MOSFET fault diagnosis method and system based on drive current data

Abstract

The application discloses a MOSFET fault diagnosis method and system based on driving current data, and relates to the technical field of device fault diagnosis. The method comprises the following steps: first, acquiring driving current data of a MOSFET power driving circuit during operation; then, decomposing the driving current signal by using a variational mode decomposition algorithm optimized by a grey wolf optimization algorithm in advance, so as to obtain a plurality of intrinsic mode components; subsequently, calculating the energy of each intrinsic mode component, selecting several intrinsic mode components with maximum energy, and calculating the normalized energy value of the driving current data; finally, outputting the fault type of the MOSFET power driving circuit by using a pre-trained fault diagnosis model. The above scheme can efficiently and accurately realize the judgment and identification of the fault and performance degradation mode of the MOSFET power driving circuit.

Description

Technical Field

[0001] This application relates to the field of electrical device fault diagnosis technology, and in particular to a MOSFET fault diagnosis method and system based on drive current data. Background Technology

[0002] MOSFETs (Metal-Oxide-Semiconductor Field-Effect Transistors) are widely used in various high-power electronic drive systems due to their excellent switching speed, low input current, low on-resistance, and strong overcurrent capability. These systems include motor H-bridge drive circuits and dual-voltage time-sharing drive circuits for fuel systems. In these applications, electronic components operate at extremely high frequencies and in harsh environments (high voltage and current with sudden changes). Over long-term operation, they are susceptible to failure and performance degradation. Both failures and performance degradation can affect the performance and safety of motors and fuel systems. Therefore, it is necessary to effectively diagnose MOSFET failures and performance degradation. Summary of the Invention

[0003] The purpose of this application is to provide a MOSFET fault diagnosis method and system based on drive current data, which can efficiently and accurately determine and identify faults and performance degradation modes of MOSFET power drive circuits.

[0004] To achieve the above objectives, this application provides the following solution:

[0005] In a first aspect, this application provides a MOSFET fault diagnosis method based on drive current data, comprising the following steps:

[0006] Obtain the drive current data when the MOSFET power drive circuit is working.

[0007] The driving current signal is decomposed using an optimized variational mode decomposition algorithm to obtain several intrinsic mode components; the optimized variational mode decomposition algorithm is a variational mode decomposition algorithm optimized by the Grey Wolf optimization algorithm for the number of decomposition layers and the penalty factor.

[0008] The energy of each intrinsic mode component is calculated, and the normalized energy value of the driving current data is obtained by selecting several intrinsic mode components with the largest energy.

[0009] The normalized energy value is input into a pre-trained fault diagnosis model to obtain the fault type of the MOSFET power drive circuit; the pre-trained fault diagnosis model is a pre-trained convolutional neural network.

[0010] Optionally, before decomposing the driving current signal using the optimized variational mode decomposition algorithm to obtain several intrinsic mode components, the following steps are also included:

[0011] The Grey Wolf optimization algorithm is used to optimize the number of decomposition levels and the penalty factor of the variational mode decomposition algorithm, resulting in an optimized variational mode decomposition algorithm. The specific steps include:

[0012] Initialize the gray wolf population and coefficient vector; the gray wolf population includes several gray wolves, each gray wolf corresponding to a set of decomposition layers and penalty factors.

[0013] Calculate the fitness value of each gray wolf, and select the three gray wolves with the highest fitness values ​​as target gray wolves.

[0014] For any gray wolf in the gray wolf population other than the target gray wolf, calculate the distance of the gray wolf relative to the three target gray wolves.

[0015] Based on the distance between the gray wolf and the three target gray wolves and the coefficient vector, the gray wolf is controlled to move in the direction of the three target gray wolves.

[0016] Update the coefficient vector and calculate the fitness value for each gray wolf.

[0017] Determine if the maximum number of iterations has been reached.

[0018] If so, then the decomposition level and penalty factor of the gray wolf with the highest fitness value are taken as the optimal decomposition level and optimal penalty factor.

[0019] If not, then the three gray wolves with the highest fitness values ​​will be selected as target gray wolves, and the process will proceed to step "Calculate the distance of any gray wolf in the gray wolf population other than the target gray wolves relative to the three target gray wolves".

[0020] Optionally, the fitness value of each gray wolf can be calculated according to the following formula:

[0021]

[0022] Where a(j) is the envelope signal obtained after Hilbert transform of the original driving current data, p j The result is the normalized envelope signal, where N is the length of the time series of the original drive current data, and E... p This is the envelope entropy, which is the fitness value.

