Hydraulic Transmission Shift Fault Identification Method, Model Training Method and Device

Abstract

An embodiment of this specification discloses a method for identifying shift faults of a hydraulic transmission, a method for model training, and a device. The method includes: obtaining an overall evaluation index and a stage evaluation index during the shift process of the hydraulic transmission, where the overall evaluation index is an index used to characterize the overall performance of the shift process, and the stage evaluation index is an index used to characterize the stage performance of each shift stage included in the shift process; inputting the overall evaluation index into a first preset fault identification model to obtain a first fault identification result, where the first fault identification result is used to characterize whether a fault affecting driving safety occurs during the shift process; if the first fault identification result indicates that no fault affecting driving safety occurs during the shift process, inputting the overall evaluation index and the stage evaluation index into a second preset fault identification model to obtain a second fault identification result. This solution can identify shift faults more comprehensively and accurately to improve the reliability of the use of the hydraulic transmission.

Description

Technical Field

[0001] The embodiments of this specification relate to the field of vehicles, and in particular, to a method for identifying shift faults of a hydraulic transmission, a method for training a model, and a device therefor. Background Art

[0002] A hydraulic transmission (Automatic Transmission, AT) is one of the components with the highest degree of automation complexity in modern vehicles. The reliability of its operation directly affects the reliability and safety of the entire vehicle. Therefore, the requirements for fault diagnosis of the hydraulic transmission are also getting higher and higher. Among them, electro-hydraulic automatic shifting is the core technology of AT, which belongs to typical complex electro-mechanical-hydraulic control technology. Due to frequent operation and complex and changeable working conditions, the probability of failure is relatively high, and it directly affects the working reliability of AT.

[0003] In the prior art, AT fault diagnosis technology tends to diagnose hardware systems such as communication, electrical components, hydraulic components, and circuits, lacking research on the dynamic characteristics of faults in the shifting process, resulting in problems of incomplete fault identification and misidentification, which in turn affects the reliability of AT use. Summary of the Invention

[0004] The embodiments of this specification provide a method for identifying shift faults of a hydraulic transmission, a method for training a model, and a device therefor.

[0005] In a first aspect, the embodiments of this specification provide a method for identifying shift faults of a hydraulic transmission, including:

[0006] Obtain an overall evaluation index and stage evaluation indexes during the shifting process of the hydraulic transmission, where the overall evaluation index is an index used to characterize the overall performance of the shifting process, and the stage evaluation indexes are indexes used to characterize the stage performance of each shifting stage included in the shifting process;

[0007] Input the overall evaluation index into a first preset fault identification model to obtain a first fault identification result, where the first fault identification result is used to characterize whether a fault affecting driving safety occurs during the shifting process;

[0008] If the first fault identification result indicates that no fault affecting driving safety occurs during the shifting process, input the overall evaluation index and the stage evaluation indexes into a second preset fault identification model to obtain a second fault identification result.

[0009] Optionally, the overall evaluation index includes one or more of the following indexes: the shifting duration of the shifting process, the peak value of the change rate of the turbine shaft speed during the shifting process, and the peak value of the change rate of the output shaft speed during the shifting process.

[0010] Optionally, the shifting process includes the following stages in chronological order: pre-oiling stage, torque stage, slow turbine speed decline stage, rapid turbine speed decline stage, and synchronization stage;

[0011] The stage evaluation indicators include one or more of the following indicators: the total turbine speed decline during the pre-oiling stage, the total turbine speed decline within one response cycle after the end of the pre-oiling stage, the actual duration of the torque stage, the total turbine speed decline during the torque stage, the actual duration of the slow turbine speed decline stage, the total turbine speed decline within one response cycle after the end of the slow turbine speed decline stage, and the actual duration of the rapid turbine speed decline stage.

[0012] Optionally, the first preset fault identification model and the second preset fault identification model are obtained in the following manner:

[0013] Obtain a training sample set, where each sample data in the training sample set includes the overall evaluation indicator and the stage evaluation indicator during one shifting process;

[0014] Based on the corresponding relationship between the preset evaluation indicators and the fault types, determine the fault type of each sample data as the label information of each sample data;

[0015] Construct a first initial fault identification model and a second initial fault identification model;

[0016] Based on the overall performance index corresponding to each sample data and the label information of each sample data, train the first initial fault identification model to obtain the trained first preset fault identification model;

[0017] Based on the overall evaluation indicator, stage evaluation indicator corresponding to each sample data, and the label information of each sample data, train the second initial fault identification model to obtain the trained second preset fault identification model.

[0018] In a second aspect, an embodiment of this specification provides a method for training a fault identification model. The fault identification model is used to identify shifting faults of a hydraulic transmission. The fault identification model includes a first preset fault identification model and a second preset fault identification model, and includes:

[0019] Obtain a training sample set, where each sample data in the training sample set includes the overall evaluation indicator and the stage evaluation indicator during one shifting process, where the overall evaluation indicator is an indicator used to characterize the overall performance of the shifting process, and the stage evaluation indicator is an indicator used to characterize the stage performance of each shifting stage included in the shifting process;

[0020] Construct a first initial fault identification model and a second initial fault identification model;

[0021] Based on the overall performance-price index corresponding to each sample data and the label information of each sample data, train the first initial fault identification model to obtain the trained first preset fault identification model;

[0022] Based on the overall evaluation index, stage evaluation index corresponding to each sample data, and the label information of each sample data, train the second initial fault identification model to obtain the trained second preset fault identification model.

[0023] Optionally, after obtaining the training sample set, the method further includes:

[0024] Based on the correspondence between the preset evaluation index and the fault type, determine the fault type of each sample data as the label information of each sample data.

[0025] Optionally, the overall evaluation index includes one or more of the following indexes: the shift duration of the shift process, the peak value of the change rate of the turbine shaft speed during the shift process, and the peak value of the change rate of the output shaft speed during the shift process.

