CN116304806A - Motor fault diagnosis method, diagnosis model training method and system - Google Patents
Motor fault diagnosis method, diagnosis model training method and system Download PDFInfo
- Publication number
- CN116304806A CN116304806A CN202310151196.3A CN202310151196A CN116304806A CN 116304806 A CN116304806 A CN 116304806A CN 202310151196 A CN202310151196 A CN 202310151196A CN 116304806 A CN116304806 A CN 116304806A
- Authority
- CN
- China
- Prior art keywords
- motor
- fault
- related parameter
- sample
- diagnosis
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000003745 diagnosis Methods 0.000 title claims abstract description 104
- 238000000034 method Methods 0.000 title claims abstract description 49
- 238000012549 training Methods 0.000 title claims abstract description 40
- 238000012545 processing Methods 0.000 claims description 27
- 238000004590 computer program Methods 0.000 claims description 14
- 238000001514 detection method Methods 0.000 claims description 10
- 238000011084 recovery Methods 0.000 claims description 5
- 101001095088 Homo sapiens Melanoma antigen preferentially expressed in tumors Proteins 0.000 claims description 3
- 102100037020 Melanoma antigen preferentially expressed in tumors Human genes 0.000 claims description 3
- 230000006870 function Effects 0.000 claims description 3
- 238000007781 pre-processing Methods 0.000 description 6
- 238000013528 artificial neural network Methods 0.000 description 5
- 238000004422 calculation algorithm Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 238000013507 mapping Methods 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 230000015556 catabolic process Effects 0.000 description 2
- 238000006731 degradation reaction Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000006866 deterioration Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/34—Testing dynamo-electric machines
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M13/00—Testing of machine parts
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Control Of Electric Motors In General (AREA)
Abstract
The application relates to a motor fault diagnosis method, a diagnosis model training method and a system. The method comprises the following steps: collecting motor-related parameter samples and corresponding fault levels, wherein each motor-related parameter sample comprises n types of motor-related parameters, and each fault level comprises s types; configuring an initial model, wherein the initial model comprises an input layer, an implicit layer and an output layer, wherein the input layer comprises n nodes, and the output layer comprises s nodes; and inputting a plurality of motor-related parameter samples into the initial model to obtain corresponding sample output results, and iteratively training and updating model parameters of the initial model according to fitting precision of the sample output results and the fault grade to obtain a diagnosis model so as to identify the motor-related parameter samples to be processed according to the diagnosis model to obtain the fault grade. Not only can the faults and the fault grades be predicted, but also the precision is improved.
Description
Technical Field
The application relates to the technical field of new energy automobiles, in particular to a motor fault diagnosis method, a diagnosis model training method and a system.
Background
With the development of new energy automobile technology, new requirements are also put forward on the running stability and reliability of the motor, at present, a motor fault diagnosis strategy usually executes corresponding treatment measures after the motor has failed, and at the moment, the fault state and fault treatment have been perceived by a user and have influence on driving of the user, and the motor fault is often difficult to predict or preprocess in advance.
Disclosure of Invention
Based on the method, the motor fault diagnosis method, the diagnosis model training method and the system are provided, and the problem that motor faults are difficult to predict in the prior art is solved.
In one aspect, a motor fault diagnosis model training method is provided, the method comprising:
collecting motor-related parameter samples and corresponding fault levels, wherein each motor-related parameter sample comprises n types of motor-related parameters, and each fault level comprises s types;
configuring an initial model, wherein the initial model comprises an input layer, an implicit layer and an output layer, wherein the input layer comprises n nodes, and the output layer comprises s nodes;
and inputting a plurality of motor-related parameter samples into the initial model to obtain corresponding sample output results, and iteratively training and updating model parameters of the initial model according to fitting precision of the sample output results and the fault grade to obtain a diagnosis model so as to identify the motor-related parameter samples to be processed according to the diagnosis model to obtain the fault grade.
In one embodiment, the fitting accuracy comprises a first fitting accuracy, the mathematical expression of which is:
R 2 for the first fitting accuracy, yi is the fault level of the ith motor-related parameter sample,outputting a result for the sample of the ith motor-related parameter sample, N being the number of motor-related parameter samples, +.>And outputting a mean value of results for samples of the N motor-related parameter samples.