[0023] Optionally, the distance between the gray wolf and the three target gray wolves can be calculated using the following formula:

[0024]

[0025] Where α, β, and δ represent the labels of the three target gray wolves, and Let α, β, and δ represent the distances of the gray wolf relative to the target gray wolf, respectively. This is the coefficient vector used in this round of optimization. and These are the locations of the three target gray wolves. This indicates the location of the gray wolf.

[0026] Optionally, the position of the gray wolf after moving towards the three target gray wolves can be calculated according to the following formula:

[0027]

[0028] in, Let represent the direction and distance the gray wolf walks towards the target gray wolf α. The direction and distance that the gray wolf moves toward the target gray wolf β. Let δ represent the direction and distance that the gray wolf will move towards the target gray wolf. and These are the coefficient vectors used in this round of optimization. This indicates the position of the gray wolf after it has moved towards the direction of the three target gray wolves.

[0029] Optionally, the coefficient vector is updated according to the following formula:

[0030]

[0031] in, and These are all coefficient vectors used in the Grey Wolf optimization algorithm. For control vectors, Decreasing linearly from 2 to 0, and It is a random vector in the range [0,1].

[0032] Optionally, before inputting the normalized energy value into a pre-trained fault diagnosis model to obtain the fault type of the MOSFET power drive circuit, the following steps are also included:

[0033] Hard faults in MOSFET power drive circuits are simulated through experiments and simulations, and drive current data under hard fault conditions are obtained.

[0034] The aging process of components is accelerated by using a high and low temperature test chamber to simulate soft faults in MOSFET power drive circuits and obtain drive current data under soft fault conditions.

[0035] The optimized variational mode decomposition algorithm is used to decompose each of the driving current signals to obtain several intrinsic mode components.

[0036] The energy of each intrinsic mode component is calculated, and the normalized energy value of the driving current data is obtained by selecting several intrinsic mode components with the largest energy.

[0037] A fault diagnosis dataset is constructed based on the normalized energy value of each drive current data and the fault type corresponding to each drive current data.

[0038] Using the fault diagnosis dataset, with the normalized energy value of the driving current data as input and the fault type corresponding to the driving current data as the target output, a convolutional neural network is trained to obtain a pre-trained fault diagnosis model.

[0039] Optionally, the convolutional neural network includes an input layer, two pairs of alternating and corresponding convolutional layers, a pooling layer, a fully connected layer, and an output layer.

[0040] Optionally, the energy of the intrinsic mode components can be calculated according to the following formula:

[0041]

[0042] Among them, E j Let u be the energy of the j-th eigenmode component. j (t) represents the j-th intrinsic mode component.

[0043] Secondly, this application provides a MOSFET fault diagnosis system based on drive current data, including the following modules:

[0044] The drive current data acquisition module is used to acquire drive current data when the MOSFET power drive circuit is working.

[0045] The drive current data decomposition module is used to decompose the drive current signal using an optimized variational mode decomposition algorithm to obtain several intrinsic mode components; the optimized variational mode decomposition algorithm is a variational mode decomposition algorithm optimized with the gray wolf optimization algorithm for the number of decomposition layers and the penalty factor.

[0046] The normalized energy value calculation module is used to calculate the energy of each intrinsic mode component, and select the intrinsic mode components with the largest energy to calculate the normalized energy value of the driving current data.

[0047] The fault type diagnosis module is used to input the normalized energy value into a pre-trained fault diagnosis model to obtain the fault type of the MOSFET power drive circuit; the pre-trained fault diagnosis model is a pre-trained convolutional neural network.

[0048] According to the specific embodiments provided in this application, the following technical effects are disclosed:

[0049] This application provides a MOSFET fault diagnosis method and system based on drive current data. The method first acquires the drive current data of the MOSFET power drive circuit during operation. Then, it decomposes the drive current signal using a variational mode decomposition algorithm (VMD) optimized with a gray wolf optimization algorithm and a penalty factor, obtaining several intrinsic mode components (EMS). Subsequently, the energy of each EMS is calculated, and the normalized energy values ​​of the drive current data are calculated from the EMS components with the highest energy. Finally, a pre-trained fault diagnosis model is used to output the determined fault type of the MOSFET power drive circuit. The above-mentioned scheme of this application uses a variational mode decomposition algorithm optimized with a gray wolf optimization algorithm and a penalty factor to decompose the drive current data and obtain EMS, which avoids misjudgments and omissions due to human experience. Furthermore, the normalized energy values ​​of the EMS are used as feature vectors of the neural network for fault type diagnosis, avoiding the direct input of EMS into the neural network, which significantly increases computational complexity and training time, leading to problems such as the curse of dimensionality and overfitting. This achieves efficient and accurate judgment and identification of faults and performance degradation modes in MOSFET power drive circuits. Attached Figure Description

[0050] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0051] Figure 1 This is a flowchart of a MOSFET fault diagnosis method based on drive current data, provided as an embodiment of this application.

