A method and system for early warning of electric drive system failure

Abstract

This invention provides a method and system for early warning of faults in electric drive systems. The method includes: when the electric drive system is detected to be running, collecting the operating parameters of the electric drive system in real time and generating a corresponding target dataset based on the operating parameters; constructing a corresponding fault prediction model based on the target dataset and a preset neural network, and inputting the operating parameters into the fault prediction model so that the fault prediction model can determine in real time whether there is a fault in the electric drive system; if the fault prediction model determines in real time that there is a fault in the electric drive system, immediately detecting the fault type of the electric drive system through the fault prediction model, and generating a corresponding early warning signal based on the fault type. This invention can effectively reduce the cost of early warning and correspondingly improve the user experience.

Description

Technical Field

[0001] This invention relates to the field of new energy electric vehicle technology, and in particular to a method and system for early warning of faults in electric drive systems. Background Technology

[0002] With the advancement of science and technology and the rapid development of productivity, new energy electric vehicle technology has become increasingly mature and has gradually gained people's recognition. It has been popularized in people's daily lives, greatly facilitating people's lives.

[0003] Among them, the electric drive system is one of the core components of new energy electric vehicles. It mainly consists of components such as batteries, motors, and electronic controls, and is used to convert the electric power input from the power battery pack into mechanical energy to drive the vehicle.

[0004] Because existing electric drive systems have numerous components, various types of faults may occur during actual operation. Therefore, it is essential to predict electric drive system faults in advance. Specifically, most existing technologies predict whether the electric drive system will fail by establishing a mathematical model of the electric drive system and then performing real-time state estimation and fault diagnosis on this model. However, the above prediction method requires a large amount of computing resources and numerous data acquisition devices, and also requires a significant amount of time to build the mathematical model of the electric drive system. This results in high early warning costs and reduces the user experience. Summary of the Invention

[0005] Based on this, the purpose of the present invention is to provide a method and system for early warning of faults in electric drive systems, so as to solve the technical problem of high early warning costs in the prior art.

[0006] The first aspect of the present invention proposes:

[0007] A fault early warning method for an electric drive system, wherein the method includes:

[0008] When the electric drive system is detected to be running, the operating parameters of the electric drive system are collected in real time, and a corresponding target dataset is generated based on the operating parameters.

[0009] A corresponding fault prediction model is constructed based on the target dataset and a preset neural network, and the operating parameters are input into the fault prediction model so that the fault prediction model can determine in real time whether there is a fault in the electric drive system.

[0010] If the fault prediction model determines in real time that there is a fault in the electric drive system, the fault prediction model will immediately detect the fault type of the electric drive system and generate a corresponding warning signal based on the fault type.

[0011] The beneficial effects of this invention are as follows: by real-time detection of the operating parameters of the electric drive system, the working status of the electric drive system can be known accordingly. Furthermore, based on the real-time collected operating parameters and the preset neural network, a fault prediction model adapted to the current electric drive system can be trained. Based on this, the fault prediction model can be used to infer in real time whether the current electric drive system will malfunction. Moreover, if so, a corresponding warning signal is immediately generated, thereby enabling the driver to know in advance. In this process, a large amount of computing resources and data acquisition equipment are saved, reducing the cost of fault prediction and improving the user experience.

[0012] Furthermore, the step of generating the corresponding target dataset based on the running parameters includes:

[0013] When the operating parameters are obtained, the operating parameters are preprocessed, and several parameter values ​​contained in the preprocessed operating parameters are extracted.

[0014] The parameter values ​​are converted into corresponding parameter vectors, and the parameter vectors are integrated into the same dataset to generate the target dataset, which is unique.

[0015] Furthermore, the step of constructing a corresponding fault prediction model based on the target dataset and a preset neural network includes:

[0016] Based on preset rules, the parameter vectors in the target dataset are divided into corresponding training sets and validation sets, and the preset neural network is trained using the training sets to generate the corresponding initial fault prediction model.