In one embodiment, the fitting accuracy comprises a second fitting accuracy, the mathematical expression of which is:
RMSE is the second fitting accuracy, yi is the failure level of the ith motor-related parameter sample,sample input for the ith motor-related parameter sampleAs a result, N is the number of motor-related parameter samples, +.>And outputting a mean value of results for samples of the N motor-related parameter samples.
In one embodiment, the fitting accuracy includes a third fitting accuracy, the mathematical expression of which is:
MAPE is the third fitting precision, yi is the fault level of the ith motor-related parameter sample,outputting a result for the sample of the ith motor-related parameter sample, N being the number of motor-related parameter samples, +.>And (5) taking the average value of sample output results of the N motor-related parameter samples, wherein max is a maximum value taking function.
On the other hand, a motor fault diagnosis method is provided, and the diagnosis model is applied and comprises the following steps:
collecting a motor associated parameter sample to be processed, inputting the motor associated parameter sample to the diagnosis model for classification processing, and obtaining a classified fault grade, wherein the number of motor associated parameter types in the motor associated parameter sample to be processed is consistent with the number of nodes of an input layer in the diagnosis model, and the classified fault grade is consistent with the number of nodes of an output layer in the diagnosis model.
In one embodiment, collecting a motor-related parameter sample to be processed, inputting the motor-related parameter sample to the diagnostic model for classification processing, and obtaining a classified fault level, wherein the steps include:
derating and/or power limiting the diagnosed motor.
In one embodiment, derating and/or power limiting the diagnosed motor, then includes:
the whole vehicle controller or the power domain controller acquires the current fault level of the motor and executes a corresponding diagnosis strategy according to the current fault level, wherein the diagnosis strategy at least comprises one of the following steps: information push, torque limit, vehicle speed limit, and exit energy recovery.
In one embodiment, a corresponding diagnosis strategy is executed according to the classified fault level, a motor associated parameter sample to be processed is collected, and the diagnosis model is input for classification processing, so as to obtain the classified fault level, and then the method comprises the following steps:
when the diagnosed new energy automobile is four-wheel drive, the classified fault levels of the first motor and the second motor are respectively obtained to execute corresponding derating processing and/or power limitation.
In yet another aspect, a motor fault diagnosis model training system is provided, the system comprising:
the system comprises an acquisition module, a fault detection module and a fault detection module, wherein the acquisition module is used for acquiring motor-related parameter samples and corresponding fault levels, each motor-related parameter sample comprises n types of motor-related parameters, and the fault levels comprise s types;
the configuration module is used for configuring an initial model, wherein the initial model comprises an input layer, an implicit layer and an output layer, the input layer comprises n nodes, and the output layer comprises s nodes;
the training module is used for inputting a plurality of motor-related parameter samples into the initial model to obtain corresponding sample output results, and iteratively training and updating model parameters of the initial model according to fitting accuracy of the sample output results and the fault grade to obtain a diagnosis model so as to identify the motor-related parameter samples to be processed according to the diagnosis model to obtain the fault grade.
There is provided a computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the steps of the method when executing the computer program.
There is provided a computer readable storage medium having stored thereon a computer program which when executed by a processor performs the steps of the method.
According to the motor fault diagnosis method, the motor fault diagnosis model training method and the motor fault diagnosis model training system, the mapping relation between the motor association parameter samples and the fault levels is obtained based on the neural network algorithm and the supervised learning, so that the motor faults and the fault levels are determined, the time sequence influence of motor operation faults is avoided, the motor faults and the fault levels can be determined by collecting the motor association parameter samples before the motor faults occur, motor fault processing can be further performed before the motor faults occur, driving experience of users and riding experience of passengers are improved, and the multidimensional motor association parameters can be used as judgment basis, single or few motor association parameters are prevented from triggering wrong numerical judgment indexes and condition judgment indexes, and the diagnosis precision of the motor faults and the fault levels is improved.
Drawings
FIG. 1 is an application environment diagram of a motor fault diagnosis model training method in one embodiment;
FIG. 2 is a flow chart of a motor fault diagnosis method in one embodiment;
FIG. 3 is a schematic diagram of a motor fault diagnosis system in one embodiment;
fig. 4 is an internal structural diagram of a computer device in one embodiment.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application will be further described in detail with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the present application.