[0026] Optionally, the shift process includes the following stages in chronological order: pre-oiling stage, torque stage, slow turbine speed decline stage, rapid turbine speed decline stage, and synchronization stage;

[0027] The stage evaluation index includes one or more of the following indexes: the total turbine speed decline amount in the pre-oiling stage, the total turbine speed decline amount within one response cycle after the end of the pre-oiling stage, the actual time used in the torque stage, the total turbine speed decline amount in the torque stage, the actual time used in the slow turbine speed decline stage, the total turbine speed decline amount within one response cycle after the end of the slow turbine speed decline stage, and the actual time used in the rapid turbine speed decline stage.

[0028] In a third aspect, an embodiment of this specification provides a shift fault identification device for a hydraulic transmission, and the device includes:

[0029] An acquisition module, configured to acquire an overall evaluation index and a stage evaluation index during the shift process of the hydraulic transmission, where the overall evaluation index is an index used to characterize the overall performance of the shift process, and the stage evaluation index is an index used to characterize the stage performance of each shift stage included in the shift process;

[0030] The first processing module is configured to input the overall evaluation index into a first preset fault identification model to obtain a first fault identification result, where the first fault identification result is used to characterize whether a fault affecting driving safety occurs during the shifting process;

[0031] The second processing module is configured to, when the first fault identification result indicates that no fault affecting driving safety occurs during the shifting process, input the overall evaluation index and the stage evaluation index into a second preset fault identification model to obtain a second fault identification result.

[0032] Optionally, the overall evaluation index includes one or more of the following indexes: the shifting duration of the shifting process, the peak value of the change rate of the turbine shaft speed during the shifting process, and the peak value of the change rate of the output shaft speed during the shifting process.

[0033] Optionally, the shifting process includes the following stages in chronological order: pre-oiling stage, torque stage, slow turbine speed decline stage, rapid turbine speed decline stage, and synchronization stage;

[0034] The stage evaluation index includes one or more of the following indexes: the total amount of turbine speed decline in the pre-oiling stage, the total amount of turbine speed decline within one response cycle after the end of the pre-oiling stage, the actual time used in the torque stage, the total amount of turbine speed decline in the torque stage, the actual time used in the slow turbine speed decline stage, the total amount of turbine speed decline within one response cycle after the end of the slow turbine speed decline stage, and the actual time used in the rapid turbine speed decline stage.

[0035] Optionally, the first preset fault identification model and the second preset fault identification model are obtained through the following methods:

[0036] Obtain a training sample set, where each sample data in the training sample set includes the overall evaluation index and the stage evaluation index during a shifting process;

[0037] Based on the corresponding relationship between the preset evaluation index and the fault type, determine the fault type of each sample data as the label information of each sample data;

[0038] Construct a first initial fault identification model and a second initial fault identification model;

[0039] Based on the overall performance index corresponding to each sample data and the label information of each sample data, train the first initial fault identification model to obtain the trained first preset fault identification model;

[0040] Based on the overall evaluation index, the stage evaluation index corresponding to each sample data, and the label information of each sample data, train the second initial fault recognition model to obtain the trained second preset fault recognition model.

[0041] Fourthly, an embodiment of this specification provides a fault recognition model training device. The fault recognition model is used to identify the shifting faults of a hydraulic transmission. The fault recognition model includes a first preset fault recognition model and a second preset fault recognition model. The device includes:

[0042] A sample acquisition module, configured to acquire a training sample set. Each sample data in the training sample set includes an overall evaluation index and a stage evaluation index during a shifting process. Among them, the overall evaluation index is an index used to characterize the overall performance of the shifting process, and the stage evaluation index is an index used to characterize the stage performance of each shifting stage included in the shifting process;

[0043] A model construction module, configured to construct a first initial fault recognition model and a second initial fault recognition model;

[0044] A first model training module, configured to train the first initial fault recognition model based on the overall performance index corresponding to each sample data and the label information of each sample data to obtain the trained first preset fault recognition model;

[0045] A second model training module, configured to train the second initial fault recognition model based on the overall evaluation index, the stage evaluation index corresponding to each sample data, and the label information of each sample data to obtain the trained second preset fault recognition model.

[0046] Optionally, the device further includes:

[0047] A label determination module, configured to determine the fault type of each sample data as the label information of each sample data based on the corresponding relationship between the preset evaluation index and the fault type.

[0048] Optionally, the overall evaluation index includes one or more of the following indexes: the shifting duration of the shifting process, the peak value of the change rate of the turbine shaft speed during the shifting process, and the peak value of the change rate of the output shaft speed during the shifting process.

[0049] Optionally, the shifting process includes the following stages in chronological order: a pre-oiling stage, a torque stage, a slow turbine speed drop stage, a fast turbine speed drop stage, and a synchronization stage;

[0050] The stage evaluation indicators include one or more of the following indicators: the total amount of turbine speed decrease during the pre-oiling stage, the total amount of turbine speed decrease within one response cycle after the end of the pre-oiling stage, the actual time used in the torque stage, the total amount of turbine speed decrease in the torque stage, the actual time used in the stage of slow turbine speed decrease, the total amount of turbine speed decrease within one response cycle after the end of the stage of slow turbine speed decrease, and the actual time used in the stage of rapid turbine speed decrease.

[0051] In a fifth aspect, an embodiment of this specification provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, where the processor executes the steps of the method described in any one of the above.

[0052] In a sixth aspect, an embodiment of this specification provides a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, the steps of the method described in any one of the above are implemented.

