AMT oil temperature sensor fault diagnosis method, device, equipment and medium

Abstract

The invention belongs to the technical field of automobiles, and particularly relates to an AMT oil temperature sensor fault diagnosis method, device and equipment and a medium, and the method comprises the steps: S1, training a prediction model through collected AMT oil temperature, vehicle speed, motor water temperature, environment temperature and output shaft torque data, and predicting AMT oil temperature values under different working conditions; s2, acquiring an oil temperature signal detected by the AMT oil temperature sensor in real time and filtering the oil temperature signal; s3, comparing the filtered oil temperature signal with a predicted oil temperature value; s4, judging whether the oil temperature signal is credible or not; if not, the step S5 is executed, and if yes, the step S6 is executed; s5, the predicted oil temperature value of the corresponding working condition replaces an untrusted oil temperature signal and is sent to a vehicle control module; and S6, sending the credible oil temperature signal to a vehicle control module. The fault that the numerical value of the AMT oil temperature sensor is not credible can be diagnosed timely and accurately, and transmission faults caused by oil temperature control errors are effectively avoided.

Description

Technical Field

[0001] The present invention belongs to the technical field of automobiles, and particularly relates to a method, device, equipment and medium for fault diagnosis of an AMT oil temperature sensor. Background Technique

[0002] AMT, short for Automated Mechanical Transmission, is an automated gearbox. The AMT oil temperature sensor is responsible for accurately converting the transmission oil temperature into an electrical signal and transmitting it to the vehicle control system. During vehicle driving, the temperature of the AMT transmission oil directly affects the performance and lifespan of the entire AMT system. Under normal operating conditions, the control system precisely regulates the key parameters of the shift timing based on the information feedback from the oil temperature sensor. However, like other precision components, the sensor itself will age over time. At the same time, circuit faults are also an important factor causing abnormal values of the oil temperature sensor. Bumps and vibrations during vehicle driving may cause the circuit connections to become loose or damaged, thereby affecting the transmission stability of the electrical signal. External interference is even more inevitable. For example, in areas with strong electromagnetic radiation, it may interfere with the normal signal acquisition and transmission of the oil temperature sensor, causing the vehicle control system to receive incorrect oil temperature information, and ultimately leading to serious deviations in AMT control. Such deviations will not only cause abnormal phenomena such as jerks and incorrect shift timing when the vehicle shifts gears, but also reduce the overall performance of the AMT transmission due to improper oil temperature control. Over time, it may even cause irreversible damage to the transmission.

[0003] Currently, the diagnostic method for the AMT oil temperature sensor is to set the upper and lower threshold values of the oil temperature. When the measured value of the sensor exceeds this range, it is determined that the sensor is faulty. For example, the low-temperature threshold is set at -40°C and the high-temperature threshold is set at 150°C. If the detected oil temperature is lower than -40°C or higher than 150°C, it is considered that the sensor may be faulty. This diagnostic method can only detect fault situations that significantly exceed the normal range and cannot effectively identify small deviations or drifts of the sensor. If the measured value of the sensor is within the threshold but there is a certain inaccuracy, the system cannot detect it, which may cause the AMT transmission control unit to control based on inaccurate oil temperature data, affecting the accuracy of the shift strategy and oil pressure control. Summary of the Invention

[0004] Aiming at the problem that the traditional detection method can only detect fault situations that significantly exceed the normal range and cannot effectively identify small deviations or drifts of the sensor, the present invention provides a method, device, equipment and medium for fault diagnosis of an AMT oil temperature sensor.

[0005] In a first aspect, the technical solution of the present invention provides a method for fault diagnosis of an AMT oil temperature sensor, including the following steps: S1: Train a prediction model using the collected data of AMT oil temperature, vehicle speed, motor water temperature, ambient temperature, and output shaft torque to predict the oil temperature values of AMT under different working conditions; S2: Obtain the oil temperature signal detected by the AMT oil temperature sensor in real time and perform filtering processing; S3: Compare the filtered oil temperature signal with the predicted oil temperature value; S4: Determine whether the oil temperature signal is credible; If not, execute step S5; if so, execute step S6; S5: Replace the uncredible oil temperature signal with the predicted oil temperature value corresponding to the working condition and send it to the vehicle control module; S6: Send the credible oil temperature signal to the vehicle control module.

[0006] As a further limitation of the technical solution of the present invention, after the step of obtaining the oil temperature signal detected by the AMT oil temperature sensor in real time and performing filtering processing, it includes: S2-3: Determine whether the filtered oil temperature signal reports an electrical fault; If so, execute step S5; if not, execute step S3.