The motor fault diagnosis model training method provided by the application can be applied to an application environment shown in figure 1. Wherein the terminal 102 communicates with the server 104 via a network. The server 104 can be utilized to provide calculation force to perform iterative training on the initial model to update model parameters so as to obtain a diagnosis model, the iterative training process can be performed by depending on supervised learning, the fitting precision is used as the basis for updating model parameters to obtain the diagnosis model, and the diagnosis model can be deployed into the terminal 102 so as to identify a motor associated parameter sample to be processed according to the diagnosis model to obtain a fault grade. The terminal 102 may be, but not limited to, a domain controller or a whole vehicle controller of a new energy automobile, or may be a detection device terminal. The server 104 may be implemented as a stand-alone server or as a server cluster of multiple servers.
With the development of related technologies of new energy automobiles, the popularity of new energy automobiles is improved, the new energy automobiles generally use electric energy and a motor as driving force and driving components, so that the state detection of the motor, especially the detection of stability and faults is particularly important, but at present, the state detection of the motor is generally whether the motor is in fault or not and the fault type is determined through the change of a motor enabling signal after the fault occurs. Based on the method, the invention provides a training method and a diagnosis method based on a fault diagnosis model based on a neural network algorithm and supervised learning.
As shown in fig. 1, the present invention provides a motor fault diagnosis model training method, which includes:
s1: collecting motor-related parameter samples and corresponding fault levels, wherein each motor-related parameter sample comprises n types of motor-related parameters, and each fault level comprises s types;
s2: configuring an initial model, wherein the initial model comprises an input layer, an implicit layer and an output layer, wherein the input layer comprises n nodes, and the output layer comprises s nodes;
s3: and inputting a plurality of motor-related parameter samples into the initial model to obtain corresponding sample output results, and iteratively training and updating model parameters of the initial model according to fitting precision of the sample output results and the fault grade to obtain a diagnosis model so as to identify the motor-related parameter samples to be processed according to the diagnosis model to obtain the fault grade.
In step S1, a motor-related parameter sample may be collected, and some motor-related parameters with a larger weight on the motor fault may be selected according to specific scene requirements and different new energy automobiles, for example, the motor-related parameters include at least one or more of the following: for example, n types of motor related parameters can be selected as one motor related parameter sample, and a fault level corresponding to the motor related parameter sample is determined, and the number s of the fault levels can also be determined according to the type and the fault degree of the motor, for example, the number s can be 3, 4, 5 or other.
Illustratively, in step S2, the neural network may be configured as an initial model including an input layer, an implicit layer, and an output layer, wherein the number of nodes of the input layer coincides with the number of types n of motor-related parameters in the motor-related parameter sample, and the number of nodes of the output layer coincides with the number of types S of failure levels.
In order to train the initial model better, in step S3, the motor-related parameter sample may be input into the initial model as an n-dimensional vector, a corresponding sample output result is obtained, the sample output result and the fault level are compared in a supervised learning manner, the fitting precision of the sample output result and the fault level is used as the basis of the initial model optimization, the model parameters of the initial model are updated through iterative training, and when the accuracy or recall rate of the diagnostic model can reach a preset threshold, the diagnostic model is obtained, and the diagnostic model may be used to process the motor-related parameter sample to be processed and obtain the corresponding fault level.
The invention obtains the mapping relation between the motor association parameter sample and the fault level based on the neural network algorithm and the supervised learning, thereby determining the motor fault and the fault level, being not influenced by the time sequence of the motor operation fault, being capable of determining the motor fault and the fault level by collecting the motor association parameter sample before the motor fault occurs, further being capable of carrying out motor fault processing before the motor fault occurs, improving the driving experience of a user and the riding experience of a passenger, being capable of taking the multi-dimensional motor association parameter as a judging basis, avoiding the single or a few motor association parameters from triggering wrong numerical judgment indexes and condition judgment indexes, and improving the diagnosis precision of the motor fault and the fault level.