[0053] The beneficial effects of the embodiments of this specification are as follows:

[0054] The shift fault identification method for a hydraulic transmission provided by the embodiments of this specification comprehensively examines the performance of each stage and the overall performance during the shift process by obtaining the overall evaluation indicators and stage evaluation indicators during the shift process of the hydraulic transmission, and determines whether there is a fault affecting driving safety through the first preset fault identification model. When there is no fault affecting driving safety, it determines whether there is a fault reducing the product reliability or lifespan through the second preset fault identification model. It can be seen that in this solution, by comprehensively monitoring the indicators during the shift process, various faults during the shift process can be identified, making the identification of shift faults more comprehensive and accurate, thereby improving the reliability of the hydraulic transmission. Description of the Drawings

[0055] By reading the detailed description of the preferred embodiments below, various other advantages and benefits will become clear to those of ordinary skill in the art. The drawings are only for the purpose of showing the preferred embodiments and are not considered to limit this specification. Moreover, throughout the drawings, the same reference numerals are used to represent the same components. In the drawings:

[0056] Figure 1 It is a flowchart of a shift fault identification method for a hydraulic transmission provided by an embodiment of this specification;

[0057] Figure 2 It is a schematic diagram for dividing stages of a shift process provided by an embodiment of this specification;

[0058] Figure 3 It is a schematic structural diagram of a neural network model provided by an embodiment of this specification;

[0059] Figure 4 Flow chart of a fault identification model training method provided by an embodiment of this specification;

[0060] Figure 5 Schematic diagram of a shift fault identification device for a hydraulic transmission provided by an embodiment of this specification;

[0061] Figure 6 Schematic diagram of a fault identification model training device provided by an embodiment of this specification;

[0062] Figure 7 Schematic diagram of an electronic device provided by an embodiment of this specification. Detailed implementation manners

[0063] To better understand the above technical solutions, the technical solutions of the embodiments of this specification will be described in detail below through the accompanying drawings and specific embodiments. It should be understood that the specific features in the embodiments of this specification and the embodiments are detailed descriptions of the technical solutions of the embodiments of this specification, rather than limitations on the technical solutions of this specification. Without conflict, the technical features in the embodiments of this specification and the embodiments can be combined with each other.

[0064] An embodiment of this specification provides a shift fault identification method for a hydraulic transmission, as Figure 1 shown, it is a flow chart of the shift fault identification method for a hydraulic transmission provided by an embodiment of this specification. The method includes the following steps:

[0065] Step S11: Obtain the overall evaluation index and the stage evaluation index during the shift process of the hydraulic transmission. Among them, the overall evaluation index is an index used to characterize the overall performance of the shift process, and the stage evaluation index is an index used to characterize the stage performance of each shift stage included in the shift process;

[0066] Step S12: Input the overall evaluation index into a first preset fault identification model to obtain a first fault identification result, where the first fault identification result is used to characterize whether a fault affecting driving safety occurs during the shift process;

[0067] Step S13: If the first fault identification result indicates that no fault affecting driving safety occurs during the shift process, input the overall evaluation index and the stage evaluation index into a second preset fault identification model to obtain a second fault identification result.

[0068] The shift fault identification method for a hydraulic transmission provided by an embodiment of this specification can be applied to a vehicle controller or an independent electronic device capable of communicating with the vehicle. For example, the vehicle sends the data collected for shift fault identification to the electronic device so that the electronic device analyzes the received data to obtain a shift fault identification result.

[0069] To identify faults in the shifting process, various indicators of the hydraulic transmission during the shifting process are obtained through step S11. In the embodiments of this specification, the shifting process can be shifting under any working conditions, such as power upshifting, power downshifting, no-power upshifting, no-power downshifting, etc. To comprehensively detect the shifting process, in the embodiments of this specification, for one shifting process, it is necessary to obtain the overall evaluation indicator and the stage evaluation indicator during this shifting process. The overall evaluation indicator and the stage evaluation indicator can be obtained by extracting the rotational speeds of the input shaft, turbine shaft, and output shaft of the AT. In this way, the identification of shift faults can get rid of the dependence on the oil pressure signal, eliminate the hidden dangers of the non-linearity and instability of the oil pressure signal, improve the scope and accuracy of shift fault identification, and at the same time, the number of oil pressure sensors on the AT product can be reduced, which is beneficial to cost control.

[0070] In the specific implementation process, the overall evaluation indicator, as an indicator for evaluating the overall performance of the shifting process, can include one or more of the following indicators: shift duration T whole , peak value of the change rate of the turbine shaft speed during the shifting process , peak value of the change rate of the output shaft speed during the shifting process

[0071] To obtain stage indicators, as Figure 2 shown, in the embodiments of this specification, the shifting process is divided into the following stages in chronological order: pre-charging stage, torque stage, slow turbine speed decline stage, rapid turbine speed decline stage, and synchronization stage.

[0072] Among them, the pre-charging oil stage is the stage from when the controller issues a shifting instruction to when the shifting valve opens and pushes the piston to eliminate the clearance of the friction plates of the clutch or brake; the torque stage is the stage from when the clutch or brake starts to transmit torque, but the transmission still maintains the original gear ratio; the stage of slow turbine speed decline is the stage when the clutch or brake transmits torque and the gear ratio of the transmission changes slowly (such as the stage when the gear ratio change rate is less than the threshold); the stage of rapid turbine speed decline is the stage when the clutch or brake transmits torque and the gear ratio of the transmission changes rapidly (such as the stage when the gear ratio change rate is greater than the threshold), and this stage is usually accompanied by the rapid release of the inertial energy of the driveline and is most likely to generate shifting shock; the synchronization stage is the stage when the gear ratio of the transmission reaches and stabilizes at the target gear ratio. It should be noted that when dividing the stage of slow turbine speed decline and the stage of rapid turbine speed decline, the size of the threshold can be set according to actual needs and is not limited here.

[0073] After dividing the shifting process into stages, the stage evaluation indexes can be determined according to each divided stage. In the specific implementation process, the stage evaluation indexes can include one or more of the following indexes: the total amount of turbine speed decline in the pre-charging oil stage The total amount of turbine speed decline within one response cycle after the end of the pre-charging oil stage The actual time t in the torque stage torque and the total amount of turbine speed decline in the torque stage The actual time t in the stage of slow turbine speed decline slow and the total amount of turbine speed decline within one response cycle after the end of the stage of slow turbine speed decline and the actual time t in the stage of rapid turbine speed decline fast .

[0074] It should be noted that since the hydraulic system needs a certain time to respond, therefore, the total amount of turbine speed decline within one response cycle after the end of the pre-charging oil stage is used to characterize the delayed response of the hydraulic system. Similarly, considering the response delay of the hydraulic system, it is necessary to obtain the total amount of turbine speed decline within one response cycle after the end of the stage of slow turbine speed decline It can be seen that the embodiments of this specification can comprehensively monitor the indexes in the shifting process, thereby ensuring the accuracy of the final fault identification.