[0017] The initial fault prediction model is validated using the validation set, and the validation result of the initial prediction model is judged in real time to determine whether it is qualified.

[0018] If the verification result of the initial prediction model is deemed satisfactory in real time, the training of the initial fault prediction model is completed, and the initial fault prediction model is set as the fault prediction model.

[0019] Furthermore, the step of training the preset neural network using the training set to generate the corresponding initial fault prediction model includes:

[0020] When the training set is obtained, the first parameter vector contained in the training set is extracted and input into the encoding layer of the preset CNN network so that the first parameter vector is encoded into the corresponding parameter matrix through the encoding layer.

[0021] The parameter matrix is ​​input into the parsing layer of the preset CNN network, and the feature values ​​corresponding to the parameter matrix are calculated through the parsing layer.

[0022] The feature values ​​are input into the learning layer of the preset CNN network, and the learning network in the learning layer is adjusted using the feature values ​​to generate the initial fault prediction model.

[0023] Furthermore, the step of adjusting the learning network in the learning layer using the feature values ​​to generate the initial fault prediction model includes:

[0024] Several network nodes contained in the learning network are detected, and each network node is initialized to extract the original network parameters contained in each network node.

[0025] Each of the original network parameters is replaced one by one with the feature value to update the learning network in real time and generate the initial fault prediction model accordingly.

[0026] Furthermore, the method also includes:

[0027] The fault level of the electric drive system is determined in real time based on the fault type, and each fault level corresponds to a fault coefficient.

[0028] The corresponding repair strategy is retrieved from a preset database based on the fault coefficient, and the repair strategy is displayed on the vehicle's dashboard. The fault coefficient is unique.

[0029] Furthermore, the method also includes:

[0030] When a fault is detected in the electric drive system, the geographical location of the vehicle is detected in real time, and a corresponding detection area is constructed with the geographical location as the center and a preset length as the radius.

[0031] Real-time determination of whether a maintenance station exists in the detection area;

[0032] If the presence of the maintenance station within the detection area is detected in real time, the communication link of the maintenance station is sent to the dashboard. There is at least one maintenance station.

[0033] The second aspect of the present invention proposes:

[0034] An electric drive system fault early warning system, wherein the system comprises:

[0035] The acquisition module is used to acquire the operating parameters of the electric drive system in real time when the electric drive system is detected to be running, and to generate a corresponding target dataset based on the operating parameters.

[0036] The judgment module is used to construct a corresponding fault prediction model based on the target dataset and a preset neural network, and input the operating parameters into the fault prediction model so that the fault prediction model can judge in real time whether there is a fault in the electric drive system;

[0037] The detection module is used to immediately detect the fault type of the electric drive system through the fault prediction model if a fault is detected in the electric drive system in real time, and generate a corresponding warning signal based on the fault type.

[0038] Furthermore, the acquisition module is specifically used for:

[0039] When the operating parameters are obtained, the operating parameters are preprocessed, and several parameter values ​​contained in the preprocessed operating parameters are extracted.

[0040] The parameter values ​​are converted into corresponding parameter vectors, and the parameter vectors are integrated into the same dataset to generate the target dataset, which is unique.

[0041] Furthermore, the determination module is specifically used for:

[0042] Based on preset rules, the parameter vectors in the target dataset are divided into corresponding training sets and validation sets, and the preset neural network is trained using the training sets to generate the corresponding initial fault prediction model.

[0043] The initial fault prediction model is validated using the validation set, and the validation result of the initial prediction model is judged in real time to determine whether it is qualified.

[0044] If the verification result of the initial prediction model is deemed satisfactory in real time, the training of the initial fault prediction model is completed, and the initial fault prediction model is set as the fault prediction model.

[0045] Furthermore, the judgment module is also specifically used for:

[0046] When the training set is obtained, the first parameter vector contained in the training set is extracted and input into the encoding layer of the preset CNN network so that the first parameter vector is encoded into the corresponding parameter matrix through the encoding layer.