Optionally, the fitting accuracy includes a first fitting accuracy, and the mathematical expression of the first fitting accuracy is:
R 2 for the first fitting accuracy, yi is the fault level of the ith motor-related parameter sample,outputting a result for the sample of the ith motor-related parameter sample, N being the number of motor-related parameter samples, +.>And outputting a mean value of results for samples of the N motor-related parameter samples.
Optionally, the fitting accuracy includes a second fitting accuracy, and a mathematical expression of the second fitting accuracy is:
RMSE is the second fitting accuracy, yi is the failure level of the ith motor-related parameter sample,outputting a result for the sample of the ith motor-related parameter sample, N being the number of motor-related parameter samples, +.>And outputting a mean value of results for samples of the N motor-related parameter samples.
Optionally, the fitting accuracy includes a third fitting accuracy, and a mathematical expression of the third fitting accuracy is:
MAPE is the third fitting precision, yi is the fault level of the ith motor-related parameter sample,outputting a result for the sample of the ith motor-related parameter sample, N being the number of motor-related parameter samples, +.>And (5) taking the average value of sample output results of the N motor-related parameter samples, wherein max is a maximum value taking function.
Different fitting precision is selected in the training process of the initial model according to the performance and the application scene of the motor, or the first fitting precision, the second fitting precision and the third fitting precision are used in a combined mode, so that the performance of the diagnosis model is improved. For example, a value of the first fitting accuracy may be required to be equal to or greater than 0.95, a value of the second fitting accuracy may be required to be less than 0.15, and a value of the third fitting accuracy may be required to be less than 0.25.
Predicting possible motor faults according to real-time data of the new energy automobile, derating the motor in advance after a prediction result is obtained, calculating the maximum available torque and the maximum available rotating speed of the motor according to capacity parameters of the motor, and carrying out small-amplitude power limitation by combining the prediction fault result, so that the occurrence of subsequent faults is avoided to the greatest extent under the condition that a user does not perceive the motor faults.
The invention provides a motor fault diagnosis method, which comprises the following steps of deploying or putting on line a trained diagnosis model, storing the diagnosis model into a domain controller or a whole vehicle controller for processing real-time motor association parameter samples, and applying the diagnosis model, wherein the method comprises the following steps:
collecting a motor associated parameter sample to be processed, inputting the motor associated parameter sample to the diagnosis model for classification processing, and obtaining a classified fault grade, wherein the number of motor associated parameter types in the motor associated parameter sample to be processed is consistent with the number of nodes of an input layer in the diagnosis model, and the classified fault grade is consistent with the number of nodes of an output layer in the diagnosis model.
And calling a corresponding diagnosis model according to motor-related parameters of different types and the number of types in the motor-related parameter sample to be processed, wherein the number of motor-related parameter types in the motor-related parameter sample to be processed is consistent with the number of nodes of an input layer in the diagnosis model, and the classified fault level is consistent with the number of nodes of an output layer in the diagnosis model.
The invention obtains the mapping relation between the motor association parameter sample and the fault level based on the neural network algorithm and the supervised learning, thereby determining the motor fault and the fault level, being not influenced by the time sequence of the motor operation fault, being capable of determining the motor fault and the fault level by collecting the motor association parameter sample before the motor fault occurs, further being capable of carrying out motor fault processing before the motor fault occurs, improving the driving experience of a user and the riding experience of a passenger, being capable of taking the multi-dimensional motor association parameter as a judging basis, avoiding the single or a few motor association parameters from triggering wrong numerical judgment indexes and condition judgment indexes, and improving the diagnosis precision of the motor fault and the fault level.
In one embodiment, collecting a motor-related parameter sample to be processed, inputting the motor-related parameter sample to the diagnostic model for classification processing, and obtaining a classified fault level, wherein the steps include:
derating and/or power limiting the diagnosed motor. Through carrying out derating processing and/or power limitation to the motor, the fault trend of the motor is relieved, meanwhile, the further deterioration of the motor fault is avoided, under the general condition, when the fault level is lower or the motor fault of the lower level is about to occur, the current motor running state and relevant running parameters can be changed through the derating processing and/or the power limitation, and under the condition that the driving of a user is noninductive and the passenger is noninductive, the processing of the motor fault is completed, and the driving experience and the riding experience are improved.