[0007] As a further limitation of the technical solution of the present invention, the step of training a prediction model using the collected data of AMT oil temperature, vehicle speed, motor water temperature, ambient temperature, and output shaft torque to predict the oil temperature values of AMT under different working conditions includes: S11: Under the normal working state of the AMT, collect a large amount of data of speed, motor water temperature, ambient temperature, output shaft torque, and the corresponding oil temperature values under different working conditions to generate a data set; S12: Divide the collected data set into an input data matrix X and an output data matrix Y; X contains multiple features of vehicle speed, motor water temperature, ambient temperature, and output shaft torque, and its dimension is , where n represents the number of samples, m represents the number of features, and Y is the corresponding AMT oil temperature value; S13: Initialize the weights and biases of the multi-layer perceptron neural network; S14: Propagate the input data forward through each layer of the network, calculate the prediction result of the output layer, and calculate the loss function according to the prediction result of the output layer and the real oil temperature value; S15: According to the loss function, calculate the gradients of each layer through the backpropagation algorithm; specifically, starting from the output layer, calculate the error term of each layer in turn, and update the weights and biases of this layer according to the error term; S16: After training, obtain the trained weight matrix and bias vector, and then obtain the oil temperature prediction model under different working conditions; S17: Predict the oil temperature values of the AMT under different working conditions based on the aforementioned prediction model.

[0008] As a further limitation of the technical solution of the present invention, the steps of collecting a large amount of data on speed, motor water temperature, ambient temperature, output shaft torque, and corresponding oil temperature values under different working conditions to generate a data set under the normal working state of the AMT include: S111: Under the normal working state of the AMT, collect a large amount of data on speed, motor water temperature, ambient temperature, output shaft torque, and corresponding oil temperature values under different working conditions; S112: Remove outliers, error values, and duplicate values from the data; S113: Perform normalization processing on the collected data to map numerical values in different ranges to the interval [0, 1] or [-1, 1]; S114: Extract features from the original data to generate a data set.

[0009] As a further limitation of the technical solution of the present invention, the steps of performing forward propagation of the input data through each layer of the network, calculating the prediction result of the output layer, and calculating the loss function based on the prediction result of the output layer and the true oil temperature value include: S141: Input the sample into the initialized neural network and calculate the linear combination from the input layer to the hidden layer ; S142: Apply the activation function to to obtain the output of the hidden layer ; Calculate the linear combination from the hidden layer to the output layer , which is the predicted output of the output layer ; S143: Use the mean square error to calculate the loss function between the predicted value and the true value , where N is the number of samples. , N is the number of samples.

[0010] As a further limitation of the technical solution of the present invention, is the sample in the input vector ; Calculate the gradient of the output layer: , , , ; Calculate the gradient of the hidden layer: , , ; Among them, is the bias vector from the hidden layer to the output layer, is the weight matrix from the hidden layer to the output layer, is the bias vector from the input layer to the hidden layer, is the weight matrix from the input layer to the hidden layer, represents element-wise multiplication, is the derivative of the activation function, when, otherwise , , , .

[0011] As a further limitation of the technical solution of the present invention, the gradient descent algorithm is used to update the parameters: , ; , ; where is the learning rate.

[0012] The prediction model is trained using multi-dimensional data (vehicle speed, motor water temperature, ambient temperature, output shaft torque), taking into account the influence of various operating conditions on the AMT oil temperature, enabling the prediction model to adapt to different driving conditions and environmental conditions. Whether under high temperature, low temperature, high load, low load, or different vehicle speed conditions, the system can accurately predict the oil temperature and reasonably evaluate and process the signals of the oil temperature sensor, ensuring the normal operation of the vehicle under various operating conditions and improving the adaptability and reliability of the vehicle under complex operating conditions.

[0013] In a second aspect, the technical solution of the present invention also provides an AMT oil temperature sensor fault diagnosis system, including an oil temperature prediction module, an oil temperature acquisition module, an oil temperature filtering module, an oil temperature credibility determination module, and an oil temperature substitution module; The oil temperature prediction module is used to train a prediction model by collecting AMT oil temperature, vehicle speed, motor water temperature, ambient temperature, and output shaft torque data, and predict the oil temperature values of the AMT under different operating conditions; The oil temperature acquisition module is used to obtain in real time the oil temperature signal detected by the AMT oil temperature sensor; The oil temperature filtering module filters the oil temperature signal obtained by the oil temperature real-time acquisition module; The oil temperature credibility determination module is used to compare the filtered oil temperature signal with the predicted oil temperature value; determine whether the oil temperature signal is credible; if the oil temperature is not credible, trigger the oil temperature substitution module, and if the oil temperature is credible, directly send the credible oil temperature signal to the vehicle control module; The oil temperature substitution module is used to substitute the non-credible oil temperature signal with the predicted oil temperature value corresponding to the operating condition and send it to the vehicle control module.