[0047] The parameter matrix is ​​input into the parsing layer of the preset CNN network, and the feature values ​​corresponding to the parameter matrix are calculated through the parsing layer.

[0048] The feature values ​​are input into the learning layer of the preset CNN network, and the learning network in the learning layer is adjusted using the feature values ​​to generate the initial fault prediction model.

[0049] Furthermore, the judgment module is also specifically used for:

[0050] Several network nodes contained in the learning network are detected, and each network node is initialized to extract the original network parameters contained in each network node.

[0051] Each of the original network parameters is replaced one by one with the feature value to update the learning network in real time and generate the initial fault prediction model accordingly.

[0052] Furthermore, the electric drive system fault early warning system also includes a display module, which is specifically used for:

[0053] The fault level of the electric drive system is determined in real time based on the fault type, and each fault level corresponds to a fault coefficient.

[0054] The corresponding repair strategy is retrieved from a preset database based on the fault coefficient, and the repair strategy is displayed on the vehicle's dashboard. The fault coefficient is unique.

[0055] Furthermore, the electric drive system fault early warning system also includes a detection module, which is specifically used for:

[0056] When a fault is detected in the electric drive system, the geographical location of the vehicle is detected in real time, and a corresponding detection area is constructed with the geographical location as the center and a preset length as the radius.

[0057] Real-time determination of whether a maintenance station exists in the detection area;

[0058] If the presence of the maintenance station within the detection area is detected in real time, the communication link of the maintenance station is sent to the dashboard. There is at least one maintenance station.

[0059] The third aspect of the present invention proposes:

[0060] A computer includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the electric drive system fault warning method as described above.

[0061] The fourth aspect of the present invention proposes:

[0062] A readable storage medium having a computer program stored thereon, wherein the program, when executed by a processor, implements the electric drive system fault warning method as described above.

[0063] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0064] Figure 1 A flowchart of the electric drive system fault early warning method provided in the first embodiment of the present invention;

[0065] Figure 2 This is a structural block diagram of an electric drive system fault early warning system provided in the sixth embodiment of the present invention.

[0066] The following detailed description, in conjunction with the accompanying drawings, will further illustrate the present invention. Detailed Implementation

[0067] To facilitate understanding of the present invention, a more complete description will be given below with reference to the accompanying drawings. Several embodiments of the invention are illustrated in the drawings. However, the invention can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete.

[0068] It should be noted that when a component is said to be "fixed to" another component, it can be directly on the other component or there may be an intervening component. When a component is said to be "connected to" another component, it can be directly connected to the other component or there may be an intervening component. The terms "vertical," "horizontal," "left," "right," and similar expressions used in this document are for illustrative purposes only.

[0069] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0070] Please see Figure 1The image shows a fault early warning method for an electric drive system provided in the first embodiment of the present invention. The fault early warning method for an electric drive system provided in this embodiment can save a lot of computing resources and data acquisition equipment, reduce the cost of fault prediction, and improve the user experience.

[0071] Specifically, the electric drive system fault early warning method provided in this embodiment includes the following steps:

[0072] Step S10: When the electric drive system is detected to be running, the operating parameters of the electric drive system are collected in real time, and a corresponding target dataset is generated based on the operating parameters.

[0073] Step S20: Construct a corresponding fault prediction model based on the target dataset and a preset neural network, and input the operating parameters into the fault prediction model so that the fault prediction model can determine in real time whether there is a fault in the electric drive system;

[0074] Step S30: If the fault prediction model determines in real time that there is a fault in the electric drive system, the fault prediction model is immediately used to detect the fault type of the electric drive system, and a corresponding warning signal is generated according to the fault type.

[0075] Specifically, in this embodiment, it should first be noted that the electric drive system fault early warning method can be applied to different types of new energy electric vehicles. Based on this, in order to accurately warn the electric drive system, it is necessary to first collect the current operating parameters of the electric drive system during the actual operation of the electric drive system. Specifically, the operating parameters include a series of values, and further, the current operating parameters are integrated into the required target dataset.