In some implementations, derating and/or power limiting the diagnosed motor, then includes:
the whole vehicle controller or the power domain controller acquires the current fault level of the motor and executes a corresponding diagnosis strategy according to the current fault level, wherein the diagnosis strategy at least comprises one of the following steps: information push, torque limit, vehicle speed limit, and exit energy recovery.
For example, the following measures can be taken:
1) For motor warning faults, log recording is carried out in a preprocessing stage without processing, when the faults cannot be recovered and reported, the faults are processed into recorded fault codes, power degradation processing is not carried out, and lighting and text prompting are not carried out on users;
2) For small-amplitude power-limiting faults of the motor, log recording is carried out in a preprocessing stage, the faults are not processed, when the faults cannot be recovered, the faults are reported, the faults are processed into recorded fault codes, power degradation processing is not carried out, and lighting and text prompting are not carried out on users;
3) The method comprises the steps of greatly limiting power faults of a motor, calculating the maximum available rotating speed and torque of the motor according to the motor capacity in a preprocessing stage, reducing the maximum available rotating speed by a small margin, recording fault codes when the faults cannot be recovered and reported, and turning on a power limiting lamp to prompt;
4) In the preprocessing stage, calculating the maximum available rotating speed and torque of the motor according to the motor capacity, reducing the maximum available rotating speed by a small margin, recording a fault code and lighting a motor fault lamp to prompt when the fault cannot be recovered and reported out;
5) For the fault of the level 1 forbidden motor, in a preprocessing stage, calculating the maximum available rotating speed and torque of the motor according to the power capacity, reducing the maximum available rotating speed and torque, recording a fault code when the fault cannot be recovered and the fault is reported, exiting the motor energy recovery, executing gradient zero torque processing, and simultaneously lighting a motor fault lamp prompt;
6) And for the fault of the motor level 2, calculating the maximum available rotating speed and torque of the motor according to the power capacity in a preprocessing stage, greatly reducing the maximum available rotating speed and torque, recording a fault code when the fault cannot be recovered, exiting the motor energy recovery, executing gradient zero torque processing, simultaneously lighting a motor fault lamp prompt, and executing N-gear cutting processing after the gradient zero torque is finished.
Under the actual working condition and the vehicle condition, the new energy automobile with double driving generally has only one front motor or only one rear motor, when fault diagnosis is carried out, only the influence of motor related parameters of one motor on motor faults and fault grades is considered, the new energy automobile with four driving generally has two motors, namely the front motor and the rear motor, when fault diagnosis is carried out, the influence of motor related parameters of the two motors on motor faults and fault grades is considered, in one embodiment, corresponding diagnosis strategies are executed according to the classified fault grades, motor related parameter samples to be processed are collected, the diagnosis model is input for classification processing, and classified fault grades are obtained, and then:
when the diagnosed new energy automobile is four-wheel drive, the classified fault levels of the first motor and the second motor are respectively obtained to execute corresponding derating processing and/or power limitation.
It should be understood that, although the steps in the flowchart of fig. 1 are shown in sequence as indicated by the arrows, the steps are not necessarily performed in sequence as indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least some of the steps in fig. 1 may include multiple sub-steps or stages that are not necessarily performed at the same time, but may be performed at different times, nor do the order in which the sub-steps or stages are performed necessarily performed in sequence, but may be performed alternately or alternately with at least a portion of other steps or sub-steps of other steps.
As shown in fig. 3, there is provided a motor fault diagnosis model training system, the system comprising:
the system comprises an acquisition module, a fault detection module and a fault detection module, wherein the acquisition module is used for acquiring motor-related parameter samples and corresponding fault levels, each motor-related parameter sample comprises n types of motor-related parameters, and the fault levels comprise s types;
the configuration module is used for configuring an initial model, wherein the initial model comprises an input layer, an implicit layer and an output layer, the input layer comprises n nodes, and the output layer comprises s nodes;
the training module is used for inputting a plurality of motor-related parameter samples into the initial model to obtain corresponding sample output results, and iteratively training and updating model parameters of the initial model according to fitting accuracy of the sample output results and the fault grade to obtain a diagnosis model so as to identify the motor-related parameter samples to be processed according to the diagnosis model to obtain the fault grade.
