefficient processing function to obtain the processed wavelet coefficient; wherein the first coefficient processing function is:

[0099]

[0100] Where u is the first adjustment coefficient, sgn() is the sign function, and wj,l is the lth wavelet coefficient under the jth decomposition level, λ is the processing threshold, α is the second adjustment coefficient, β is the third adjustment coefficient, γ is the fourth adjustment coefficient, p is the wavelet decomposition level, is the wavelet coefficient after processing;

[0101] When the value of the wavelet coefficient is less than the processing threshold, the wavelet coefficient is processed by a second coefficient processing function to obtain a processed wavelet coefficient; wherein the second coefficient processing function is:

[0102]

[0103] Specifically, this embodiment can effectively suppress the noise in the signal by thresholding the wavelet coefficients. Wavelet transform can decompose the signal into components of different frequencies, and the noise is usually concentrated in the high-frequency part. By setting a suitable processing threshold, the high-frequency noise coefficient can be suppressed to near zero, thereby improving the signal-to-noise ratio of the signal.

[0104] When processing wavelet coefficients, the first coefficient processing function is used to process coefficients greater than or equal to the processing threshold, and the main features of the signal can be retained. This is because these coefficients usually correspond to the effective components of the signal, rather than noise. Through appropriate adjustment coefficients, the degree of retention of signal features can be adjusted. The introduction of the first adjustment coefficient, the second adjustment coefficient, the third adjustment coefficient and the fourth adjustment coefficient makes the processing process flexible. Users can adjust these coefficients according to specific application scenarios and signal characteristics to achieve the best denoising effect and signal retention effect. And this embodiment can reduce the constant deviation between the original wavelet coefficients and the estimated wavelet coefficients by increasing the adjustment coefficient, so that the processing function can better denoise the noisy wavelet coefficients.

[0105] Preferably, the processing threshold calculation formula is:

[0106]

[0107] Wherein, λ is the processing threshold, δ is the standard deviation of Gaussian noise, M is the length of the real-time signal, and z is the preset coefficient.

[0108] Preferably, the first preprocessed signal is subjected to noise reduction preprocessing using an SSA-SGMD algorithm to obtain a second preprocessed signal, comprising:

[0109] The optimization parameters of symplectic geometric mode decomposition are optimized using the sparrow search algorithm to obtain the SSA-SGMD algorithm;

[0110] The first preprocessed signal is subjected to noise reduction preprocessing by using the SSA-SGMD algorithm to obtain a second preprocessed signal.

[0111] Preferably, after the first preprocessed signal is subjected to noise reduction preprocessing by adopting the SSA-SGMD algorithm to obtain the second preprocessed signal, the method further comprises:

[0112] The second preprocessed signal is segmented, the length of each segment is determined, and overlapping sampling is used to obtain the second preprocessed signal converted into a two-dimensional image.

[0113] Specifically, this embodiment can collect current and voltage signals in real time by installing current and voltage sensors at the incoming line of the motor drive board of the electric motorcycle. The data acquisition system will filter and extract features of these signals, and then analyze the signal characteristics through pattern recognition technology to determine whether there is an abnormality. Common faults include overcurrent (which may indicate motor overload or short circuit), undervoltage (which may indicate low battery or poor connection), overvoltage (which may indicate voltage regulator failure), and signal imbalance (which may indicate motor or controller failure). Once a fault is detected, the system will issue an alarm and record relevant data for subsequent analysis and maintenance.

[0114] The beneficial effects of the present invention are as follows:

[0115] (1) The present invention can monitor the working status of the vehicle in real time and detect potential faults in time by collecting real-time current and voltage signals of the electric motorcycle during operation, thereby improving safety.

[0116] (2) The present invention utilizes a wireless communication module to transmit information to a remote server in real time, thereby realizing remote fault diagnosis, reducing the time and cost of manual inspection, and improving maintenance efficiency.

[0117] (3) The present invention can significantly improve the quality of the signal through denoising preprocessing and noise reduction processing using the SSA-SGMD algorithm, ensuring the accuracy of subsequent analysis, which is very critical for accurate diagnosis.

[0118] (4) The present invention uses a two-dimensional feature image obtained by synchronous squeezing wavelet transform, which can better extract signal features and use the trained model for classification, so as to accurately identify and diagnose various fault types.