[0052] Figure 2 This is a flowchart of step A1 in a MOSFET fault diagnosis method based on drive current data, provided in an embodiment of this application.

[0053] Figure 3 This is a flowchart illustrating the construction of a dataset and training of a convolutional neural network in a MOSFET fault diagnosis method based on drive current data, provided in one embodiment of this application.

[0054] Figure 4 This is a schematic diagram of a dual-voltage time-sharing drive circuit used in a MOSFET fault diagnosis method based on drive current data, provided in another embodiment of this application.

[0055] Figure 5A schematic diagram of the current waveform of a dual-voltage time-sharing drive circuit used in a MOSFET fault diagnosis method based on drive current data, provided in another embodiment of this application.

[0056] Figure 6 This is a schematic diagram of the current waveform after a high-side bootstrap floating drive fault occurs in a MOSFET fault diagnosis method based on drive current data, provided in another embodiment of this application.

[0057] Figure 7 This is a schematic diagram of two fault modes of MOSFET1 and the current waveform after the fault in a MOSFET fault diagnosis method based on drive current data provided in another embodiment of this application.

[0058] Figure 8 This is a schematic diagram of two fault modes of MOSFET2 and the current waveform after the fault in a MOSFET fault diagnosis method based on drive current data provided in another embodiment of this application.

[0059] Figure 9 This is a schematic diagram of two fault modes of MOSFET3 and the current waveform after the fault in a MOSFET fault diagnosis method based on drive current data provided in another embodiment of this application.

[0060] Figure 10 This is a schematic diagram of an accelerated aging test platform used in a MOSFET fault diagnosis method based on drive current data, provided in another embodiment of this application.

[0061] Figure 11 This is a schematic diagram of the data measurement point distribution of a dual-voltage time-sharing drive circuit in a MOSFET fault diagnosis method based on drive current data, provided in another embodiment of this application.

[0062] Figure 12 This is a schematic diagram of the functional modules of a MOSFET fault diagnosis system based on drive current data, provided in an embodiment of this application.

[0063] Figure 13 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0064] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0065] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0066] In one exemplary embodiment, such as Figure 1 As shown, a MOSFET fault diagnosis method based on drive current data is provided, including the following steps:

[0067] S1. Obtain the drive current data when the MOSFET power drive circuit is working.

[0068] S2. The driving current signal is decomposed using an optimized variational mode decomposition algorithm to obtain several intrinsic mode components; the optimized variational mode decomposition algorithm is a variational mode decomposition algorithm optimized by the gray wolf optimization algorithm for the number of decomposition layers and the penalty factor.

[0069] S3. Calculate the energy of each intrinsic mode component, and select the intrinsic mode components with the largest energy to calculate the normalized energy value of the driving current data.

[0070] S4. Input the normalized energy value into the pre-trained fault diagnosis model to obtain the fault type of the MOSFET power drive circuit; the pre-trained fault diagnosis model is a pre-trained convolutional neural network.

[0071] Specifically, in this embodiment, before decomposing the drive current signal using the optimized variational mode decomposition algorithm in step S2 to obtain several intrinsic mode components, the MOSFET fault diagnosis method based on drive current data further includes the following steps:

[0072] A1. The Grey Wolf optimization algorithm is used to optimize the number of decomposition levels and the penalty factor of the variational mode decomposition algorithm, resulting in an optimized variational mode decomposition algorithm; such as Figure 2 The flowchart shown illustrates that, in this embodiment, step A1 specifically includes the following steps:

[0073] A11. Initialize the gray wolf population and coefficient vector; the gray wolf population includes several gray wolves, each gray wolf corresponding to a set of decomposition layers and penalty factors.

[0074] A12. Calculate the fitness value of each gray wolf, and select the three gray wolves with the highest fitness values ​​as target gray wolves. In this embodiment, the fitness value of each gray wolf is calculated according to the following formula:

[0075]

[0076] Where a(j) is the envelope signal obtained after Hilbert transform of the original driving current data, p jThe result is the normalized envelope signal, where N is the length of the time series of the original drive current data, and E... p This is the envelope entropy, which is the fitness value.

[0077] A13. For any gray wolf in the gray wolf population other than the target gray wolf, calculate the distance of the gray wolf relative to the three target gray wolves. In this embodiment, the distance of the gray wolf relative to the three target gray wolves is calculated according to the following formula:

[0078]

[0079] Where α, β, and δ represent the labels of the three target gray wolves, and Let α, β, and δ represent the distances of the gray wolf relative to the target gray wolf, respectively. This is the coefficient vector used in this round of optimization. and These are the locations of the three target gray wolves. This indicates the location of the gray wolf.