[0014] As a further limitation of the technical solution of the present invention, the system further includes an electrical fault judgment module, which is used to judge whether an electrical fault is reported in the oil temperature signal after filtering processing. If so, the oil temperature substitution module is triggered; if not, the reliable oil temperature signal is sent to the vehicle control module.

[0015] As a further limitation of the technical solution of the present invention, the oil temperature prediction module is specifically used to collect a large amount of data on speed, motor water temperature, ambient temperature, output shaft torque, and corresponding oil temperature values under different working conditions to generate a data set in the normal working state of the AMT; divide the collected data set into an input data matrix X and an output data matrix Y; X contains multiple features such as vehicle speed, motor water temperature, ambient temperature, and output shaft torque, and its dimension is , where n represents the number of samples and m represents the number of features, and Y is the corresponding AMT oil temperature value; initialize the weights and biases of the multi-layer perceptron neural network; perform forward propagation of the input data through each layer of the network to calculate the prediction result of the output layer, and calculate the loss function according to the prediction result of the output layer and the true oil temperature value; calculate the gradients of each layer through the backpropagation algorithm according to the loss function; specifically start from the output layer, calculate the error term of each layer in turn, and update the weights and biases of this layer according to the error term; after training, obtain the trained weight matrix and bias vector, and then obtain the oil temperature prediction model under different working conditions; predict the oil temperature value of the AMT under different working conditions based on the prediction model.

[0016] As a further limitation of the technical solution of the present invention, the oil temperature prediction module is specifically further used to input the sample into the initialized neural network, and calculate the linear combination from the input layer to the hidden layer ; apply the activation function to to obtain the output of the hidden layer ; calculate the linear combination from the hidden layer to the output layer , which is the predicted output of the output layer ; use the mean square error to calculate the loss function between the predicted value and the true value , , and N is the number of samples.

[0017] As a further limitation of the technical solution of the present invention, the oil temperature prediction module is specifically used to collect a large amount of data on speed, motor water temperature, ambient temperature, output shaft torque, and corresponding oil temperature values under different working conditions in the normal working state of the AMT; remove outliers, error values, and duplicate values from the data; perform normalization processing on the collected data to map numerical values in different ranges to the interval [0,1] or [-1,1]; perform feature extraction on the original data to generate a data set.

[0018] As a further limitation of the technical solution of the present invention, is the input vector in the sample; Calculate the output layer gradient: , , ; Calculate the gradient of the hidden layer: , , ; Among them, is the bias vector from the hidden layer to the output layer, is the weight matrix from the hidden layer to the output layer, is the bias vector from the input layer to the hidden layer, is the weight matrix from the input layer to the hidden layer, represents element-wise multiplication, is the derivative of the activation function, When, , otherwise , , , .

[0019] Since the system can detect the failure of the oil temperature sensor in time, it enables maintenance personnel to perform maintenance and repair targeted according to the failure information, avoiding blind replacement of components and reducing the maintenance cost. At the same time, by predicting the oil temperature value to assist the control module to work, the vehicle can continue to run normally for a certain period of time until the appropriate maintenance time, improving the maintainability and operation efficiency of the vehicle. Especially for long-haul transport vehicles or operation fleets, this fault diagnosis method can bring significant economic benefits and operation convenience.

[0020] In a third aspect, the technical solution of the present invention further provides an electronic device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; the memory stores computer program instructions executable by the at least one processor, and the computer program instructions are executed by the at least one processor so that the at least one processor can execute the AMT oil temperature sensor fault diagnosis method as described in the first aspect.

[0021] In a fourth aspect, the technical solution of the present invention further provides a non-transitory computer-readable storage medium, and the non-transitory computer-readable storage medium stores computer instructions, and the computer instructions cause the computer to execute the AMT oil temperature sensor fault diagnosis method as described in the first aspect.

[0022] The beneficial effects of the technical solution of the present invention are as follows: By establishing a prediction model to predict the AMT oil temperature under different working conditions, an additional reference index is provided for the system. When the oil temperature sensor signal obtained in real time fails or is untrustworthy, the vehicle control system no longer completely relies on the possibly inaccurate sensor signal, but can use the predicted oil temperature value as a substitute, avoiding incorrect control commands caused by oil temperature sensor failures, thereby significantly improving the overall reliability and safety of the vehicle control system. For example, it prevents the vehicle from operating at too high or too low an oil temperature due to abnormal oil temperature signals, thereby avoiding damage to AMT components and potential driving safety hazards.