[0075] In the embodiments of this specification, according to the severity of the shifting fault, the shifting process faults can be divided into two categories: the first category is the faults that may cause the loss of product function or driving safety; the second category is the faults that have no impact on the transmission function, but frequent occurrence of such faults for a long time may reduce the product reliability and lifespan, such as reducing the reliability of the transmission system components and friction plates. Since the first category of faults is relatively serious and measures need to be taken in a timely manner to prevent driving hazards, the first category of faults can be identified in real time. For the second category of faults, since they do not have a great impact on the current driving, in order to save computing resources, the second category of faults can be identified offline. For example, a fault identification period can be set for the identification of the second category of faults, such as calculating once every day, or once every week, etc. Of course, the second category of faults can also be identified in real time, which is not limited here.

[0076] In the specific implementation process, the first category of faults can be identified through the first preset fault identification model, and the second category of faults can be identified through the second preset fault identification model. Among them, the first category of faults can be identified based on the overall evaluation index, and the second category of faults can be identified based on the overall evaluation index and the stage evaluation index. Of course, which specific index is used to identify which kind of fault can be set according to actual needs. For example, the second category of faults can be identified through the stage evaluation index, or the first category of faults can be identified through some overall evaluation indexes and some stage evaluation indexes, or a specific fault included in the first category of faults or the second category of faults can be identified through the specified evaluation index, etc., which is not limited here. For the sake of illustration, here, taking the identification of the first category of faults through the overall evaluation index and the identification of the second category of faults through the overall evaluation index and the stage evaluation index as an example for explanation.

[0077] In the embodiments of this specification, the faults in the AT shifting process include the following: normal shifting, power interference, power interruption, large impact, general impact, large slip friction, and general slip friction. Next, for each of the above faults, the parameter changes corresponding to this fault in the upshift and downshift processes will be specifically described.

[0078] Normal shifting: During the upshift process, the angular acceleration of the output shaft is less than the first preset angular acceleration of the output shaft, and the output speed remains basically unchanged; during the downshift process, the output angular acceleration is less than the first preset angular acceleration of the output shaft, and the turbine speed rises steadily.

[0079] Power interference: During the upshift process, the angular acceleration of the output shaft is greater than or equal to the second preset angular acceleration of the output shaft, and the output speed shows an abnormal decrease; during the downshift process, the angular acceleration of the output shaft is greater than or equal to the second preset angular acceleration of the output shaft, and the turbine speed shows an abnormal decrease.

[0080] Power interruption: During the upshift process, the shift duration is greater than the first preset shift duration, and the angular acceleration of the output shaft is less than the third preset angular acceleration of the output shaft; during the downshift process, the shift duration is greater than the preset shift duration, and the angular acceleration of the output shaft is less than the third preset angular acceleration of the output shaft.

[0081] Large shock: During the upshift process, the ratio of the angular acceleration of the turbine shaft to the angular acceleration of the output shaft is greater than the first preset ratio, and the rotational speed of the output shaft shows an abnormal decrease; during the downshift process, the ratio of the angular acceleration of the turbine shaft to the angular acceleration of the output shaft is greater than the first preset ratio, and the actual time used in the torque stage is greater than the first preset duration.

[0082] General shock: During the upshift process, the ratio of the angular acceleration of the turbine shaft to the angular acceleration of the output shaft is greater than the second preset ratio, and the rotational speed of the output shaft shows an abnormal decrease; during the downshift process, the ratio of the angular acceleration of the turbine shaft to the angular acceleration of the output shaft is greater than the second preset ratio, and the actual time used in the torque stage is greater than the second preset duration. Among them, the second preset ratio is less than the first preset ratio, and the second preset duration is less than the first preset duration.

[0083] Large slip friction: During the upshift process, the shift duration is greater than the second preset shift duration, and the ratio of the angular acceleration of the turbine shaft to the angular acceleration of the output shaft is less than the third preset ratio; during the downshift process, the shift duration is greater than the second preset shift duration, and the ratio of the angular acceleration of the turbine shaft to the angular acceleration of the output shaft is less than the third preset ratio.

[0084] General slip friction: During the upshift process, the shift duration is greater than the third preset shift duration, and the ratio of the angular acceleration of the turbine shaft to the angular acceleration of the output shaft is less than the fourth preset ratio; during the downshift process, the shift duration is greater than the third preset shift duration, and the ratio of the angular acceleration of the turbine shaft to the angular acceleration of the output shaft is less than the fourth preset ratio. Among them, the third preset shift duration is less than the second preset shift duration, and the fourth preset ratio is less than the third preset ratio.

[0085] It should be noted that the specific values of the above-mentioned first preset angular acceleration of the output shaft, second preset angular acceleration of the output shaft, third preset angular acceleration of the output shaft, first preset shift duration, second preset shift duration, third preset shift duration, first preset ratio, second preset ratio, third preset ratio, fourth preset ratio, first preset duration, and second preset duration can be set according to actual needs. For different manufacturers and different vehicle models, the specific values of the above parameters may vary.

[0086] In the embodiments of this specification, the first type of faults includes: power interference and power interruption; the second type of faults includes: large shock, general shock, large slip friction, and general slip friction.

[0087] For the first type of fault, it can be identified by the first preset fault identification model. Through step S12, the overall evaluation index during the gearshift process is obtained, and the overall evaluation index of this gearshift process is input into the first fault identification model to output the first fault identification result. Herein, the first fault identification result is used to indicate whether the first type of fault occurs during this gearshift process. Specifically, the first fault identification result may include three situations: normal gearshift, power interference, and power interruption. When the first fault identification result is power interference or power interruption, it indicates that the first type of fault occurs during the gearshift process. At this time, the TCU (Transmission Control Unit) can directly implement active fault protection measures. When the first fault identification result is normal gearshift, it indicates that the first type of fault does not occur during the gearshift process. The gearshift process can be further evaluated by the second preset fault to determine whether there is a second type of fault, so as to locate the part that causes poor gearshift quality based on the identified second type of fault, provide a basis for the correction direction of the gearshift index, and play a preventive role in improving the reliability of the system.