[0080] A14. Based on the distance and coefficient vector of the gray wolf relative to the three target gray wolves, control the gray wolf to move towards the three target gray wolves. In this embodiment, the position of the gray wolf after moving towards the three target gray wolves is calculated according to the following formula:

[0081]

[0082] in, Let represent the direction and distance the gray wolf walks towards the target gray wolf α. The direction and distance that the gray wolf moves toward the target gray wolf β. Let δ represent the direction and distance that the gray wolf will move towards the target gray wolf. and These are the coefficient vectors used in this round of optimization. This indicates the position of the gray wolf after it has moved towards the direction of the three target gray wolves.

[0083] A15. Update the coefficient vector and calculate the fitness value for each gray wolf. In this embodiment, the coefficient vector is updated according to the following formula:

[0084]

[0085] in, and These are all coefficient vectors used in the Grey Wolf optimization algorithm. For control vectors, Decreasing linearly from 2 to 0, and It is a random vector in the range [0,1].

[0086] A16. Determine if the maximum number of iterations has been reached. If yes, proceed to step A17; otherwise, proceed to step A18.

[0087] A17. The number of decomposition layers and the penalty factor of the gray wolf with the highest fitness value are taken as the optimal number of decomposition layers and the optimal penalty factor.

[0088] A18. Select the three gray wolves with the highest fitness values ​​as target gray wolves, and jump to step A13 "Calculate the distance of any gray wolf in the gray wolf population other than the target gray wolves relative to the three target gray wolves".

[0089] Specifically, in this embodiment, before inputting the normalized energy value into the pre-trained fault diagnosis model in step S4 to obtain the fault type of the MOSFET power drive circuit, the MOSFET fault diagnosis method based on drive current data further includes the process of constructing a dataset and training a convolutional neural network, such as... Figure 3 The flowchart shown includes the following steps:

[0090] B1. Simulate hard faults in the MOSFET power drive circuit through experiments and simulations, and obtain drive current data under hard fault conditions.

[0091] B2. Accelerate the aging process of components through a high and low temperature test chamber, simulate soft faults in MOSFET power drive circuits, and obtain drive current data under soft fault conditions.

[0092] B3. The optimized variational mode decomposition algorithm is used to decompose each driving current signal to obtain several intrinsic mode components.

[0093] B4. Calculate the energy of each intrinsic mode component, and select the intrinsic mode components with the highest energy to calculate the normalized energy value of the drive current data. In this embodiment, the energy of the intrinsic mode components is calculated according to the following formula:

[0094]

[0095] Among them, E j Let u be the energy of the j-th eigenmode component. j (t) represents the j-th intrinsic mode component.

[0096] B5. Construct a fault diagnosis dataset based on the normalized energy value of each drive current data and the fault type corresponding to each drive current data.

[0097] B6. Using a fault diagnosis dataset, with the normalized energy value of the driving current data as input and the fault type corresponding to the driving current data as the target output, a convolutional neural network is trained to obtain a pre-trained fault diagnosis model. Specifically, in this embodiment, the convolutional neural network includes an input layer, two pairs of alternating and corresponding convolutional layers, pooling layers, fully connected layers, and an output layer.

[0098] Next, to facilitate the explanation of the improved route of the MOSFET fault diagnosis method based on drive current data provided in this embodiment, a typical MOSFET drive circuit, such as the dual-voltage time-sharing drive circuit in the fuel system drive circuit, will be used as an example. This circuit is relatively complex and can fully illustrate the technical solution and effect of this application. However, it is understood that the application scope of the method of this application is not limited to this and can also be applied to all other MOSFET power drive circuits.

[0099] The specific technical solution will be explained in conjunction with the dual-voltage time-sharing drive circuit. A schematic diagram of the dual-voltage time-sharing drive circuit for the fuel system drive circuit is shown below. Figure 4 As shown, D2, D3, and D4 are all diodes. Diode D2 prevents the high-voltage source from charging the low-voltage source. Diode D3 allows current to flow freely from GND. Diode D4 prevents the high-voltage source from charging the area between the solenoid valve and MOSFET3. The current waveform is shown below. Figure 5 As shown, where I peak_max I represents the maximum peak current. peak_min I represents the minimum peak current. hold_max Indicates the maximum holding current, I hold_min This indicates that the minimum current is maintained.

[0100] Before performing specific fault diagnosis, it is necessary to first train the convolutional neural network model to be used. Before training, a fault diagnosis dataset needs to be constructed. The steps for constructing the fault diagnosis dataset include:

[0101] 1) Acquisition of original fault and performance degradation data.