For specific limitations on the motor fault diagnosis model training system, reference may be made to the above limitation on the motor fault diagnosis model training method, and no further description is given here. The above-described individual modules in the motor fault diagnosis model training system may be implemented in whole or in part by software, hardware, and combinations thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.
In one embodiment, a computer device is provided, which may be a server, the internal structure of which may be as shown in fig. 4. The computer device includes a processor, a memory, a network interface, and a database connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The database of the computer device is used for storing motor fault diagnosis model training method data. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program, when executed by a processor, implements a motor fault diagnosis model training method.
Those skilled in the art will appreciate that the structures shown in FIG. 4 are block diagrams only and do not constitute a limitation of the computer device on which the present aspects apply, and that a particular computer device may include more or less components than those shown, or may combine some of the components, or have a different arrangement of components.
In one embodiment, a computer device is provided comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, the processor implementing the steps of when executing the computer program:
collecting motor-related parameter samples and corresponding fault levels, wherein each motor-related parameter sample comprises n types of motor-related parameters, and each fault level comprises s types;
configuring an initial model, wherein the initial model comprises an input layer, an implicit layer and an output layer, wherein the input layer comprises n nodes, and the output layer comprises s nodes;
and inputting a plurality of motor-related parameter samples into the initial model to obtain corresponding sample output results, and iteratively training and updating model parameters of the initial model according to fitting precision of the sample output results and the fault grade to obtain a diagnosis model so as to identify the motor-related parameter samples to be processed according to the diagnosis model to obtain the fault grade.
Alternatively, execution:
the diagnostic model comprises:
collecting a motor associated parameter sample to be processed, inputting the motor associated parameter sample to the diagnosis model for classification processing, and obtaining a classified fault grade, wherein the number of motor associated parameter types in the motor associated parameter sample to be processed is consistent with the number of nodes of an input layer in the diagnosis model, and the classified fault grade is consistent with the number of nodes of an output layer in the diagnosis model.
In one embodiment, a computer readable storage medium is provided having a computer program stored thereon, which when executed by a processor, performs the steps of:
collecting motor-related parameter samples and corresponding fault levels, wherein each motor-related parameter sample comprises n types of motor-related parameters, and each fault level comprises s types;
configuring an initial model, wherein the initial model comprises an input layer, an implicit layer and an output layer, wherein the input layer comprises n nodes, and the output layer comprises s nodes;
and inputting a plurality of motor-related parameter samples into the initial model to obtain corresponding sample output results, and iteratively training and updating model parameters of the initial model according to fitting precision of the sample output results and the fault grade to obtain a diagnosis model so as to identify the motor-related parameter samples to be processed according to the diagnosis model to obtain the fault grade.
Alternatively, execution:
the diagnostic model comprises:
collecting a motor associated parameter sample to be processed, inputting the motor associated parameter sample to the diagnosis model for classification processing, and obtaining a classified fault grade, wherein the number of motor associated parameter types in the motor associated parameter sample to be processed is consistent with the number of nodes of an input layer in the diagnosis model, and the classified fault grade is consistent with the number of nodes of an output layer in the diagnosis model.
Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in the various embodiments provided herein may include non-volatile and/or volatile memory. The nonvolatile memory can include Read Only Memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double Data Rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchi nk) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), among others.
The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.
The above examples merely represent a few embodiments of the present application, which are described in more detail and are not to be construed as limiting the scope of the invention. It should be noted that it would be apparent to those skilled in the art that various modifications and improvements could be made without departing from the spirit of the present application, which would be within the scope of the present application. Accordingly, the scope of protection of the present application is to be determined by the claims appended hereto.
Claims (11)
1. A motor fault diagnosis model training method, comprising:
collecting motor-related parameter samples and corresponding fault levels, wherein each motor-related parameter sample comprises n types of motor-related parameters, and each fault level comprises s types;
configuring an initial model, wherein the initial model comprises an input layer, an implicit layer and an output layer, wherein the input layer comprises n nodes, and the output layer comprises s nodes;
and inputting a plurality of motor-related parameter samples into the initial model to obtain corresponding sample output results, and iteratively training and updating model parameters of the initial model according to fitting precision of the sample output results and the fault grade to obtain a diagnosis model so as to identify the motor-related parameter samples to be processed according to the diagnosis model to obtain the fault grade.