[0076] Furthermore, after obtaining the required target dataset, it is input into a pre-configured CNN neural network in real time for adaptive training, simultaneously generating the necessary fault prediction model. Going further, the current operating parameters are input into the fault prediction model again, enabling real-time judgment. Specifically, these operating parameters may include voltage, current, and temperature. Based on this, if a fault is detected in the electric drive system in real time, the corresponding fault type is immediately identified, and a corresponding warning signal is issued based on that fault type. Specifically, the warning signal may be a light signal, a text signal, or an icon signal.

[0077] Second Embodiment

[0078] Specifically, in this embodiment, it should be noted that the step of generating the corresponding target dataset based on the running parameters includes:

[0079] When the operating parameters are obtained, the operating parameters are preprocessed, and several parameter values ​​contained in the preprocessed operating parameters are extracted.

[0080] The parameter values ​​are converted into corresponding parameter vectors, and the parameter vectors are integrated into the same dataset to generate the target dataset, which is unique.

[0081] Specifically, in this embodiment, it should be noted that after obtaining the required operating parameters through the above steps, in order to facilitate subsequent identification and processing, the current operating parameters need to be preprocessed by filtering and normalization in sequence. Furthermore, several parameter values ​​contained in the current preprocessed operating parameters are extracted.

[0082] Furthermore, by using a pre-set conversion script, the current parameter values ​​are converted into corresponding parameter vectors. Even further, the current parameter vectors are integrated into the same dataset, thus enabling the simple and quick generation of the target dataset for subsequent processing.

[0083] Specifically, in this embodiment, it should also be noted that the steps of constructing the corresponding fault prediction model based on the target dataset and the preset neural network include:

[0084] Based on preset rules, the parameter vectors in the target dataset are divided into corresponding training sets and validation sets, and the preset neural network is trained using the training sets to generate the corresponding initial fault prediction model.

[0085] The initial fault prediction model is validated using the validation set, and the validation result of the initial prediction model is judged in real time to determine whether it is qualified.

[0086] If the verification result of the initial prediction model is deemed satisfactory in real time, the training of the initial fault prediction model is completed, and the initial fault prediction model is set as the fault prediction model.

[0087] Specifically, in this embodiment, it should also be noted that after obtaining the required target dataset through the above steps, the parameter vectors in the current target dataset are divided into corresponding training sets and validation sets according to a pre-set ratio of 7:3. Based on this, the pre-set CNN network can be trained using the current training set, and the required initial fault prediction model can be trained. Furthermore, the current fault prediction model can be validated using the aforementioned validation set. On this basis, it is only necessary to judge in real time whether the output validation result is qualified, and the aforementioned fault prediction model can be trained accordingly for subsequent processing.

[0088] Third Embodiment

[0089] Furthermore, in this embodiment, it should be noted that the step of training the preset neural network using the training set to generate the corresponding initial fault prediction model includes:

[0090] When the training set is obtained, the first parameter vector contained in the training set is extracted and input into the encoding layer of the preset CNN network so that the first parameter vector is encoded into the corresponding parameter matrix through the encoding layer.

[0091] The parameter matrix is ​​input into the parsing layer of the preset CNN network, and the feature values ​​corresponding to the parameter matrix are calculated through the parsing layer.

[0092] The feature values ​​are input into the learning layer of the preset CNN network, and the learning network in the learning layer is adjusted using the feature values ​​to generate the initial fault prediction model.

[0093] In addition, it should be noted in this embodiment that the CNN network provided in this embodiment specifically includes an encoding layer, a parsing layer and a learning layer. Based on this, the first parameter vector contained in the training set is first input into the current encoding layer so that the Transformer encoder in the encoding layer encodes the current first parameter vector into the corresponding parameter matrix, and then inputs it into the parsing layer.