[0102] The main fault modes of MOSFET power drive circuits include structural faults (hard faults) and parameter faults (soft faults). Hard faults severely affect the drive current waveform. By simulating hard fault conditions through experiments and simulations, current waveform data when hard faults occur can be obtained.

[0103] 1.1) Acquisition of raw data under hard faults.

[0104] Hard faults are categorized into high-side bootstrap floating drive faults, MOSFET open-circuit faults, and MOSFET short-circuit faults. For high-side bootstrap floating drive faults and MOSFET open-circuit faults, the drive current waveform is smaller than normal, so it can be simulated experimentally. However, for MOSFET short-circuit faults, the drive current waveform is larger than normal, making it impossible to simulate experimentally; therefore, simulation is used. For high-side bootstrap floating drive faults, the current waveform after the fault occurs is as follows: Figure 6 As shown, for MOSFET open-circuit faults and MOSFET short-circuit faults, the fault modes and post-fault current waveforms of the three MOSFET devices are respectively as follows: Figures 7-9 As shown.

[0105] 1.2) Acquisition of raw data under soft faults.

[0106] Soft failures refer to component aging caused by MOSFETs during long-term use. Under normal use, the aging process is relatively slow; therefore, this embodiment designs an accelerated aging test method for MOSFETs. For semiconductor chips, to increase the degradation rate and shorten the aging time, it is necessary to increase the chip junction temperature T0 as much as possible. j And the chip junction temperature T j Since it is difficult to measure directly, this embodiment uses the following formula to approximate the chip junction temperature T. j :

[0107] T j =T c +ψ JT P D .

[0108] Among them, T j T is the junction temperature of a semiconductor device. c For ambient temperature, ψ JT P represents the thermal characteristic parameter between the node and the center of the package upper surface. Its magnitude is related to the intrinsic properties of the semiconductor device. D This refers to the power consumption of the package.

[0109] Therefore, in this embodiment, T is increased by raising the ambient temperature and increasing the operating frequency of the MOSFET, respectively. c With P D This achieves the goal of accelerating the aging of components. In a further improved embodiment, the following can be used: Figure 10 The accelerated aging test platform shown has data measurement points distributed as follows: Figure 11As shown in the figure, ①②③ represent the locations of three data measurement points. The operating procedure includes: setting the high and low temperature test chamber to a high temperature relative to the standard state, such as 85℃, setting a high-frequency drive operating frequency, and actively cutting off the power supply of the high and low temperature test chamber after every 500,000 power drives, allowing it to cool down to room temperature, and then reheating it to 85℃, thereby accelerating all bonding and interface degradation between the semiconductor chip and the mounting surface again through active temperature cycling.

[0110] 2) Extraction of current characteristic signals.

[0111] The driving current signal is essentially a time-domain signal. When a hard or soft fault occurs in the driving circuit, the response of the current signal flowing through the actuator will also change during the current abrupt change process. This change is particularly noticeable in the frequency domain. In this application, the time series of the current signal flowing through the solenoid valve is considered a non-stationary signal. Based on the center frequency, the original signal is decomposed into K IMF components under the optimization condition of minimizing the bandwidth of K IMF components, as shown in the following expression:

[0112]

[0113] Among them, u k The modal component sequence is obtained after VMD decomposition, k = 1, 2, ..., K; ω k The center frequency corresponding to each modal component; Let be the partial derivative of the function with respect to time; δ(t) is the unit impulse function; e -jωt is a complex exponential signal; f is the original drive current signal.

[0114] To find the optimal solution to the above variational problem, a quadratic penalty term α and a Lagrange multiplier λ are introduced. The quadratic penalty term constrains the smoothness of the signal decomposition during the optimization process, avoiding over-decomposition and oscillations. The Lagrange multiplier ensures that certain conditions and constraints are met in the signal decomposition. The introduction of these two factors transforms the constrained variational problem into an unconstrained variational problem.

[0115] In the Variational Mode Decomposition (VMD) algorithm described above, the setting of the decomposition level K and the penalty factor α affects the VMD performance. Improper selection can lead to information loss or mode aliasing, and the parameter selection is often irregular. However, in hardware and software fault identification, there are many types of faults and a large number of samples. Determining parameters based on experience is time-consuming and lacks evaluation criteria. Therefore, this embodiment introduces the Grey Wolf Optimizer (GWO) to accelerate and optimize VMD. The Grey Wolf Optimizer (GWO) is a swarm intelligence optimization algorithm that simulates the social behavior and hierarchical structure of a grey wolf pack to solve optimization problems. Specifically, the optimization process is as follows:

[0116] Initially, n vectors are randomly generated within a defined allowable range, representing n gray wolves. The fitness function value of each individual in the population is calculated. The three gray wolves with the best fitness are labeled as α wolf, β wolf, and δ wolf, respectively, and the remaining wolves are labeled as ω wolf. Since the location of the optimal solution is unknown, α wolf, β wolf, and δ wolf are temporarily considered to represent the potential location of the optimal solution. According to the above hierarchy, higher-ranking gray wolves have absolute dominance over lower-ranking gray wolves. For each ω wolf, the distance of ω wolf relative to α wolf, β wolf, and δ wolf is calculated using the following formula:

[0117]

[0118] in, Let this be the current position vector of the gray wolf. This represents the approximate distance from the current gray wolf to the α wolf, β wolf, or δ wolf. Let be the coefficient vector. This process uses the different positions of α wolf, β wolf, and δ wolf to predict the position of the prey, and then changes the position of ω wolf according to the distance calculated above, as shown in the following formula:

[0119]

[0120] in, Let ω represent the direction and distance that wolf ω moves toward wolf α, wolf β, and wolf δ. This represents the final position of wolf ω. (Coefficient vector) The update equation is shown below:

[0121]

[0122] in, Decreasing linearly from 2 to 0, The direction of the gray wolf's movement is randomized to a certain extent, which can help avoid the occurrence of local optima.

[0123] The fitness function used in the above optimization process is selected as follows. During parameter optimization, the evaluation criterion for the decomposition effect of the VMD method is the envelope entropy E proposed by Tang Guiji et al. p The envelope entropy is defined as follows:

[0124]

[0125] Where a(j) is the envelope signal obtained after Hilbert transform of the original driving current data time series f(j), and p j This is the result after normalization. If the obtained components still contain high-frequency noise, the sparsity of the component signal will be weakened, the envelope entropy will be relatively large, and the extracted fault information will be submerged in the noise signal. Therefore, the fitness function for optimizing the VMD parameters is set to minE.p .

[0126] 3) Selection of feature vectors.

[0127] The original signal, after VMD decomposition, yields several intrinsic mode components (IMFs). Each IMF component consists of thousands of data points. Directly inputting these into a neural network would significantly increase computational complexity and training time, leading to the curse of dimensionality and overfitting. Therefore, in this embodiment, the energy value of each decomposed component is first calculated using the following formula:

[0128]

[0129] Among them, u j (t) is the signal sequence of intrinsic mode components after VMD decomposition.

[0130] Then, take the normalized energy value Q of the first four IMF components according to the following formula. * As a feature vector input to the neural network:

[0131]

[0132] That is, the values ​​of the four components E1 to E4 are divided by the total energy value to obtain the normalized energy value of each of the four components, and the resulting Q is... 1 Q 2 Q 3 Q 4 As a feature vector input to the neural network.

[0133] In summary, a fault diagnosis dataset is constructed based on the normalized energy values ​​of each drive current data point and the corresponding fault types. The normalized energy values ​​serve as input data, and the corresponding fault types serve as the model labels.

[0134] 4) Training of convolutional neural networks.

[0135] In this embodiment, a convolutional neural network (CNN) is used for fault and performance degradation judgment. The CNN consists of five parts: an input layer, two pairs of alternating and corresponding convolutional layers, a pooling layer, a fully connected layer, and an output layer. The convolutional layers use convolutional kernels to extract feature parameters. To prevent feature parameter loss, max pooling is used to reduce the dimensionality of the high-dimensional data after convolution. During the training of the CNN using the fault diagnosis dataset, a learning rate reduction method is used to prevent overfitting. The fully connected layer uses the Softmax function as the activation function of the output layer and Adam as the optimizer, which can effectively improve the training speed, convergence effect, and generalization ability of the model. The training process of the neural network is relatively conventional and will not be described in detail here. After the model training is completed, it can be applied to the MOSFET fault diagnosis method based on drive current data provided in the above embodiment. Through steps S1 to S4 of the above method, the diagnosis of MOSFET fault modes based on drive current data is completed.

[0136] The MOSFET fault diagnosis method based on drive current data proposed in this application optimizes the parameters of the variational mode decomposition algorithm using the Grey Wolf optimization algorithm, obtains the intrinsic mode components based on the variational mode decomposition, and calculates the energy values ​​of the intrinsic mode components as feature vectors of a CNN for fault discrimination. This avoids misjudgments and omissions due to human experience and enables the judgment and identification of short-circuit, open-circuit, and performance degradation modes of dual-voltage time-division drive circuits. Specifically, it has the following advantages:

[0137] (1) The driving current signal is essentially a time-domain signal. When a hard or soft fault occurs in the driving circuit, the response of the current signal flowing through the actuator will also change during the current change process. This change is particularly obvious in the frequency domain. This application starts from the time-domain signal, extracts the driving current data under the two major fault types of hard and soft faults, and extracts the characteristic parameters in the frequency domain from the current time-domain signal, thus avoiding the problem of easy interference based on the time-domain signal.