[0088] Specifically, when identifying the fault of the gearshift process through step S13, since the second preset fault identification model is more complex, in the embodiments of this specification, both the overall evaluation index and the stage evaluation index can be used as the input of the second preset fault identification model to obtain the second fault identification result. Among them, the second fault identification result includes one of the above 4 types of second faults.

[0089] In addition, since the first type of fault will affect driving safety, it is necessary to detect the first type of fault in real time, that is, the first preset fault identification model continuously identifies the first type of fault during the gearshift process. Since the second type of fault has little impact on driving safety, the second preset fault identification model can identify the second fault offline to save computing resources. Of course, the second preset fault identification model can also identify the second type of fault in real time, which is not limited here.

[0090] To illustrate the first preset fault identification model and the second preset fault identification model provided in the embodiments of this specification, the formation processes of the first preset fault identification model and the second preset fault identification model are described below.

[0091] In the specific implementation process, the first preset fault identification model and the second preset fault identification model are obtained in the following manner: Obtain a training sample set, where each sample data in the training sample set includes the overall evaluation index and the stage evaluation index during a gearshift process; Based on the corresponding relationship between the preset evaluation index and the fault type, determine the fault type of each sample data as the label information of each sample data; Construct the first initial fault identification model and the second initial fault identification model; Based on the overall performance index corresponding to each sample data and the label information of each sample data, train the first initial fault identification model to obtain the trained first preset fault identification model; Based on the overall evaluation index, the stage evaluation index corresponding to each sample data, and the label information of each sample data, train the second initial fault identification model to obtain the trained second preset fault identification model.

[0092] Specifically, each sample in the training sample set corresponds to an overall evaluation index and a stage evaluation index during a gearshift process. Specifically, the overall evaluation index and the stage evaluation index during each gearshift process of the target vehicle are statistically analyzed. For example, each gearshift process of the target vehicle within half a year is recorded, including various indexes during the upshift process and the downshift process, and the indexes corresponding to each gearshift process are used as a sample data.

[0093] In the embodiments of this specification, the model training is carried out in a supervised manner. Therefore, each sample needs to be labeled. In the specific implementation process, for each vehicle model, a set of standard overall evaluation indexes and stage indexes can be set. By comparing the indexes in the actually measured sample data with the standard indexes, the label information of each sample is determined. Among them, the standard overall evaluation index and the stage evaluation index can be the indexes during normal gearshift determined according to empirical values or experimental values. For different gearshift faults, the errors between each index and the standard index are also different. In the embodiments of this specification, the corresponding relationship between the preset evaluation index and the fault type can be used to label the label information for each sample data. The corresponding relationship between the preset evaluation index and the fault type is shown in Table 1.

[0094] Table 1

[0095]

[0096] Among them, "too small" in Table 1 indicates that the corresponding index in the sample is less than the standard index, and the difference between the standard index and the sample index is less than the first preset value. Correspondingly, "too large" in Table 1 indicates that the corresponding index in the sample is greater than the standard index, and the difference between the sample index and the standard index is greater than the second preset value. "Slightly smaller" in Table 1 indicates that the corresponding index in the sample is less than the standard index, and the difference between the standard index and the sample index is less than the third preset value. Correspondingly, "slightly larger" in Table 1 indicates that the corresponding index in the sample is greater than the standard index, and the difference between the sample index and the standard index is greater than the fourth preset value. "Large" in Table 1 indicates that the corresponding index in the sample is greater than the standard index, and the difference between the sample index and the standard index is greater than the fifth preset value. Correspondingly, "small" in Table 1 indicates that the corresponding index in the sample is less than the standard index, and the difference between the standard index and the sample index is less than the sixth preset value. It should be noted that the specific values of the above first preset value to sixth preset value can be set according to actual needs and are not limited here.

[0097] For example, if the indicators of a sample data, compared with the standard indicators, meet the fault type of dynamic interference, the label information of the sample is marked as dynamic interference. In this way, the label information of each training sample can be determined.

[0098] Before model training, it is necessary to first construct an initial model, that is, to construct a first initial fault identification model and a second initial fault identification model. The type of the model can be selected according to actual needs and is not limited here. For the convenience of description, in the embodiments of this specification, the first initial fault identification model and the second initial fault identification model are both neural network models as an example for explanation.

[0099] As Figure 3 shown, it is a schematic structural diagram of a neural network model, where X is an m-dimensional vector of the input layer, H is a k-dimensional vector of the hidden layer, Y is an n-dimensional vector of the output layer, and W 1 is the weight matrix from the input layer to the hidden layer (with a size of l×m), b 1 is the l-dimensional threshold vector from the input layer to the hidden layer, W 2 is the weight matrix from the hidden layer to the output layer (with a size of n×l), b 2 is the n-dimensional threshold vector from the hidden layer to the output layer, is the logsig activation function.

[0100] For the input layer of the neural network, the number of input layer nodes depends on the dimension m of the input vector. In the embodiments of this specification, the first type of fault is identified based on three overall evaluation indicators, and the second type of fault is identified based on three overall evaluation indicators and seven stage evaluation indicators. Therefore, for the first initial fault identification model, three overall evaluation indicators are used as the input of the model, that is, m = 3; for the second initial fault identification model, three overall evaluation indicators and seven stage evaluation indicators are used as the input of the model, that is, m = 10. Of course, the number of input nodes of the model can be adjusted according to the actual situation, which is not limited here.

[0101] For the hidden layer of the neural network, the number of stages of the hidden layer can be adjusted according to the error performance during the training process. In the embodiments of this specification, the first initial fault identification model selects 4 nodes, and the second initial fault identification model selects 9 nodes. Of course, the number of nodes in the hidden layer can also be set to other values according to actual needs, which is not limited here.

