CN114821856A - Intelligent auxiliary device connected in parallel to automobile traveling computer for rapid automobile maintenance - Google Patents
Intelligent auxiliary device connected in parallel to automobile traveling computer for rapid automobile maintenance Download PDFInfo
- Publication number
- CN114821856A CN114821856A CN202210402808.7A CN202210402808A CN114821856A CN 114821856 A CN114821856 A CN 114821856A CN 202210402808 A CN202210402808 A CN 202210402808A CN 114821856 A CN114821856 A CN 114821856A
- Authority
- CN
- China
- Prior art keywords
- fault
- data
- database
- intelligent
- vehicle
- 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.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07C—TIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE
- G07C5/00—Registering or indicating the working of vehicles
- G07C5/08—Registering or indicating performance data other than driving, working, idle, or waiting time, with or without registering driving, working, idle or waiting time
- G07C5/0808—Diagnosing performance data
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01D—MEASURING NOT SPECIALLY ADAPTED FOR A SPECIFIC VARIABLE; ARRANGEMENTS FOR MEASURING TWO OR MORE VARIABLES NOT COVERED IN A SINGLE OTHER SUBCLASS; TARIFF METERING APPARATUS; MEASURING OR TESTING NOT OTHERWISE PROVIDED FOR
- G01D21/00—Measuring or testing not otherwise provided for
- G01D21/02—Measuring two or more variables by means not covered by a single other subclass
-
- 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
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- 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
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- 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
- G06N3/086—Learning methods using evolutionary algorithms, e.g. genetic algorithms or genetic programming
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Physiology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Testing And Monitoring For Control Systems (AREA)
Abstract
The invention belongs to the research and development field of automobile maintenance equipment, discloses an intelligent auxiliary device for rapid automobile maintenance connected in parallel with a traveling computer, and can solve the problems that manual maintenance cannot be performed on specific fault parts accurately and failure of fault diagnosis cannot be caused due to networking. The method comprises the following steps: the automobile running data information is obtained by additionally installing the quick plug-pull sensor and combining the original vehicle-mounted sensor and the vehicle-mounted computer, the automobile running data information is transmitted to the intelligent fault diagnosis embedded system by means of the CAN bus technology, and fault diagnosis is completed by means of the API model downloaded from the cloud. And (3) completing API model training at the cloud server, selecting a BP neural network with a genetic algorithm optimized weight threshold, and obtaining initial training data from a vehicle operation information historical database tested and constructed under laboratory conditions. And updating the database through user feedback. The device is mainly used for fault state identification and fault classification positioning of automobile operation.
Description
Technical Field
The invention belongs to the research and development field of automobile maintenance equipment, and particularly relates to an automobile rapid maintenance intelligent auxiliary device which comprises an automobile external rapid plugging sensor, a cloud service neural network model training system and an intelligent fault diagnosis embedded system, is connected in parallel with a traveling computer and has the functions of fault state identification and fault classification positioning.
Background
With the development of automobile manufacturing industry and the increase of automobile function demands of users, the total amount of automobiles is continuously increased, and the structures of all parts of the automobiles are gradually complicated. The complicated structure leads to more and more automobile fault types, and the difficulty of identifying and positioning automobile faults is increased. If a practical and effective fault diagnosis method is lacked, the automobile overhaul process is complicated, the waiting time of a user is increased, and the automobile overhaul cost is also increased. Therefore, how to quickly identify and locate the automobile fault becomes a concern.
In the existing common local fault diagnosis method, an automobile fault diagnosis instrument is used for reading a fault code in a memory of an on-board computer, so that maintenance personnel can be helped to find out the cause of the vehicle fault. However, the automobile fault diagnosis instrument only completes the explanation of the fault code and indicates the system where the fault is located (for example, the fault of the power module-battery management system), but the fault cannot be accurately detected to a specific fault component, and a maintenance worker still needs to further judge according to experience to find out a specific fault point. In the remote fault diagnosis method under the popularization background of the internet of vehicles, a vehicle-mounted end sends a fault signal to a cloud end through wireless transmission, and then detection and feedback are carried out through a detection technology stored in the cloud end. However, the car networking can only process the feedback information of the traditional vehicle-mounted sensor, and key targeted sensing information is lacked, so that the car networking diagnosis method has the defects of deficient congenital diagnosis key clue information and the like.