[0094] Furthermore, the parsing layer performs corresponding matrix operations on the parameter matrices and calculates the feature values ​​that fit each parameter matrix. Based on this, the current feature values ​​are input into the learning layer, and the parameters in the learning network are adjusted in real time using these feature values ​​to generate the required initial fault prediction model.

[0095] Furthermore, in this embodiment, it should be noted that the step of adjusting the learning network in the learning layer using the feature values ​​to generate the initial fault prediction model includes:

[0096] Several network nodes contained in the learning network are detected, and each network node is initialized to extract the original network parameters contained in each network node.

[0097] Each of the original network parameters is replaced one by one with the feature value to update the learning network in real time and generate the initial fault prediction model accordingly.

[0098] Furthermore, in this embodiment, it should be noted that after obtaining the required feature values ​​through the above steps, the learning network can be adjusted. Specifically, the network nodes contained in the current learning network are detected in real time, that is, the network structure of the current learning network is detected in real time. Further, the original network parameters contained in each network node, i.e., the factory parameters, are extracted. Based on this, each original network parameter is replaced one by one with the aforementioned feature values, thereby enabling real-time updates to the network structure of the current learning network and correspondingly training the aforementioned initial fault prediction model.

[0099] Fourth embodiment

[0100] In this embodiment, it should be noted that the method further includes:

[0101] The fault level of the electric drive system is determined in real time based on the fault type, and each fault level corresponds to a fault coefficient.

[0102] The corresponding repair strategy is retrieved from a preset database based on the fault coefficient, and the repair strategy is displayed on the vehicle's dashboard. The fault coefficient is unique.

[0103] In this embodiment, it should be noted that the fault type determined in real time can correspond to the fault level of the current electric drive system. Specifically, for example, if the fault type is determined to be "overvoltage" or "overcurrent," the corresponding fault level can be medium. Furthermore, the corresponding fault coefficient is 0.5, where the fault coefficient is between 0 and 1, and the larger the fault coefficient, the more severe the fault. The fault levels are specifically divided into minor, medium, and severe.

[0104] Furthermore, the system retrieves the required maintenance strategy from a preset database based on the fault coefficient and displays it on the dashboard so that the driver can perform the corresponding operation.

[0105] Fifth embodiment

[0106] In this embodiment, it should be noted that the method further includes:

[0107] When a fault is detected in the electric drive system, the geographical location of the vehicle is detected in real time, and a corresponding detection area is constructed with the geographical location as the center and a preset length as the radius.

[0108] Real-time determination of whether a maintenance station exists in the detection area;

[0109] If the presence of the maintenance station within the detection area is detected in real time, the communication link of the maintenance station is sent to the dashboard. There is at least one maintenance station.

[0110] In this embodiment, it should be noted that, in order to facilitate the driver in troubleshooting, the current vehicle will be further located, and a corresponding detection area will be constructed with the current vehicle as the center and a radius of 2 kilometers or 5 kilometers, etc.

[0111] Furthermore, it can determine in real time whether there is a repair station in the current detection area. Specifically, if so, it can immediately obtain the contact information of the current repair station and send it to the dashboard of the current vehicle to help the driver resolve the fault.

[0112] Please see Figure 2 The sixth embodiment of the present invention provides:

[0113] An electric drive system fault early warning system, wherein the system comprises:

[0114] The acquisition module is used to acquire the operating parameters of the electric drive system in real time when the electric drive system is detected to be running, and to generate a corresponding target dataset based on the operating parameters.

[0115] The judgment module is used to construct a corresponding fault prediction model based on the target dataset and a preset neural network, and input the operating parameters into the fault prediction model so that the fault prediction model can judge in real time whether there is a fault in the electric drive system;

[0116] The detection module is used to immediately detect the fault type of the electric drive system through the fault prediction model if a fault is detected in the electric drive system in real time, and generate a corresponding warning signal based on the fault type.