[0119] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0120] The principles and implementation methods of the present invention are described in this article using specific examples. The description of the above embodiments is only used to help understand the method and core idea of ​​the present invention. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. An online fault diagnosis method for electric motorcycles based on real-time current and voltage signals, characterized in that: include: Use current sensors and voltage sensors to collect real-time signals of incoming wires on the motor drive board when the target electric motorcycle is running; The collected real-time signals are transmitted to the remote server in real time through the wireless communication module; In the remote server, performing denoising preprocessing on the real-time signal to obtain a first preprocessed signal; Performing noise reduction preprocessing on the first preprocessed signal using an SSA-SGMD algorithm to obtain a second preprocessed signal; Processing the second preprocessed signal using synchronous squeezing wavelet transform to obtain the two-dimensional feature image; The trained fault diagnosis model is used to extract and classify the two-dimensional feature image to distinguish and diagnose various fault types and obtain a fault diagnosis result.

2. The electric motorcycle online fault diagnosis method based on real-time current and voltage signals according to claim 1 is characterized in that: Also includes: Using the remote server to provide the fault diagnosis result to the user and maintenance personnel of the target electric motorcycle through a user interface; Store operating data and diagnostic results in a cloud database for future data analysis and pattern recognition.

3. The electric motorcycle online fault diagnosis method based on real-time current and voltage signals according to claim 1 is characterized in that: The method for constructing the fault diagnosis model comprises: Get the preset annotation dataset; Standardizing and data enhancing the labeled data set to obtain an enhanced data set; Using a deep convolutional generative adversarial network to perform image synthesis on the enhanced dataset to obtain a training dataset; The training data set is input into the improved VGG network for training to obtain the fault diagnosis model; the VGG network includes: a convolution layer, a pooling layer, a fully connected layer and an output layer; the pooling layer is connected after each of the convolution layers; the output of the pooling layer is connected to the fully connected layer; the output of the fully connected layer is connected to the output layer.

4. The electric motorcycle online fault diagnosis method based on real-time current and voltage signals according to claim 1 is characterized in that: In the remote server, performing denoising preprocessing on the real-time signal to obtain a first preprocessed signal includes: Performing multi-level decomposition on the real-time signal to obtain wavelet coefficients; Performing threshold processing on the wavelet coefficients to obtain processed wavelet coefficients; Signal reconstruction is performed on the processed wavelet coefficients to obtain the first preprocessed signal.

5. The electric motorcycle online fault diagnosis method based on real-time current and voltage signals according to claim 4 is characterized in that: Thresholding the wavelet coefficients to obtain processed wavelet coefficients includes: When the value of the wavelet coefficient is greater than or equal to the preset processing threshold, the wavelet coefficient is processed by using the first coefficient processing function to obtain the processed wavelet coefficient; wherein the first coefficient processing function is: Where u is the first adjustment coefficient, sgn() is the sign function, and w j,l is the lth wavelet coefficient under the jth decomposition level, λ is the processing threshold, α is the second adjustment coefficient, β is the third adjustment coefficient, γ is the fourth adjustment coefficient, p is the wavelet decomposition level, is the wavelet coefficient after processing; When the value of the wavelet coefficient is less than the processing threshold, the wavelet coefficient is processed by a second coefficient processing function to obtain a processed wavelet coefficient; wherein the second coefficient processing function is:

6. The electric motorcycle online fault diagnosis method based on real-time current and voltage signals according to claim 5 is characterized in that: The processing threshold calculation formula is: Wherein, λ is the processing threshold, δ is the standard deviation of Gaussian noise, M is the length of the real-time signal, and z is the preset coefficient.

7. The electric motorcycle online fault diagnosis method based on real-time current and voltage signals according to claim 1 is characterized in that: The first preprocessed signal is subjected to noise reduction preprocessing by using an SSA-SGMD algorithm to obtain a second preprocessed signal, including: The optimization parameters of symplectic geometric mode decomposition are optimized using the sparrow search algorithm to obtain the SSA-SGMD algorithm; The first preprocessed signal is subjected to noise reduction preprocessing by using the SSA-SGMD algorithm to obtain a second preprocessed signal.

8. The electric motorcycle online fault diagnosis method based on real-time current and voltage signals according to claim 7 is characterized in that: After the first preprocessed signal is subjected to noise reduction preprocessing by using the SSA-SGMD algorithm to obtain a second preprocessed signal, the method further includes: The second preprocessed signal is segmented, the length of each segment is determined, and overlapping sampling is used to obtain the second preprocessed signal converted into a two-dimensional image.

Images