[0023] By adopting the method of comparing the real-time detected oil temperature signal with the predicted oil temperature value, it is possible to quickly and accurately determine whether the oil temperature sensor fails or provides an untrustworthy signal. This mechanism of comparison and judgment makes the fault diagnosis more intelligent and precise, and can detect problems in a timely manner when the oil temperature sensor has a minor fault or is interfered, rather than waiting until the sensor completely fails or outputs an obviously incorrect signal to detect the fault, which helps vehicle maintenance personnel to carry out maintenance and repair in advance, reducing the vehicle downtime and repair costs caused by sensor failures.

[0024] The introduction of real-time filtering processing can effectively filter out the noise in the oil temperature signal, enabling the system to receive more accurate and stable oil temperature information. More accurate oil temperature information helps the vehicle control module to make precise control decisions, such as adjusting the shift strategy, optimizing the working parameters of the engine and motor, etc., thereby improving the overall performance of the vehicle, including fuel economy, power performance, and transmission efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the technical solution of the present invention, the drawings required for description will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0026] Figure 1 It is a schematic flowchart of the method provided by the embodiment of the present invention.

[0027] Figure 2 It is a schematic flowchart of the method provided by another embodiment of the present invention.

[0028] Figure 3 It is a connection block diagram of the system provided by the embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0029] To make the objectives, features, and advantages of the present invention more obvious and understandable, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the specific embodiments of the present invention. Obviously, the embodiments described below are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in this application without making creative efforts belong to the scope of protection of this application.

[0030] As Figure 1 shown, the first aspect provided by the embodiment of the present invention is that the technical solution of the present invention provides a method for diagnosing faults in an AMT oil temperature sensor, including the following steps: S1: Train a prediction model with the collected AMT oil temperature, vehicle speed, motor water temperature, ambient temperature, and output shaft torque data to predict the oil temperature value of the AMT under different working conditions; S2: Obtain the oil temperature signal detected by the AMT oil temperature sensor in real time and perform filtering processing; When the temperature value monitored by the oil temperature sensor undergoes an instantaneous mutation and then quickly returns to the temperature level before the mutation within a very short time, such temperature mutation values caused by abnormal situations will be identified and eliminated after filtering processing, so as to ensure the stability and reliability of the oil temperature data and provide more accurate oil temperature information support for the precise operation of the system.

[0031] S3: Compare the filtered oil temperature signal with the predicted oil temperature value; S4: Determine whether the oil temperature signal is credible; If not, execute step S5; if so, execute step S6; S5: Replace the uncredible oil temperature signal with the predicted oil temperature value corresponding to the working condition and send it to the vehicle control module; S6: Send the credible oil temperature signal to the vehicle control module.

[0032] In some embodiments, as Figure 2 shown, after the step of obtaining the oil temperature signal detected by the AMT oil temperature sensor in real time and performing filtering processing, it includes: S2-3: Determine whether the filtered oil temperature signal reports an electrical fault; If so, execute step S5; if not, execute step S3.

[0033] In some embodiments, the step of training a prediction model with the collected AMT oil temperature, vehicle speed, motor water temperature, ambient temperature, and output shaft torque data to predict the oil temperature value of the AMT under different working conditions includes: S11: Under the normal working condition of the AMT, collect a large amount of data on speed, motor water temperature, ambient temperature, output shaft torque, and corresponding oil temperature values under different working conditions to generate a data set; S12: Divide the collected data set into an input data matrix X and an output data matrix Y; X contains multiple features such as vehicle speed, motor water temperature, ambient temperature, and output shaft torque, and its dimension is , where n represents the number of samples and m represents the number of features, and Y is the corresponding AMT oil temperature value; S13: Initialize the weights and biases of the multi-layer perceptron neural network; S14: Propagate the input data forward through the layers of the network to calculate the prediction result of the output layer. According to the prediction result of the output layer and the true oil temperature value, calculate the loss function; S15: According to the loss function, calculate the gradients of each layer through the backpropagation algorithm; specifically, starting from the output layer, calculate the error term of each layer in turn, and update the weights and biases of this layer according to the error term; S16: After training, obtain the trained weight matrix and bias vector, and then obtain the oil temperature prediction model under different working conditions; S17: Predict the oil temperature values of the AMT under different working conditions based on the predicted model.