In summary, the existing device diagnosis technology lacks the accurate positioning capability for the fault component, and meanwhile, the car networking diagnosis technology cannot deploy sensors in a targeted manner to obtain key diagnosis clue information, so that the diagnosis efficiency and the intelligent diagnosis degree need to be improved.
Disclosure of Invention
The invention aims to design an intelligent auxiliary device for fast repairing of an automobile, which is connected in parallel with a traveling computer, aiming at the industrial problems that the existing equipment diagnosis technology lacks the accurate positioning capability of a fault part, meanwhile, the Internet of vehicles diagnosis technology cannot deploy a sensor in a targeted manner to obtain key diagnosis clue information, and the diagnosis efficiency and the intelligent diagnosis degree need to be improved. Firstly, training a plurality of conventional faults of an automobile system in an early stage by utilizing a leading edge neural network technology based on a large amount of proprietary experimental data; then, the model obtained by training is arranged in an embedded system, so that the embedded system device can identify the conventional faults of various systems of various automobiles; then, designing a device which can be quickly installed and disassembled by simply connecting the device in parallel with the vehicle-mounted computer, and effectively reading sensing information and vehicle-mounted computer alarm information in a vehicle-mounted system; finally, designing various sensors which can be rapidly deployed at key parts of the automobile and can be rapidly plugged in and pulled out of an embedded computer, and realizing direct and rapid acquisition of key fault diagnosis clue information of the automobile; the intelligent auxiliary device for automobile maintenance capable of achieving offline accurate information acquisition is constructed comprehensively, functions of fault state identification and fault classification and positioning of accident vehicles for maintenance engineers are achieved rapidly, and the intelligent auxiliary device has the advantages of being low in cost, high in precision, high in efficiency and strong in applicability.
An intelligent auxiliary device for fast maintenance of an automobile connected in parallel with a traveling crane computer comprises a fast plug-in sensor, an intelligent fault diagnosis embedded system and a cloud platform;
the quick plug-in sensor and the intelligent fault diagnosis embedded system are connected with the original vehicle-mounted sensor and the original vehicle-mounted computer of the automobile in parallel in a quick plug-in mode; the modularized quick plug-pull sensor is used for carrying out auxiliary measurement on vehicle state data; the intelligent fault diagnosis embedded system is used for receiving a quick plug-pull sensor signal, an original vehicle-mounted sensor signal of an automobile and original accident alarm information of a vehicle-mounted computer, and carrying out automobile fault state identification and fault classification positioning according to the received information;
the cloud platform comprises cloud storage and cloud computing; the cloud storage part constructs a vehicle operation information historical database which comprises original vehicle-mounted sensor information, rapid plugging and unplugging sensor information, vehicle-mounted computer accident alarm information and corresponding fault state and fault classification and positioning information; the cloud computing part uses a cloud-stored vehicle operation information historical database to perform fault state recognition and fault positioning core API model training;
the cloud storage part updates the database according to the feedback of the user diagnosis result;
the cloud computing part is combined with a genetic algorithm to carry out BP neural network parameter optimization when fault state identification and fault positioning core API model training are carried out;
the cloud computing part updates the database according to the user diagnosis result feedback in the cloud storage part, then carries out fault state recognition and fault positioning core API model training again, and downloads and updates the trained model to the intelligent fault diagnosis embedded system in a networking state; downloading a fault state identification and fault positioning core API model after cloud training to an intelligent fault diagnosis embedded system of a vehicle-mounted end for judging the fault state of the vehicle and classifying and positioning the faults;
original vehicle-mounted sensor information, rapid plugging sensor information and vehicle-mounted computer accident alarm information are transmitted to an intelligent fault diagnosis embedded system through a CAN bus, and the intelligent fault diagnosis embedded system performs data preprocessing operation on the information in a database to obtain input data of a fault state identification and fault positioning core API model;
the method comprises the steps of firstly training a constructed vehicle operation information historical database in an Ali cloud server, and establishing a model for performing fault state identification and fault classification positioning by using vehicle operation information data, namely mapping from vehicle operation information to definite faults.