[0117] In the aforementioned electric drive system fault early warning system, the acquisition module is specifically used for:

[0118] When the operating parameters are obtained, the operating parameters are preprocessed, and several parameter values ​​contained in the preprocessed operating parameters are extracted.

[0119] The parameter values ​​are converted into corresponding parameter vectors, and the parameter vectors are integrated into the same dataset to generate the target dataset, which is unique.

[0120] In the aforementioned electric drive system fault early warning system, the judgment module is specifically used for:

[0121] Based on preset rules, the parameter vectors in the target dataset are divided into corresponding training sets and validation sets, and the preset neural network is trained using the training sets to generate the corresponding initial fault prediction model.

[0122] The initial fault prediction model is validated using the validation set, and the validation result of the initial prediction model is judged in real time to determine whether it is qualified.

[0123] If the verification result of the initial prediction model is deemed satisfactory in real time, the training of the initial fault prediction model is completed, and the initial fault prediction model is set as the fault prediction model.

[0124] In the aforementioned electric drive system fault early warning system, the judgment module is further specifically used for:

[0125] When the training set is obtained, the first parameter vector contained in the training set is extracted and input into the encoding layer of the preset CNN network so that the first parameter vector is encoded into the corresponding parameter matrix through the encoding layer.

[0126] The parameter matrix is ​​input into the parsing layer of the preset CNN network, and the feature values ​​corresponding to the parameter matrix are calculated through the parsing layer.

[0127] The feature values ​​are input into the learning layer of the preset CNN network, and the learning network in the learning layer is adjusted using the feature values ​​to generate the initial fault prediction model.

[0128] In the aforementioned electric drive system fault early warning system, the judgment module is further specifically used for:

[0129] Several network nodes contained in the learning network are detected, and each network node is initialized to extract the original network parameters contained in each network node.

[0130] Each of the original network parameters is replaced one by one with the feature value to update the learning network in real time and generate the initial fault prediction model accordingly.

[0131] In the aforementioned electric drive system fault early warning system, the electric drive system fault early warning system further includes a display module, which is specifically used for:

[0132] The fault level of the electric drive system is determined in real time based on the fault type, and each fault level corresponds to a fault coefficient.

[0133] The corresponding repair strategy is retrieved from a preset database based on the fault coefficient, and the repair strategy is displayed on the vehicle's dashboard. The fault coefficient is unique.

[0134] In the aforementioned electric drive system fault early warning system, the electric drive system fault early warning system further includes a detection module, which is specifically used for:

[0135] When a fault is detected in the electric drive system, the geographical location of the vehicle is detected in real time, and a corresponding detection area is constructed with the geographical location as the center and a preset length as the radius.

[0136] Real-time determination of whether a maintenance station exists in the detection area;

[0137] If the presence of the maintenance station within the detection area is detected in real time, the communication link of the maintenance station is sent to the dashboard. There is at least one maintenance station.

[0138] The seventh embodiment of the present invention provides a computer, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the electric drive system fault early warning method provided in the above embodiments.

[0139] The eighth embodiment of the present invention provides a readable storage medium having a computer program stored thereon, wherein the program, when executed by a processor, implements the electric drive system fault early warning method provided in the above embodiments.

[0140] In summary, the electric drive system fault early warning method and system provided by the above embodiments of the present invention can save a lot of computing resources and data acquisition equipment, reduce the cost of fault prediction, and improve the user experience.

[0141] It should be noted that the above modules can be functional modules or program modules, and can be implemented through software or hardware. For modules implemented through hardware, the above modules can reside in the same processor; or the above modules can be located in different processors in any combination.