2. The motor fault diagnosis model training method according to claim 1, wherein the fitting accuracy includes a first fitting accuracy, and the mathematical expression of the first fitting accuracy is:
R 2 for the first fitting accuracy, yi is the fault level of the ith motor-related parameter sample,outputting a result for the sample of the ith motor-related parameter sample, N being the number of motor-related parameter samples, +.>And outputting a mean value of results for samples of the N motor-related parameter samples.
3. The motor fault diagnosis model training method according to claim 1, wherein the fitting accuracy includes a second fitting accuracy, and the mathematical expression of the second fitting accuracy is:
RMSE is the second fitting accuracy, yi is the failure level of the ith motor-related parameter sample,outputting a result for the sample of the ith motor-related parameter sample, N being the number of motor-related parameter samples, +.>And outputting a mean value of results for samples of the N motor-related parameter samples.
4. The motor fault diagnosis model training method according to claim 1, wherein the fitting accuracy includes a third fitting accuracy, and the mathematical expression of the third fitting accuracy is:
MAPE is the third fitting precision, yi is the result of the ith motor-related parameter sampleThe level of the obstacle is determined,outputting a result for the sample of the ith motor-related parameter sample, N being the number of motor-related parameter samples, +.>And (5) taking the average value of sample output results of the N motor-related parameter samples, wherein max is a maximum value taking function.
5. A motor fault diagnosis method, characterized by applying the diagnosis model according to any one of claims 1 to 4, comprising:
collecting a motor associated parameter sample to be processed, inputting the motor associated parameter sample to the diagnosis model for classification processing, and obtaining a classified fault grade, wherein the number of motor associated parameter types in the motor associated parameter sample to be processed is consistent with the number of nodes of an input layer in the diagnosis model, and the classified fault grade is consistent with the number of nodes of an output layer in the diagnosis model.
6. The motor fault diagnosis method according to claim 5, wherein collecting a motor-related parameter sample to be processed, inputting the motor-related parameter sample to the diagnosis model for classification processing, obtaining a classified fault level, and then comprising:
derating and/or power limiting the diagnosed motor.
7. The motor fault diagnosis method according to claim 6, characterized in that derating and/or power limiting of the motor being diagnosed is performed, after which it comprises:
the whole vehicle controller or the power domain controller acquires the current fault level of the motor and executes a corresponding diagnosis strategy according to the current fault level, wherein the diagnosis strategy at least comprises one of the following steps: information push, torque limit, vehicle speed limit, and exit energy recovery.
8. The motor fault diagnosis method according to claim 5, wherein the corresponding diagnosis strategy is executed according to the classified fault level, the motor-related parameter sample to be processed is collected, and the diagnosis model is input for classification processing, so as to obtain the classified fault level, and then the method comprises the steps of:
when the diagnosed new energy automobile is four-wheel drive, the classified fault levels of the first motor and the second motor are respectively obtained to execute corresponding derating processing and/or power limitation.
9. A motor fault diagnosis model training system, the system comprising:
the system comprises an acquisition module, a fault detection module and a fault detection module, wherein the acquisition module is used for acquiring motor-related parameter samples and corresponding fault levels, each motor-related parameter sample comprises n types of motor-related parameters, and the fault levels comprise s types;
the configuration module is used for configuring an initial model, wherein the initial model comprises an input layer, an implicit layer and an output layer, the input layer comprises n nodes, and the output layer comprises s nodes;
the training module is used for inputting a plurality of motor-related parameter samples into the initial model to obtain corresponding sample output results, and iteratively training and updating model parameters of the initial model according to fitting accuracy of the sample output results and the fault grade to obtain a diagnosis model so as to identify the motor-related parameter samples to be processed according to the diagnosis model to obtain the fault grade.
10. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the steps of the method according to any one of claims 1 to 8 when the computer program is executed by the processor.
11. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 8.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310151196.3A CN116304806A (en) | 2023-02-22 | 2023-02-22 | Motor fault diagnosis method, diagnosis model training method and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310151196.3A CN116304806A (en) | 2023-02-22 | 2023-02-22 | Motor fault diagnosis method, diagnosis model training method and system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN116304806A true CN116304806A (en) | 2023-06-23 |
Family
ID=86819741
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202310151196.3A Pending CN116304806A (en) | 2023-02-22 | 2023-02-22 | Motor fault diagnosis method, diagnosis model training method and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN116304806A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117930028A (en) * | 2024-03-21 | 2024-04-26 | 成都赛力斯科技有限公司 | Method, system, equipment and medium for predicting thermal failure of new energy vehicle battery |
CN118035882A (en) * | 2024-01-11 | 2024-05-14 | 太极计算机股份有限公司 | Electrical fault diagnosis method and system based on AI |
-
2023
- 2023-02-22 CN CN202310151196.3A patent/CN116304806A/en active Pending
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN118035882A (en) * | 2024-01-11 | 2024-05-14 | 太极计算机股份有限公司 | Electrical fault diagnosis method and system based on AI |
CN117930028A (en) * | 2024-03-21 | 2024-04-26 | 成都赛力斯科技有限公司 | Method, system, equipment and medium for predicting thermal failure of new energy vehicle battery |
CN117930028B (en) * | 2024-03-21 | 2024-05-17 | 成都赛力斯科技有限公司 | Method, system, equipment and medium for predicting thermal failure of new energy vehicle battery |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN116304806A (en) | Motor fault diagnosis method, diagnosis model training method and system | |
CN109163913A (en) | A kind of Diagnosis method of automobile faults and relevant device | |
CN111931639B (en) | Driver behavior detection method and device, electronic equipment and storage medium | |
CN116108717B (en) | Traffic transportation equipment operation prediction method and device based on digital twin | |
CN111695692A (en) | Apparatus and method for predicting damage degree | |
US20190339330A1 (en) | Method of estimating a charge state for a battery cell | |
CN115980585A (en) | Battery fault detection method and device, computer equipment and storage medium | |
CN113743528A (en) | Voltage source inverter fault diagnosis method and system based on convolutional neural network | |
CN111310804B (en) | Method, device, equipment and medium for processing vehicle fault problem based on decision tree | |
CN106557773B (en) | Vehicle fault diagnosis method and system | |
CN111273634A (en) | Arrangement of an at least partially automatic control system of a motor vehicle | |
CN115563503A (en) | Fault prediction method, system and storage medium based on vehicle historical data | |
CN111271183A (en) | Method and system for self-adaptive online prediction of state of diesel engine | |
CN113799788A (en) | Method, computer program and device for processing a signal | |
CN116968556A (en) | Power and energy storage battery fault diagnosis method based on fuzzy entropy | |
CN111523609A (en) | Vehicle data processing method and device, computer equipment and storage medium | |
CN114911982A (en) | Vehicle fault early warning method and device, terminal equipment and storage medium | |
CN116009515A (en) | Method, device, system and medium for managing faults of vehicle | |
CN114750773B (en) | Drive-by-wire chassis control method, system, device, controller and storage medium | |
CN114715054B (en) | Pressure stability detection method, vehicle-mounted ECU, processor and vehicle | |
CN114750773A (en) | Line control chassis control method, system, device, controller and storage medium | |
CN110077234B (en) | Electric vehicle safety tree updating method and electric vehicle | |
CN117572852B (en) | Vehicle component fault analysis method, device, equipment, medium and product | |
Varghese et al. | A Feature-Based Machine Learning Approach for Mixed-Criticality Systems | |
CN118013231B (en) | Intelligent automobile fault diagnosis method and device and intelligent automobile instrument panel system |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
TA01 | Transfer of patent application right | ||
TA01 | Transfer of patent application right |
Effective date of registration: 20240113 Address after: No. 13 Xingxiang Road, Zengjia Town, High tech Zone, Jiulongpo District, Chongqing, 400039 Applicant after: Chongqing Selis Phoenix Intelligent Innovation Technology Co.,Ltd. Address before: 610095 No. 2901, floor 29, unit 1, building 1, No. 151, Tianfu Second Street, high tech Zone, China (Sichuan) pilot Free Trade Zone, Chengdu, Sichuan Province Applicant before: Chengdu Thalys Technology Co.,Ltd. |