In the fault state identification and fault location core API model training, firstly, training samples need to be extracted from a vehicle operation information historical database, and extracted data are divided into a training set, a testing set and a verification set;
for extracted data, data preprocessing is firstly required before training, and the specific steps are as follows:
step 1: deleting samples corresponding to the detected data repetition and data deletion from the database, and re-extracting new data from the database to supplement the training samples;
step 2: in consideration of the existence of data noise, the data noise needs to be removed by means including sliding filtering;
and step 3: selecting various characteristic data by adopting principal component analysis;
and 4, step 4: and (3) carrying out min-max normalization on the characteristic data selected by adopting the principal component analysis method, finishing the random extraction of the training sample, and preprocessing the data after the random extraction.
Using the preprocessed data to train the model for fault state identification and fault classification and location, and adopting a BP neural network combined with a genetic algorithm, wherein the specific steps are as follows:
step 1: constructing a plurality of BP neural network models by using the preprocessed data, wherein the activation function of the hidden layer is a tanh function, and the activation function of the output layer adopts a softmax function;
step 2: initializing a neural network connection weight and a threshold, and optimizing the weight and the threshold by adopting a genetic algorithm;
and step 3: carrying out real number coding on the weight and the threshold of each BP neural network diagnosis model obtained by building, and randomly selecting 100 initial individuals of the weight and the threshold corresponding to the real number coding to form an initial population;
and 4, step 4: calculating a loss function expressed by the sum of squared errors; taking the reciprocal of the loss function as an individual fitness function;
and 5: selecting, crossing and mutating individuals in the current population to form a new population of the next generation;
step 6: judging whether the new population obtained in the step 5 reaches a convergence condition, and finishing weight and threshold optimization if the new population reaches the convergence condition; if the convergence condition is not reached, returning to the step 5 for recalculation;
and 7: taking the data of the optimal individuals in the population as the initial weight and the threshold of the optimized BP neural network model, starting iterative training on the BP neural network model until the loss function value is smaller than the preset threshold or the number of iterations is reached, and finishing the training of the BP neural network model;
and 8: inputting the verification set into a plurality of trained BP neural network models, and selecting the neural network model with the best performance as a fault state identification and fault positioning core API model;
and step 9: and obtaining the accuracy of the API model through the test set.
In the updating process of the database content, data updating is carried out according to the following principles:
(1) when the diagnosis result of the intelligent fault diagnosis embedded system is correct and the data of the same fault state identification and fault classification positioning in the database does not reach the capacity value, directly updating the vehicle operation information data, the fault state identification and fault classification positioning data into the database;
(2) when the intelligent fault diagnosis embedded system has correct fault diagnosis results and the data corresponding to fault state identification and fault classification positioning in the database reaches a capacity value, the database is not updated;
(3) when the intelligent fault diagnosis embedded system has wrong fault diagnosis results and data corresponding to fault state identification and fault classification positioning in the database does not reach a capacity value, directly updating vehicle operation information data, fault state identification and fault classification positioning data to the database;
(4) when the intelligent fault diagnosis embedded system has wrong fault diagnosis results and the data corresponding to fault state identification and fault classification positioning in the database reaches a capacity value, the vehicle operation information data and the fault state identification and fault classification positioning data during diagnosis are used for randomly replacing a group of data corresponding to the same fault state identification and fault classification positioning in the database.
And buffering by adopting a cloud data message queue: in order to ensure that the information transmission of the intelligent fault diagnosis embedded system is matched with the updating process of the cloud database, an information queue is additionally arranged in the middle to serve as a data buffer area, so that the access pressure of a cloud server is reduced; the cloud platform receives operation information simultaneously sent by the intelligent fault diagnosis embedded systems, and the operation information is pushed to a message queue through the data storage module; the message queue is designed into a circular queue data structure, and the queue stores vehicle operation history information according to a first-in first-out sequence; starting a resident process, monitoring the data storage condition of the message queue in real time, taking out the data for updating the database once finding that new data information arrives in the queue, and then deleting the processed information in the queue.