[0142] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-including system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0143] More specific examples of computer-readable media (a non-exhaustive list) include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0144] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0145] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0146] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

Claims

1. A method for early warning of faults in an electric drive system, characterized in that, The method includes: When the electric drive system is detected to be running, the operating parameters of the electric drive system are collected in real time, and a corresponding target dataset is generated based on the operating parameters. A corresponding fault prediction model is constructed based on the target dataset and a preset neural network, and the operating parameters are input into the fault prediction model so that the fault prediction model can determine in real time whether there is a fault in the electric drive system. If the fault prediction model determines in real time that there is a fault in the electric drive system, the fault prediction model will immediately detect the fault type of the electric drive system and generate a corresponding warning signal based on the fault type. The step of generating the corresponding target dataset based on the running parameters includes: When the operating parameters are obtained, the operating parameters are preprocessed, and several parameter values ​​contained in the preprocessed operating parameters are extracted. The parameter values ​​are converted into corresponding parameter vectors, and the parameter vectors are integrated into the same dataset to generate the target dataset, which is unique. The step of constructing a corresponding fault prediction model based on the target dataset and a preset neural network includes: Based on preset rules, the parameter vectors in the target dataset are divided into corresponding training sets and validation sets, and the preset neural network is trained using the training sets to generate the corresponding initial fault prediction model. The initial fault prediction model is validated using the validation set, and the validation result of the initial fault prediction model is judged in real time to determine whether it is qualified. If the verification result of the initial prediction model is deemed satisfactory in real time, the training of the initial fault prediction model is completed, and the initial fault prediction model is set as the fault prediction model. The step of training the preset neural network using the training set to generate the corresponding initial fault prediction model includes: When the training set is obtained, the first parameter vector contained in the training set is extracted, and the first parameter vector is input into the encoding layer in the preset CNN network so that the first parameter vector is encoded into the corresponding parameter matrix through the encoding layer. The parameter matrix is ​​input into the parsing layer of the preset CNN network, and the feature values ​​corresponding to the parameter matrix are calculated through the parsing layer. The feature values ​​are input into the learning layer of the preset CNN network, and the learning network in the learning layer is adjusted by the feature values ​​to generate the initial fault prediction model accordingly. The step of adjusting the learning network in the learning layer using the feature values ​​to generate the initial fault prediction model includes: Several network nodes contained in the learning network are detected, and each network node is initialized to extract the original network parameters contained in each network node. Each of the original network parameters is replaced one by one with the feature value to update the learning network in real time and generate the initial fault prediction model accordingly.

2. The fault early warning method for an electric drive system according to claim 1, characterized in that: The method further includes: The fault level of the electric drive system is determined in real time based on the fault type, and each fault level corresponds to a fault coefficient. The corresponding repair strategy is retrieved from a preset database based on the fault coefficient, and the repair strategy is displayed on the vehicle's dashboard. The fault coefficient is unique.

3. The electric drive system fault early warning method according to claim 2, characterized in that: The method further includes: When a fault is detected in the electric drive system, the geographical location of the vehicle is detected in real time, and a corresponding detection area is constructed with the geographical location as the center and a preset length as the radius. Real-time determination of whether a maintenance station exists in the detection area; If the presence of the maintenance station within the detection area is detected in real time, the communication link of the maintenance station is sent to the dashboard. There is at least one maintenance station.

4. A fault early warning system for an electric drive system, characterized in that, For implementing the electric drive system fault early warning method as described in any one of claims 1 to 3, the system comprises: The acquisition module is used to acquire the operating parameters of the electric drive system in real time when the electric drive system is detected to be running, and to generate a corresponding target dataset based on the operating parameters. The judgment module is used to construct a corresponding fault prediction model based on the target dataset and a preset neural network, and input the operating parameters into the fault prediction model so that the fault prediction model can judge in real time whether there is a fault in the electric drive system; The detection module is used to immediately detect the fault type of the electric drive system through the fault prediction model if a fault is detected in the electric drive system in real time, and generate a corresponding warning signal based on the fault type.

5. A computer, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the electric drive system fault early warning method as described in any one of claims 1 to 3.

6. A readable storage medium having a computer program stored thereon, characterized in that, When executed by the processor, the program implements the electric drive system fault early warning method as described in any one of claims 1 to 3.

Images