[0034] In some embodiments, the step of collecting a large amount of data on speed, motor water temperature, ambient temperature, output shaft torque, and corresponding oil temperature values under different working conditions to generate a data set under the normal working condition of the AMT includes: S111: Under the normal working condition of the AMT, collect a large amount of data on speed, motor water temperature, ambient temperature, output shaft torque, and corresponding oil temperature values under different working conditions; S112: Remove outliers, error values, and duplicate values from the data; S113: Normalize the collected data to map numerical values in different ranges to the interval [0, 1] or [-1, 1]; Map the data to the interval [0, 1], and the formula is , where x is the original data, x min and x max are the minimum and maximum values of this feature data.

[0035] S114: Extract features from the original data to generate a data set.

[0036] In some embodiments, the step of propagating the input data forward through the layers of the network to calculate the prediction result of the output layer and calculating the loss function according to the prediction result of the output layer and the true oil temperature value includes: S141: The sample Input it into the initialized neural network and calculate the linear combination from the input layer to the hidden layer ; S142: Apply the activation function to to obtain the output of the hidden layer ; Calculate the linear combination from the hidden layer to the output layer , which is the predicted output of the output layer ; S143: Use the mean square error to calculate the loss function between the predicted value and the true value , where N is the number of samples .

[0037] In some embodiments, is the sample in the input vector ; Calculate the gradient of the output layer: , , ; Calculate the gradient of the hidden layer: , , ; Among them, is the bias vector from the hidden layer to the output layer, is the weight matrix from the hidden layer to the output layer, is the bias vector from the input layer to the hidden layer, is the weight matrix from the input layer to the hidden layer, represents element-wise multiplication, is the derivative of the activation function, when , otherwise , , , .

[0038] Update the parameters using the gradient descent algorithm: , ; , ; Among them is the learning rate

[0039] In some embodiments, since there are four types of temperature-related data in the input layer, the number of input layer nodes is 4, and each node corresponds to a type of input data. A hidden layer with 10 nodes is set. The activation function of the hidden layer nodes can be selected as the ReLU (Rectified Linear Unit) function, f(x) = max(0, x), which can effectively alleviate the problem of gradient disappearance and has high computational efficiency. The output layer has only 1 node, which is used to output the predicted temperature value, and the Sigmoid function is used to map the output to an appropriate interval. The formula of the Sigmoid function: The AMT oil temperature range of -40°C to 150°C is denoted as [a, b]. Let the value obtained through the Sigmoid function be y_sigmoid, then the output predicted temperature = a + (b - a)y_sigmoid Initialize the weights and biases of the neural network. The weights can be randomly initialized, for example, sampled from a normal distribution with a mean of 0 and a standard deviation of a small value (such as 0.01). The biases can be initialized to 0 or a small value.

[0040] Input the training set data into the network, calculate the output through forward propagation, and then calculate the loss function based on the output and the true temperature value. Then use the backpropagation algorithm to calculate the gradient and use the optimizer to update the weights and biases of the network. During the training process, prevent overfitting by observing the loss function value on the validation set. Use the test set to evaluate the trained model. Calculate various evaluation metrics between the predicted temperature value and the true temperature value, the root mean square error (RMSE), and the formula is ; the mean absolute error (MAE), and the formula is . These metrics can measure the accuracy of the model prediction, and judge whether the model meets the requirements according to the evaluation results.

[0041] As Figure 3 shown, an AMT oil temperature sensor fault diagnosis system is further provided in an embodiment of the present invention, including an oil temperature prediction module, an oil temperature acquisition module, an oil temperature filtering module, an oil temperature credibility determination module, and an oil temperature replacement module; The oil temperature prediction module is used to train a prediction model through the collected AMT oil temperature, vehicle speed, motor water temperature, ambient temperature, and output shaft torque data, and predict the oil temperature value of the AMT under different working conditions; The oil temperature acquisition module is used to obtain the oil temperature signal detected by the AMT oil temperature sensor in real time; The oil temperature filtering module filters the oil temperature signal obtained by the oil temperature real-time acquisition module; The oil temperature reliability determination module is used to compare the filtered oil temperature signal with the predicted oil temperature value, determine whether the oil temperature signal is reliable. If the oil temperature is not reliable, it triggers the oil temperature substitution module. If the oil temperature is reliable, it directly sends the reliable oil temperature signal to the vehicle control module; The oil temperature substitution module is used to substitute the unreliable oil temperature signal with the predicted oil temperature value under the corresponding working condition and send it to the vehicle control module.