The invention has the beneficial effects that: the sensors are quickly plugged and pulled, different sensors can be selectively used for different types of automobiles, and the flexibility and the universality of data acquisition are high; by combining the automobile sensor, the vehicle-mounted computer and other components, the fault state identification and fault classification positioning can be more accurate and comprehensive; the construction of the core API model is realized by applying a cloud training method, the cost for purchasing a server is saved, and the updating of the algorithm model is more convenient by using a cloud training mode; for an embedded system of a vehicle-mounted end, a model with functions of fault state recognition and fault classification positioning is downloaded from a cloud end, so that the phenomenon that fault diagnosis cannot be performed when networking cannot be performed is avoided, and meanwhile, the embedded system does not need model training, so that the computational cost is saved; the updating of the database is completed through the later feedback of the user, the overfitting problem of the model to the original fault and the classification and positioning problem of the model to the new fault are solved to a certain extent, and the real-time performance and the accuracy of the model are improved.
Drawings
FIG. 1 is a general block diagram of the present invention;
FIG. 2 is a fault state identification and fault location core API model training method of the present invention;
FIG. 3 is a database update method of the present invention for all vehicles;
fig. 4 is a process of API model training for fault diagnosis by using a BP cloud neural network platform according to the present invention.
Detailed Description
The invention will be further illustrated with reference to specific embodiments. It is to be understood that the examples are for illustrative purposes only and are not intended to limit the present invention.
Example 1:
the embodiment provides an intelligent auxiliary device with high reliability, which is connected in parallel with a traveling computer and has the functions of fault state identification and fault classification positioning for rapid maintenance of an automobile.
Before fault diagnosis analysis, the raw data of the operation of the vehicle is firstly acquired: the acquisition of the original data of the running of the automobile is completed by quickly plugging and unplugging the sensing sensor, the original vehicle-mounted sensor, the vehicle-mounted computer, the data acquisition card, the signal conditioner and other components. The method comprises the steps of collecting original data of automobile operation through a plug-in sensor, a vehicle-mounted original sensor and a vehicle-mounted computer, converting gain parameters of a combined program control amplifier into voltage signals, then performing discrete time sequence signal conversion on output continuous signals by using a data acquisition card, and outputting discrete voltage values.
Hardware equipment adopted for data acquisition is as follows:
(1) the sensor and the vehicle-mounted original sensor are quickly plugged and pulled out.
Corresponding sensors are respectively additionally arranged for the faults of components such as a fuel supply system, a cooling system, a starting system, an ignition system, a lubrication system and the like, for example, the sensors for measuring signals in the fuel supply system are as follows: oil tank level sensor: detecting whether the fuel in the fuel tank is too little or the fuel level is lower than the lower opening of the upper oil pipe hole to judge whether the fuel quantity is sufficient; oil pipe vibration signal sensor that oils: the vibration signal sensor is bound around the oil feeding pipe, and if the oil feeding pipe has the phenomena of welding failure, crack, fracture or loosening of an oil pipe joint, an abnormal vibration signal can be detected; fuel pipeline liquid pressure gauge: the fuel pressure is abnormal if the fuel pressure is blocked when the fuel pressure is placed on a pipeline passage where the gasoline filter is located.
(2) A signal conditioner.
(3) A data acquisition card.
As shown in fig. 2, the intelligent auxiliary device for quickly repairing an automobile with high reliability, which is connected in parallel to a traveling computer and has the functions of fault state identification and fault classification positioning, of the embodiment first needs to use vehicle operation information data, fault state identification and fault classification positioning data measured under laboratory conditions to form a vehicle operation information historical database.
As shown in fig. 2, a training set, a test set and a verification set are obtained by randomly extracting data from a vehicle operation information historical database and distributing the extracted data according to the proportion of 70%, 15% and 15%. In the process of randomly extracting data, the fault state parameter data corresponding to the fault of each category needs to be extracted.
For the extracted data, firstly, checking is needed, repeated data and missing data are deleted from a sample, meanwhile, information is reported to a database, and the repeated data and the missing data are deleted from the database; meanwhile, in order to ensure that the total amount of the extracted samples is unchanged, a corresponding number of samples need to be extracted from the database again to be used as supplement; and then, checking repeated data and missing data, and performing cycle operation in sequence until all the data in the sample are qualified.