[0102] For the output layer of the neural network, in the embodiments of this specification, a multi-output type can be selected, that is, the number of output nodes is equal to the number of types. For the first initial fault identification model, since the model is used to identify the first type of fault and uses the second preset fault identification model for further fault identification when it is identified that it is not the first type of fault. Therefore, the output of the first initial fault identification model includes three categories: normal shifting, power interference, and power interruption, that is, the output dimension of the first initial fault identification model is 3. For the second initial fault identification model, since this model is used to identify the second type of fault, and the second type of fault includes four categories: large impact, general impact, large sliding friction, and general sliding friction, the output dimension of the second initial fault identification model is 4.

[0103] After constructing the initial models, the first initial fault identification model and the second initial fault identification model are trained based on the training sample set. It should be noted that to ensure the final training effect of the models, the more samples in the training sample set, the better. In the embodiments of this specification, the sample data of each fault type in the training sample set is not less than 100 groups. During the specific model training process, the training accuracy and classification accuracy can be set according to actual needs. In the embodiments of this specification, the training accuracy can be set to 0.1, and the output accuracy can be set to be greater than 85%.

[0104] After the initial model training is completed, the first preset fault identification model and the second preset fault identification model are obtained, and the first preset fault identification model and the second preset fault identification model can be applied to the fault identification in the shifting process. Different output results correspond to different types. For example, when the output result of the first preset fault identification model is 001, the corresponding type is normal shifting; when the output result of the first preset fault identification model is 010, the corresponding type is power interference; when the output result of the first preset fault identification model is 100, the corresponding type is power interruption. When the output result of the second preset fault identification model is 0001, the corresponding type is large impact; when the output result of the second preset fault identification model is 0010, the corresponding type is general impact; when the output result of the second preset fault identification model is 0100, the corresponding type is large slip friction; when the output result of the second preset fault identification model is 1000, the corresponding type is general slip friction.

[0105] Based on the same inventive concept, an embodiment of the present specification provides a method for training a fault identification model. The fault identification model is used to identify shifting faults of a hydraulic transmission. The fault identification model includes a first preset fault identification model and a second preset fault identification model, as Figure 4 shown, the method includes:

[0106] Step S41: Obtain a training sample set. Each sample data in the training sample set includes an overall evaluation index and a stage evaluation index during a shifting process. Among them, the overall evaluation index is an index used to characterize the overall performance of the shifting process, and the stage evaluation index is an index used to characterize the stage performance of each shifting stage included in the shifting process;

[0107] Step S42: Construct a first initial fault identification model and a second initial fault identification model;

[0108] Step S43: Train the first initial fault identification model based on the overall performance index corresponding to each sample data and the label information of each sample data to obtain the trained first preset fault identification model;

[0109] Step S44: Train the second initial fault identification model based on the overall evaluation index, the stage evaluation index corresponding to each sample data, and the label information of each sample data to obtain the trained second preset fault identification model.

[0110] Optionally, after obtaining the training sample set, the method further includes:

[0111] Based on the correspondence between the preset evaluation indicators and the fault types, determine the fault type of each sample data as the label information of each sample data.

[0112] Optionally, the overall evaluation indicator includes one or more of the following indicators: the shift duration of the shifting process, the peak value of the change rate of the turbine shaft speed during the shifting process, and the peak value of the change rate of the output shaft speed during the shifting process.

[0113] Optionally, the shifting process includes the following stages in chronological order: pre-oiling stage, torque stage, slow turbine speed decline stage, rapid turbine speed decline stage, and synchronization stage;

[0114] The stage evaluation indicators include one or more of the following indicators: the total amount of turbine speed decline in the pre-oiling stage, the total amount of turbine speed decline within one response cycle after the end of the pre-oiling stage, the actual time used in the torque stage, the total amount of turbine speed decline in the torque stage, the actual time used in the slow turbine speed decline stage, the total amount of turbine speed decline within one response cycle after the end of the slow turbine speed decline stage, and the actual time used in the rapid turbine speed decline stage.

[0115] Regarding the above method, the specific implementation manners of each step have been described in detail in the embodiments of the fault identification method of the hydraulic transmission provided in this specification, and will not be elaborated here.

[0116] Based on the same inventive concept as the shifting fault identification method of the hydraulic transmission, the embodiments of this specification provide a shifting fault identification device for a hydraulic transmission, as Figure 5 shown, the device includes:

[0117] An acquisition module 51, configured to acquire the overall evaluation indicator and the stage evaluation indicator during the shifting process of the hydraulic transmission, where the overall evaluation indicator is an indicator used to characterize the overall performance of the shifting process, and the stage evaluation indicator is an indicator used to characterize the stage performance of each shifting stage included in the shifting process;

[0118] A first processing module 52, configured to input the overall evaluation indicator into a first preset fault identification model to obtain a first fault identification result, where the first fault identification result is used to characterize whether a fault affecting driving safety occurs during the shifting process;

[0119] A second processing module 53, configured to input the overall evaluation indicator and the stage evaluation indicator into a second preset fault identification model to obtain a second fault identification result when the first fault identification result indicates that no fault affecting driving safety occurs during the shifting process.

[0120] Optionally, the overall evaluation index includes one or more of the following indexes: the shift duration of the shifting process, the peak value of the change rate of the turbine shaft speed during the shifting process, and the peak value of the change rate of the output shaft speed during the shifting process.

[0121] Optionally, the shifting process includes the following stages in chronological order: pre-oiling stage, torque stage, slow turbine speed decline stage, rapid turbine speed decline stage, and synchronization stage;

[0122] The stage evaluation index includes one or more of the following indexes: the total turbine speed decline amount in the pre-oiling stage, the total turbine speed decline amount within one response cycle after the end of the pre-oiling stage, the actual time used in the torque stage, the total turbine speed decline amount in the torque stage, the actual time used in the slow turbine speed decline stage, the total turbine speed decline amount within one response cycle after the end of the slow turbine speed decline stage, and the actual time used in the rapid turbine speed decline stage.