[0042] The oil temperature filtering module filters the temperature signal of the AMT oil temperature sensor to remove instantaneous interference signals. The filtered AMT oil temperature signal is compared with the oil temperature in the model. The oil temperature prediction module uses multi-dimensional data such as the AMT oil temperature, speed, motor water temperature, ambient temperature, output shaft torque, etc. of the factory vehicles collected as the input layer neurons, and performs complex non-linear transformations through the neurons in the hidden layer, and finally outputs the predicted oil temperature value under the corresponding working condition. It regularly (T) receives the data of the signal collection and storage unit, corrects and optimizes the established normal working model to ensure that the model can always accurately reflect the current actual operating state of the vehicle and improve the accuracy of fault diagnosis. The oil temperature reliability determination module compares the predicted temperature in the AMT oil temperature prediction module with the real-time temperature of the oil temperature sensor to determine whether the value is reliable. The oil temperature substitution module is used to substitute the unreasonable oil temperature signal with the oil temperature signal under the corresponding working condition of the oil temperature prediction module when there is an electrical fault diagnosis of the oil temperature or the unreliable signal determined by the AMT oil temperature unreliable determination module and send it to the vehicle control module. The acquisition module collects the working condition information of the AMT oil temperature sensor during normal operation in real time and sends the data to the oil temperature prediction module regularly. The oil temperature reliability determination module determines the oil temperature signal collected by the oil temperature sensor. The oil temperature substitution module substitutes the unreliable oil temperature signal with the set oil temperature substitution value in the oil temperature prediction module when the oil temperature sensor reports an electrical fault or the AMT oil temperature unreliable determination module determines it to be unreliable. The oil temperature filtering module filters the collected AMT oil temperature signal to remove instantaneous interference signals and prevent false alarms of unreliable numerical values. It greatly improves the operating reliability of the vehicle. In practical applications, it can timely and accurately diagnose the fault of unreliable numerical values of the AMT oil temperature sensor and effectively avoid transmission faults caused by oil temperature control errors.

[0043] In some embodiments, the system further includes an electrical fault judgment module, which is used to judge whether the filtered oil temperature signal reports an electrical fault. If so, it triggers the oil temperature substitution module; if not, it sends the reliable oil temperature signal to the vehicle control module.

[0044] In some embodiments, the oil temperature prediction module is specifically configured to collect data on speed, motor water temperature, ambient temperature, output shaft torque, and corresponding oil temperature values under a large number of different working conditions to generate a data set in the normal working state of the AMT; divide the collected data set into an input data matrix X and an output data matrix Y; X contains multiple features such as vehicle speed, motor water temperature, ambient temperature, and output shaft torque, and its dimension is , where n represents the number of samples and m represents the number of features, and Y is the corresponding AMT oil temperature value; initialize the weights and biases of the multi-layer perceptron neural network; perform forward propagation of the input data through each layer of the network, calculate the prediction result of the output layer, and calculate the loss function based on the prediction result of the output layer and the true oil temperature value; calculate the gradients of each layer through the backpropagation algorithm; specifically starting from the output layer, calculate the error term of each layer in turn, and update the weights and biases of this layer according to the error term; after training, obtain the trained weight matrix and bias vector, and then obtain the oil temperature prediction model under different working conditions; predict the oil temperature values of the AMT under different working conditions based on the prediction model.

[0045] In some embodiments, the oil temperature prediction module is further specifically configured to input the sample into the initialized neural network, and calculate the linear combination from the input layer to the hidden layer ; apply the activation function to to obtain the output of the hidden layer ; calculate the linear combination from the hidden layer to the output layer , which is the predicted output of the output layer ; use the mean square error to calculate the loss function between the predicted value and the true value , , where N is the number of samples.

[0046] In some embodiments, the oil temperature prediction module is specifically configured to collect data on speed, motor water temperature, ambient temperature, output shaft torque, and corresponding oil temperature values under a large number of different working conditions in the normal working state of the AMT; remove outliers, error values, and duplicate values from the data; perform normalization processing on the collected data to map numerical values in different ranges to the interval [0,1] or [-1,1]; perform feature extraction on the original data to generate a data set.

[0047] is the sample in the input vector ; Calculate the output layer gradient: , , ; Calculate the gradient of the hidden layer: , , ; Among them, is the bias vector from the hidden layer to the output layer, is the weight matrix from the hidden layer to the output layer, is the bias vector from the input layer to the hidden layer, is the weight matrix from the input layer to the hidden layer, represents element-wise multiplication, is the derivative of the activation function, When , otherwise , , , .

[0048] Update the parameters using the gradient descent algorithm: , ; , ; Among them is the learning rate.