[0138] (2) This application optimizes the feature extraction method. In the VMD calculation process, the setting of the decomposition layer number K and the penalty factor α will affect the VMD effect. Improper selection will lead to information loss or modal aliasing, and the selection of parameters is irregular. However, in soft and hard fault discrimination, there are many types of faults and a large number of samples. The method of determining parameters based on experience is time-consuming and lacks evaluation criteria. Therefore, the Grey Wolf algorithm is introduced to accelerate and optimize VMD.

[0139] (3) During feature extraction optimization, if the obtained components still contain high-frequency noise, the sparsity of the component signal will be weakened, the envelope entropy will be relatively large, and the extracted fault information will be submerged in the noise signal. Therefore, the fitness function for optimizing VMD parameters is set to minE. p .

[0140] (4) The original driving current signal can be decomposed into several intrinsic mode components (IMFs) by VMD. Since a single IMF component consists of thousands of data points, directly inputting it into a neural network would greatly increase the computational complexity and training time, leading to the curse of dimensionality and overfitting. Therefore, this application first calculates the energy value of each component after decomposition and takes the normalized energy value Q of the first four IMF components. * As a feature variable input to the neural network, it avoids the aforementioned problems.

[0141] Based on the same inventive concept, this application also provides a system for implementing the MOSFET fault diagnosis method based on drive current data as described above. The solution provided by this system is similar to the implementation described in the above method; therefore, the specific limitations in one or more system embodiments provided below can be found in the limitations of the MOSFET fault diagnosis method based on drive current data described above, and will not be repeated here.

[0142] In one exemplary embodiment, such as Figure 12 As shown, a MOSFET fault diagnosis system based on drive current data is provided, including the following modules:

[0143] The drive current data acquisition module is used to acquire drive current data when the MOSFET power drive circuit is working.

[0144] The drive current data decomposition module is used to decompose the drive current signal using an optimized variational mode decomposition algorithm to obtain several intrinsic mode components; the optimized variational mode decomposition algorithm is a variational mode decomposition algorithm optimized with the gray wolf optimization algorithm for the number of decomposition layers and the penalty factor.

[0145] The normalized energy value calculation module is used to calculate the energy of each intrinsic mode component, and select the intrinsic mode components with the largest energy to calculate the normalized energy value of the driving current data.

[0146] The fault type diagnosis module is used to input the normalized energy value into a pre-trained fault diagnosis model to obtain the fault type of the MOSFET power drive circuit; the pre-trained fault diagnosis model is a pre-trained convolutional neural network.

[0147] certainly, Figure 12 The architecture shown is merely exemplary; it can be omitted as needed when implementing different functionalities. Figure 12 One or at least two components of the system shown.

[0148] In one exemplary embodiment, a computer device is provided, which may be a server or a terminal, and its internal structure diagram may be as follows. Figure 13 As shown, the computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communicating with external terminals via a network connection. When the computer program is executed by the processor, it implements the MOSFET fault diagnosis method based on drive current data provided in the above embodiment.

[0149] Those skilled in the art will understand that Figure 13 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0150] In one exemplary embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0151] In one exemplary embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0152] In one exemplary embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0153] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0154] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0155] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0156] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0157] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A MOSFET fault diagnosis method based on drive current data, characterized in that, include: Obtain the drive current data when the MOSFET power drive circuit is working; The driving current signal is decomposed using an optimized variational mode decomposition algorithm to obtain several intrinsic mode components; the optimized variational mode decomposition algorithm is a variational mode decomposition algorithm optimized by the Grey Wolf optimization algorithm for the number of decomposition layers and the penalty factor; Calculate the energy of each intrinsic mode component, and select the intrinsic mode components with the largest energy to calculate the normalized energy value of the driving current data; The normalized energy value is input into a pre-trained fault diagnosis model to obtain the fault type of the MOSFET power drive circuit; the pre-trained fault diagnosis model is a pre-trained convolutional neural network. The pre-trained fault diagnosis model is obtained through the following steps: Hard faults in MOSFET power drive circuits are simulated through experiments and simulations, and drive current data under hard fault conditions are obtained. The aging process of components is accelerated by using a high and low temperature test chamber to simulate soft faults in MOSFET power drive circuits and obtain drive current data under soft fault conditions. The optimized variational mode decomposition algorithm is used to decompose each of the driving current signals to obtain several intrinsic mode components; Calculate the energy of each intrinsic mode component, and select the intrinsic mode components with the largest energy to calculate the normalized energy value of the driving current data; A fault diagnosis dataset is constructed based on the normalized energy value of each drive current data and the fault type corresponding to each drive current data. Using the fault diagnosis dataset, with the normalized energy value of the driving current data as input and the fault type corresponding to the driving current data as the target output, a convolutional neural network is trained to obtain a pre-trained fault diagnosis model. The energy of the intrinsic modal components can be calculated using the following formula: ； in, For the first j The energy of each intrinsic mode component For the first j The four eigenmode components with the highest energy are selected, and the normalized energy value of the driving current data is calculated according to the following formula: ； in, E 1. E 2. E 3. E 4 represents the energies of the four eigenmode components with the highest energy; E 1. E 2. E 3. E 4. Divide the values ​​of these four components by the total energy value to obtain the normalized energy values ​​for each of the four components. Q 1 , Q 2 , Q 3 , Q 4 As a feature vector input to the convolutional neural network.