[0123] Optionally, the first preset fault identification model and the second preset fault identification model are obtained by the following methods:

[0124] Obtain a training sample set, and each sample data in the training sample set includes the overall evaluation index and the stage evaluation index during a shifting process;

[0125] Based on the correspondence between the preset evaluation index and the fault type, determine the fault type of each sample data as the label information of each sample data;

[0126] Construct a first initial fault identification model and a second initial fault identification model;

[0127] Based on the overall performance index corresponding to each sample data and the label information of each sample data, train the first initial fault identification model to obtain the trained first preset fault identification model;

[0128] Based on the overall evaluation index, stage evaluation index corresponding to each sample data, and the label information of each sample data, train the second initial fault identification model to obtain the trained second preset fault identification model.

[0129] Regarding the above device, the specific functions of each module have been described in detail in the embodiments of the shifting fault identification method of the hydraulic transmission provided in this specification, and will not be elaborated here.

[0130] Based on the same inventive concept as the fault identification model training method, an embodiment of this specification provides a fault identification model training device. The fault identification model is used to identify shifting faults of a hydraulic transmission. The fault identification model includes a first preset fault identification model and a second preset fault identification model. As Figure 6 shown, the device includes:

[0131] A sample acquisition module 61, configured to acquire a training sample set. Each sample data in the training sample set includes an overall evaluation index and a stage evaluation index during a shifting process. Among them, the overall evaluation index is an index used to characterize the overall performance of the shifting process, and the stage evaluation index is an index used to characterize the stage performance of each shifting stage included in the shifting process;

[0132] A model construction module 62, configured to construct a first initial fault identification model and a second initial fault identification model;

[0133] A first model training module 63, configured to train the first initial fault identification model based on the overall performance index corresponding to each sample data and the label information of each sample data, to obtain the trained first preset fault identification model;

[0134] A second model training module 64, configured to train the second initial fault identification model based on the overall evaluation index, the stage evaluation index, and the label information of each sample data corresponding to each sample data, to obtain the trained second preset fault identification model.

[0135] Optionally, the device further includes:

[0136] A label determination module, configured to determine the fault type of each sample data as the label information of each sample data based on the corresponding relationship between the preset evaluation index and the fault type.

[0137] Optionally, the overall evaluation index includes one or more of the following indexes: the shifting duration of the shifting process, the peak value of the change rate of the turbine shaft speed during the shifting process, and the peak value of the change rate of the output shaft speed during the shifting process.

[0138] Optionally, the shifting process includes the following stages in chronological order: a pre-oiling stage, a torque stage, a slow turbine speed decline stage, a rapid turbine speed decline stage, and a synchronization stage;

[0139] The stage evaluation indicators include one or more of the following indicators: the total decrease in turbine speed during the pre-oiling stage, the total decrease in turbine speed within one response cycle after the end of the pre-oiling stage, the actual time taken for the torque stage, the total decrease in turbine speed during the torque stage, the actual time taken for the slow turbine speed decrease stage, the total decrease in turbine speed within one response cycle after the end of the slow turbine speed decrease stage, and the actual time taken for the rapid turbine speed decrease stage.

[0140] Regarding the above device, the specific functions of each module have been described in detail in the embodiments of the shift fault identification method for the hydraulic transmission provided in this specification, and will not be elaborated here.

[0141] Based on the same inventive concept as the shift fault identification method and the fault identification model training method for the hydraulic transmission in the foregoing embodiments, the embodiments of this specification also provide an electronic device, such as Figure 7 shown, including a memory 404, a processor 402, and a computer program stored on the memory 404 and executable on the processor 402. When the processor 402 executes the program, it implements the steps of any one of the shift fault identification method and the fault identification model training method for the hydraulic transmission described above.

[0142] Among them, in Figure 7 , the bus architecture (represented by bus 400), bus 400 can include any number of interconnected buses and bridges. Bus 400 links various circuits including one or more processors represented by processor 402 and a memory represented by memory 404 together. Bus 400 can also link various other circuits such as peripheral devices, voltage regulators, and power management circuits, which are well known in the art, and therefore will not be further described herein. Bus interface 406 provides an interface between bus 400 and receiver 401 and transmitter 403. Receiver 401 and transmitter 403 can be the same element, that is, a transceiver, providing a unit for communicating with various other devices on the transmission medium. Processor 402 is responsible for managing bus 400 and general processing, while memory 404 can be used to store data used by processor 402 when performing operations.

[0143] Based on the same inventive concept as the shift fault identification method and the fault identification model training method for the hydraulic transmission in the foregoing embodiments, the embodiments of this specification also provide a computer-readable storage medium, on which a computer program is stored. When the program is executed by a processor, it implements the steps of any one of the shift fault identification method and the fault identification model training method for the hydraulic transmission described above.

[0144] This specification is described with reference to the flowcharts and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the specification. It should be understood that each flow and / or block in the flowchart and / or block diagram, and combinations of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processors of general-purpose computers, special-purpose computers, embedded processors, or other programmable data processing devices to produce a machine, such that the instructions executed by the processors of the computer or other programmable data processing devices produce a device for implementing the functions specified in one or more of the flows Figure 1 one or more of the flows and / or blocks Figure 1 or one or more of the blocks.

[0145] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to operate in a specific manner, such that the instructions stored in the computer-readable memory produce a manufacture including an instruction device that implements the functions specified in one or more of the flows Figure 1 one or more of the flows and / or blocks Figure 1 or one or more of the blocks.

[0146] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one or more of the flows Figure 1 one or more of the flows and / or blocks Figure 1 or one or more of the blocks.

[0147] Although the preferred embodiments of this specification have been described, additional changes and modifications can be made to these embodiments by those skilled in the art once they learn of the basic creative concept. Therefore, the appended claims are intended to be construed to include the preferred embodiments and all changes and modifications that fall within the scope of this specification.

[0148] Obviously, those skilled in the art can make various changes and modifications to this specification without departing from the spirit and scope of this specification. Thus, if these modifications and variations of this specification fall within the scope of the claims of this specification and their equivalent technologies, this specification is also intended to include these changes and modifications.