And for the sample data subjected to repeated data and missing data detection, the data is subjected to filtering processing in consideration of the existence of data noise.
Performing feature extraction on the filtered data; considering that the extracted data features are too many and high redundancy possibly exists, selecting various feature data by adopting principal component analysis and performing data dimension reduction; and finishing the steps of randomly extracting the training samples and preprocessing the randomly extracted data.
After the data preprocessing is completed, as shown in fig. 4, the BP neural network model is trained by combining a genetic algorithm.
(1) And constructing a plurality of BP neural network models, wherein each BP neural network comprises n input neurons, d hidden layer neurons (the number of the hidden layer neurons of different BP neural network models is different) and m output neurons, the activation function of the hidden layer is a tanh activation function, and the activation function of the output layer is a softmax function.
(2) Firstly, optimizing parameters of a BP neural network model by adopting a genetic algorithm, and carrying out real number coding on the weight and the threshold of each BP neural network diagnosis model established in the step 1, wherein the coding length, namely the length of an individual chromosome is S; then a group of population with an individual specification number of 100 is randomly generated as 100 random solutions. Namely: and randomly selecting initial individuals of the weights and thresholds corresponding to 100 real number codes to form an initial population. Each initial individual represents an initial solution for finding the optimal initial weight and initial threshold.
(3) Calculating the fitness of each individual in the population (initial population) in step 2, firstly calculating a loss function, which is expressed by the sum of squared errors j (i), and the formula is:
where i 1.. said, N is the number of chromosomes, m is the number of nodes of the output layer, k is the number of training samples, C m Representing the actual value of the m-th output node, y m A predicted value representing the mth output node;
(4) calculating the fitness of the individual, and taking the reciprocal of the loss function as a fitness function F (i) of the individual:
(5) and (4) carrying out selection, crossing and mutation operations on individuals in the current population (the population generated in the step 2 or the population returned in the step 6) to form a new population of the next generation. The higher the fitness of the individual, the greater the probability of being selected, the probability of each individual being selected p (i):
for the cross probability, 0.4 is taken in the training; for the mutation probability, 0.1 was taken in the training.
(6) And (5) judging whether the new population obtained in the step (5) reaches a convergence condition, finishing genetic algorithm optimization if the new population reaches the convergence condition, and returning to the step (5) to calculate again if the new population does not reach the convergence condition.
(7) And taking the optimal individual data in the population which reaches the convergence condition as the initial weight and the threshold of the corresponding BP neural network model, optimizing the initial weight and the threshold by the constructed BP neural networks through a genetic algorithm, and then training the BP neural network model.
(8) Performing iterative training on the BP neural network models, and quitting the training of the neural network when the loss function is smaller than a preset threshold value or reaches a target iteration number to obtain a plurality of neural network models;
(9) after obtaining a plurality of neural network models, in order to verify the performance of the models, each model is tested by using a test set, the model which has the best performance to the test set is selected, the test set is input into the model which has the best performance, and the generalization capability of the model is preliminarily known.
Downloading the neural network model with the best performance to an intelligent fault diagnosis embedded system, connecting a quick plug-pull sensor with an original vehicle-mounted sensor and an original vehicle-mounted computer in parallel, inputting vehicle operation information data and original accident alarm information data into the embedded system for preprocessing (extracting the same characteristics used in the process of establishing an initial model), and then using a fault state recognition and fault positioning core API model in the embedded system for fault state recognition and fault classification positioning.
After the fault state identification and the fault classification positioning are finished, the user feeds back whether the diagnosis result is correct or not and the correct fault category through actual maintenance; the feedback result is audited by the diagnostician to ensure the validity of the data.
The database updating principle is as shown in fig. 3, along with the continuous updating of the database, the original laboratory collected data is replaced by the actual fault state identification and fault classification data, the data volume in the database is also increased to the rated capacity, and the accuracy of the fault state identification and fault positioning core API model fault identification obtained by training is improved; the intelligent auxiliary device for the rapid maintenance of the automobile is used for fault diagnosis of various automobiles, and the applicability of the trained fault state identification and fault positioning core API model is gradually enhanced along with the increase of data acquisition in the fault diagnosis process of various automobiles.