[0049] Establish a mechanism for regular model updates. Because during the long-term use of the vehicle, due to the wear and aging of components and the change of the use environment, its overall performance will gradually change. Regularly collect the latest data of the vehicle during actual operation, and use these data to correct and optimize the established normal working model to ensure that the model can always accurately reflect the current actual operation state of the vehicle and improve the accuracy of fault diagnosis.

[0050] If the oil temperature substitution module determines through the model that the value of the oil temperature sensor is truly unreliable, it replaces the value of the oil temperature sensor with the temperature value in the model. Temporarily maintain the normal operation of the vehicle until it reaches the repair point. Greatly improve the operating reliability of the vehicle. In practical applications, this invention application can timely and accurately diagnose the fault that the value of the AMT oil temperature sensor is unreliable, and effectively avoid transmission faults caused by oil temperature control errors.

[0051] Reduce the repair cost and parking time. Since it can quickly locate the fault point and provide accurate repair suggestions, maintenance personnel can carry out maintenance work more efficiently. This not only shortens the parking time of the vehicle, improves the utilization rate of the vehicle, but also reduces the component replacement cost and labor cost during the repair process.

[0052] An embodiment of the present invention further provides an electronic device, which includes: a processor, a communication interface, a memory, and a communication bus. Among them, the processor, the communication interface, and the memory complete mutual communication through the communication bus. The communication bus can be used for information transmission between the electronic device and the sensor. The processor can call the logical instructions in the memory to execute the following method: S1: Train a prediction model through the collected AMT oil temperature, vehicle speed, motor water temperature, ambient temperature, and output shaft torque data to predict the oil temperature value of the AMT under different working conditions; S2: Real-time obtain the oil temperature signal detected by the AMT oil temperature sensor and perform filtering processing; S3: Compare the filtered oil temperature signal with the predicted oil temperature value; S4: Determine whether the oil temperature signal is credible; if not, execute step S5, if so, execute step S6; S5: Replace the uncredible oil temperature signal with the predicted oil temperature value corresponding to the working condition and send it to the vehicle control module; S6: Send the credible oil temperature signal to the vehicle control module.

[0053] In addition, when the logical instructions in the above-mentioned memory are implemented in the form of software functional units and sold or used as an independent product, they can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs that can store program codes.

[0054] An embodiment of the present invention provides a non-transitory computer-readable storage medium, which stores computer instructions that enable a computer to execute the method provided by the above method embodiment, for example, including: S1: Train a prediction model through the collected AMT oil temperature, vehicle speed, motor water temperature, ambient temperature, and output shaft torque data to predict the oil temperature value of the AMT under different working conditions; S2: Real-time obtain the oil temperature signal detected by the AMT oil temperature sensor and perform filtering processing; S3: Compare the filtered oil temperature signal with the predicted oil temperature value; S4: Determine whether the oil temperature signal is credible; if not, execute step S5, if so, execute step S6; S5: Replace the uncredible oil temperature signal with the predicted oil temperature value corresponding to the working condition and send it to the vehicle control module; S6: Send the credible oil temperature signal to the vehicle control module.

[0055] It should be understood that the sequence numbers of the steps in the above embodiments do not imply the order of execution, and the execution order of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.

[0056] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present invention. Various modifications to these embodiments will be obvious to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for diagnosing faults of an AMT oil temperature sensor, characterized in that: The steps include: S1: The prediction model is trained by collecting data on AMT oil temperature, vehicle speed, motor water temperature, ambient temperature, and output shaft torque to predict the oil temperature value of the AMT under different working conditions; S2: Obtain the oil temperature signal detected by the AMT oil temperature sensor in real time and perform filtering processing; S3: comparing the filtered oil temperature signal with the predicted oil temperature value; S4: Determine whether the oil temperature signal is credible; If not, go to step S5, if so, go to step S6; S5: replacing the unreliable oil temperature signal with the predicted oil temperature value of the corresponding working condition and sending it to the vehicle control module; S6: Send the reliable oil temperature signal to the vehicle control module.

2. The AMT oil temperature sensor fault diagnosis method according to claim 1, characterized in that: The steps of obtaining the oil temperature signal detected by the AMT oil temperature sensor in real time and filtering the signal include: S2-3: Determine whether the oil temperature signal after filtering has reported an electrical fault; If yes, go to step S5; if no, go to step S3.