Claims

1. A method for identifying shifting faults of a hydraulic transmission, characterized in that, The method includes: Obtaining an overall evaluation index and stage evaluation indexes during the gear shifting process of a hydraulic transmission, wherein the overall evaluation index is an index used to characterize the overall performance of the gear shifting process, and the stage evaluation indexes are indexes used to characterize the stage performance of each gear shifting stage included in the gear shifting process; Inputting the overall evaluation index into a first preset fault identification model to obtain a first fault identification result, where the first fault identification result is used to characterize whether a fault affecting driving safety occurs during the gear shifting process; If the first fault identification result indicates that no fault affecting driving safety occurs during the gear shifting process, inputting the overall evaluation index and the stage evaluation indexes into a second preset fault identification model to obtain a second fault identification result; Wherein, the first preset fault identification model and the second preset fault identification model are obtained through the following methods: Obtaining a training sample set, where each sample data in the training sample set includes the overall evaluation index and stage evaluation indexes during a gear shifting process; Based on the corresponding relationship between the preset evaluation indexes and fault types, determining the fault type of each sample data as the label information of each sample data; Constructing a first initial fault identification model and a second initial fault identification model; Training the first initial fault identification model based on the overall performance index corresponding to each sample data and the label information of each sample data to obtain the trained first preset fault identification model; Training the second initial fault identification model based on the overall evaluation index, stage evaluation indexes corresponding to each sample data, and the label information of each sample data to obtain the trained second preset fault identification model.

2. The method according to claim 1, characterized in that, The overall evaluation index includes one or more of the following indexes: the gear shifting duration of the gear shifting process, the peak value of the change rate of the turbine shaft speed during the gear shifting process, and the peak value of the change rate of the output shaft speed during the gear shifting process.

3. The method according to claim 1 or 2, characterized in that, The gear shifting process includes the following stages in chronological order: a pre-oiling stage, a torque stage, a stage of slow turbine speed decline, a stage of rapid turbine speed decline, and a synchronization stage; The stage evaluation indexes include one or more of the following indexes: the total amount of turbine speed decline in the pre-oiling stage, the total amount of turbine speed decline within one response cycle after the end of the pre-oiling stage, the actual time used in the torque stage, the total amount of turbine speed decline in the torque stage, the actual time used in the stage of slow turbine speed decline, the total amount of turbine speed decline within one response cycle after the end of the stage of slow turbine speed decline, and the actual time used in the stage of rapid turbine speed decline.

4. A method for training a fault identification model, characterized in that, The fault identification model is used to identify the gear shifting faults of a hydraulic transmission. The fault identification model includes a first preset fault identification model and a second preset fault identification model. The method includes: Obtain a training sample set, where each sample data in the training sample set includes an overall evaluation index and a stage evaluation index during a gear shifting process. Among them, the overall evaluation index is an index used to characterize the overall performance of the gear shifting process, and the stage evaluation index is an index used to characterize the stage performance of each gear shifting stage included in the gear shifting process; Construct a first initial fault identification model and a second initial fault identification model; Based on the overall performance index corresponding to each sample data and the label information of each sample data, train the first initial fault identification model to obtain the trained first preset fault identification model; Based on the overall evaluation index, stage evaluation index corresponding to each sample data, and the label information of each sample data, train the second initial fault identification model to obtain the trained second preset fault identification model.

5. The method according to claim 4, characterized in that, After obtaining the training sample set, the method further includes: Based on the correspondence between the preset evaluation index and the fault type, determine the fault type of each sample data as the label information of each sample data.

6. A shift fault identification device for a hydrodynamic transmission, characterized in that, The device includes: An acquisition module, configured to acquire an overall evaluation index and a stage evaluation index during the gear shifting process of a hydraulic transmission. Among them, the overall evaluation index is an index used to characterize the overall performance of the gear shifting process, and the stage evaluation index is an index used to characterize the stage performance of each gear shifting stage included in the gear shifting process; A first processing module, configured to input the overall evaluation index into a first preset fault identification model to obtain a first fault identification result, where the first fault identification result is used to characterize whether a fault affecting driving safety occurs during the gear shifting process; A second processing module, configured to, when the first fault identification result indicates that no fault affecting driving safety occurs during the gear shifting process, input the overall evaluation index and the stage evaluation index into a second preset fault identification model to obtain a second fault identification result; Among them, the first preset fault identification model and the second preset fault identification model are obtained through the following methods: Obtain a training sample set, where each sample data in the training sample set includes an overall evaluation index and a stage evaluation index during a gear shifting process; Based on the correspondence between the preset evaluation index and the fault type, determine the fault type of each sample data as the label information of each sample data; Construct a first initial fault identification model and a second initial fault identification model; Based on the overall performance index corresponding to each sample data and the label information of each sample data, train the first initial fault identification model to obtain the trained first preset fault identification model; Based on the overall evaluation index, stage evaluation index corresponding to each sample data, and the label information of each sample data, train the second initial fault identification model to obtain the trained second preset fault identification model.

7. A fault identification model training device, characterized in that The fault identification model is used to identify the shifting faults of a hydraulic transmission. The fault identification model includes a first preset fault identification model and a second preset fault identification model. The device includes: A sample acquisition module, configured to acquire a training sample set. Each sample data in the training sample set includes an overall evaluation index and a stage evaluation index during a shifting process. Among them, the overall evaluation index is an index used to characterize the overall performance of the shifting process, and the stage evaluation index is an index used to characterize the stage performance of each shifting stage included in the shifting process; A model construction module, configured to construct a first initial fault identification model and a second initial fault identification model; A first model training module, configured to train the first initial fault identification model based on the overall performance index corresponding to each sample data and the label information of each sample data, to obtain the trained first preset fault identification model; A second model training module, configured to train the second initial fault identification model based on the overall evaluation index, the stage evaluation index corresponding to each sample data, and the label information of each sample data, to obtain the trained second preset fault identification model.

8. An electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, the steps of the method according to any one of claims 1-5 are implemented.

9. A computer-readable storage medium, on which a computer program is stored. When the program is executed by a processor, the steps of the method according to any one of claims 1-5 are implemented.

Images