2. The MOSFET fault diagnosis method based on drive current data according to claim 1, characterized in that, Before decomposing the driving current signal using the optimized variational mode decomposition algorithm to obtain several intrinsic mode components, the following steps are also included: The Grey Wolf optimization algorithm is used to optimize the number of decomposition levels and the penalty factor of the variational mode decomposition algorithm, resulting in an optimized variational mode decomposition algorithm; specifically including: Initialize the gray wolf population and coefficient vector; the gray wolf population includes several gray wolves, each gray wolf corresponding to a set of decomposition layers and penalty factors; Calculate the fitness value of each gray wolf, and select the three gray wolves with the highest fitness values ​​as target gray wolves; For any gray wolf in the gray wolf population other than the target gray wolf, calculate the distance of the gray wolf relative to the three target gray wolves; Based on the distance of the gray wolf relative to the three target gray wolves and the coefficient vector, the gray wolf is controlled to move in the direction of the three target gray wolves; Update the coefficient vector and calculate the fitness value for each gray wolf; Determine if the maximum number of iterations has been reached. If so, then the number of decomposition layers and the penalty factor of the gray wolf with the highest fitness value are taken as the optimal number of decomposition layers and the optimal penalty factor; If not, then the three gray wolves with the highest fitness values ​​will be selected as target gray wolves, and the process will proceed to step "Calculate the distance of any gray wolf in the gray wolf population other than the target gray wolves relative to the three target gray wolves".

3. The MOSFET fault diagnosis method based on drive current data according to claim 2, characterized in that, The distance between the gray wolf and the three target gray wolves is calculated using the following formula: ； in, α , β , δ These are the labels representing the three target gray wolves. , and They are gray wolves relative to the target gray wolf. α , β , δ distance, This is the coefficient vector used in this round of optimization. , and These are the locations of the three target gray wolves. This indicates the location of the gray wolf.

4. The MOSFET fault diagnosis method based on drive current data according to claim 3, characterized in that, The position of the gray wolf after moving towards the three target gray wolves is calculated using the following formula: ； in, For the gray wolf to target the gray wolf α The direction and distance of the step. For the gray wolf to target the gray wolf β The direction and distance of the step. For the gray wolf to target the gray wolf δ The direction and distance of the step. , and These are the coefficient vectors used in this round of optimization. This indicates the position of the gray wolf after it has moved towards the direction of the three target gray wolves.

5. The MOSFET fault diagnosis method based on drive current data according to claim 4, characterized in that, The coefficient vector is updated according to the following formula: ； in, and These are all coefficient vectors used in the Grey Wolf optimization algorithm. For control vectors, Decreasing linearly from 2 to 0, and It is a random vector in the range [0,1].

6. The MOSFET fault diagnosis method based on drive current data according to claim 1, characterized in that, The convolutional neural network includes an input layer, two pairs of alternating and corresponding convolutional layers, a pooling layer, a fully connected layer, and an output layer.

7. A MOSFET fault diagnosis system based on drive current data, characterized in that, For implementing the MOSFET fault diagnosis method based on drive current data as described in any one of claims 1-6, the MOSFET fault diagnosis system based on drive current data comprises: The drive current data acquisition module is used to acquire drive current data when the MOSFET power drive circuit is working. The drive current data decomposition module is used to decompose the drive current signal using an optimized variational mode decomposition algorithm to obtain several intrinsic mode components; the optimized variational mode decomposition algorithm is a variational mode decomposition algorithm optimized with the gray wolf optimization algorithm for the number of decomposition layers and the penalty factor. The normalized energy value calculation module is used to calculate the energy of each intrinsic mode component, and select the intrinsic mode components with the largest energy to calculate the normalized energy value of the driving current data. The fault type diagnosis module is used to input the normalized energy value into a pre-trained fault diagnosis model to obtain the fault type of the MOSFET power drive circuit; the pre-trained fault diagnosis model is a pre-trained convolutional neural network.

Images