3. The AMT oil temperature sensor fault diagnosis method according to claim 2, characterized in that: The steps of training the prediction model by collecting the AMT oil temperature, vehicle speed, motor water temperature, ambient temperature, and output shaft torque data to predict the AMT oil temperature value under different working conditions include: S11: Under normal working state of AMT, a large amount of speed, motor water temperature, ambient temperature, output shaft torque and corresponding oil temperature value data under different working conditions are collected to generate a data set; S12: Divide the collected data set into input data matrix X and output data matrix Y; X includes multiple features such as vehicle speed, motor water temperature, ambient temperature and output shaft torque, and its dimension is , where n represents the number of samples, m represents the number of features, and Y is the corresponding AMT oil temperature value; S13: Initialize the weights and biases of the multilayer perceptron neural network; S14: forward propagating the input data through each layer of the network, calculating the prediction result of the output layer, and calculating the loss function according to the prediction result of the output layer and the actual oil temperature value; S15: According to the loss function, the gradient of each layer is calculated through the back propagation algorithm; starting from the output layer, the error term of each layer is calculated in turn, and the weight and bias of the layer are updated according to the error term; S16: After the training is completed, the trained weight matrix and bias vector are obtained, and then the oil temperature prediction model under different working conditions is obtained; S17: Predicting the oil temperature value of the AMT under different working conditions based on the prediction model.

4. The AMT oil temperature sensor fault diagnosis method according to claim 3, characterized in that: When the AMT is in normal working state, the steps of collecting a large amount of speed, motor water temperature, ambient temperature, output shaft torque and corresponding oil temperature value data under different working conditions to generate a data set include: S111: Under normal working state of AMT, collect a large amount of data on speed, motor water temperature, ambient temperature, output shaft torque and corresponding oil temperature under different working conditions; S112: Remove outliers, erroneous values ​​and duplicate values ​​from the data; S113: normalizing the collected data, mapping values ​​in different ranges to the interval [0,1] or [-1,1]; S114: Extract features from the original data to generate a data set.

5. The AMT oil temperature sensor fault diagnosis method according to claim 4, characterized in that: The input data is forward propagated through each layer of the network to calculate the prediction result of the output layer. According to the prediction result of the output layer and the actual oil temperature value, the steps of calculating the loss function include: S141: Sample Input into the initialized neural network and calculate the linear combination from the input layer to the hidden layer ; is the input vector Samples in S142: Yes Apply activation function to get the output of hidden layer ; Calculate the linear combination of hidden layer to output layer , which is the predicted output of the output layer ; S143: Calculate predicted values ​​using mean square error and the true value The loss function between , N is the sample size.

6. The AMT oil temperature sensor fault diagnosis method according to claim 5, characterized in that: Calculate the output layer gradient: , , ; Calculate the gradient of the hidden layer: , , ; in, is the bias vector from the hidden layer to the output layer, is the weight matrix from the hidden layer to the output layer, is the bias vector from the input layer to the hidden layer, is the weight matrix from the input layer to the hidden layer, represents element-wise multiplication, is the derivative of the activation function, hour, ,otherwise , , , .

7. An AMT oil temperature sensor fault diagnosis system, characterized in that: It includes an oil temperature prediction module, an oil temperature acquisition module, an oil temperature filtering module, an oil temperature reliable determination module and an oil temperature substitution module; The oil temperature prediction module is used to train the prediction model through the collected AMT oil temperature, vehicle speed, motor water temperature, ambient temperature, and output shaft torque data to predict the oil temperature value of the AMT under different working conditions; The oil temperature acquisition module is used to obtain the oil temperature signal detected by the AMT oil temperature sensor in real time; The oil temperature filtering module filters the oil temperature signal acquired by the oil temperature real-time acquisition module; The oil temperature credibility judgment module is used to compare the oil temperature signal after filtering with the predicted oil temperature value to determine whether the oil temperature signal is credible; If the oil temperature is not credible, the oil temperature substitution module is triggered. If the oil temperature is credible, the credible oil temperature signal is directly sent to the vehicle control module. The oil temperature substitution module is used to replace the unreliable oil temperature signal with the predicted oil temperature value of the corresponding working condition and send it to the vehicle control module.

8. The AMT oil temperature sensor fault diagnosis system according to claim 7, characterized in that: The system also includes an electrical fault judgment module for judging whether the oil temperature signal after filtering has reported an electrical fault. If so, the oil temperature substitution module is triggered; if not, a reliable oil temperature signal is sent to the vehicle control module.

9. An electronic device, characterized in that: The electronic device includes: at least one processor; and a memory communicatively connected to the at least one processor; the memory stores computer program instructions that can be executed by the at least one processor, and the computer program instructions are executed by the at least one processor so that the at least one processor can execute the AMT oil temperature sensor fault diagnosis method as described in any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium, characterized in that: The non-transitory computer-readable storage medium stores computer instructions, and the computer instructions enable the computer to execute the AMT oil temperature sensor fault diagnosis method as described in any one of claims 1 to 7.

Images