In the updating process of the database, when the accumulated diagnosis error quantity of a certain fault type reaches a set threshold value, the updated database is used for training the BP neural network model, the BP neural network model for fault state identification and fault classification positioning is obtained again, and the model is updated to the intelligent fault diagnosis embedded system of the vehicle.
TABLE 1 Fault location and Classification function Table of the invention
Claims (5)
1. An intelligent auxiliary device for fast maintenance of an automobile, which is connected in parallel with a traveling computer, is characterized by comprising a fast plugging sensor, an intelligent fault diagnosis embedded system and a cloud platform;
the quick plug-in sensor and the intelligent fault diagnosis embedded system are connected with the original vehicle-mounted sensor and the original vehicle-mounted computer of the automobile in parallel in a quick plug-in mode; the modularized quick plug-pull sensor is used for carrying out auxiliary measurement on vehicle state data; the intelligent fault diagnosis embedded system is used for receiving a quick plug-pull sensor signal, an original vehicle-mounted sensor signal of an automobile and original accident alarm information of a vehicle-mounted computer, and carrying out automobile fault state identification and fault classification positioning according to the received information;
the cloud platform comprises cloud storage and cloud computing; the cloud storage part constructs a vehicle operation information historical database which comprises original vehicle-mounted sensor information, rapid plugging and unplugging sensor information, vehicle-mounted computer accident alarm information and corresponding fault state and fault classification and positioning information; the cloud computing part uses a cloud-stored vehicle operation information historical database to perform fault state recognition and fault positioning core API model training;
the cloud storage part updates the database according to the feedback of the user diagnosis result;
the cloud computing part is combined with a genetic algorithm to carry out BP neural network parameter optimization when fault state identification and fault positioning core API model training are carried out;
the cloud computing part updates the database according to the user diagnosis result feedback in the cloud storage part, then carries out fault state recognition and fault positioning core API model training again, and downloads and updates the trained model to the intelligent fault diagnosis embedded system in a networking state; downloading a fault state identification and fault positioning core API model after cloud training to an intelligent fault diagnosis embedded system of a vehicle-mounted end for judging the fault state of the vehicle and classifying and positioning the faults;
original vehicle-mounted sensor information, rapid plugging sensor information and vehicle-mounted computer accident alarm information are transmitted to an intelligent fault diagnosis embedded system through a CAN bus, and the intelligent fault diagnosis embedded system performs data preprocessing operation on the information in a database to obtain input data of a fault state identification and fault positioning core API model;
the method comprises the steps of firstly training a constructed vehicle operation information historical database in an Ali cloud server, and establishing a model for performing fault state identification and fault classification positioning by using vehicle operation information data, namely mapping from vehicle operation information to definite faults.
2. The intelligent auxiliary device for automobile rapid maintenance according to claim 1, wherein in the fault state identification and fault location core API model training, training samples are firstly extracted from a vehicle operation information historical database, and the extracted data is divided into a training set, a testing set and a verification set;
for extracted data, data preprocessing is firstly required before training, and the specific steps are as follows:
step 1: deleting samples corresponding to the detected data repetition and data deletion from the database, and re-extracting new data from the database to supplement the training samples;
step 2: in consideration of the existence of data noise, the data noise needs to be removed by means including sliding filtering;
and step 3: selecting various characteristic data by adopting principal component analysis;
and 4, step 4: and (3) carrying out min-max normalization on the characteristic data selected by adopting the principal component analysis method, finishing the random extraction of the training sample, and preprocessing the data after the random extraction.
3. The intelligent auxiliary device for automobile rapid maintenance according to claim 2, wherein the model for fault state recognition and fault classification and location is trained by using preprocessed data, and a BP neural network is adopted in combination with a genetic algorithm, and the method comprises the following specific steps:
step 1: constructing a plurality of BP neural network models by using the preprocessed data, wherein the activation function of the hidden layer is a tanh function, and the activation function of the output layer adopts a softmax function;
step 2: initializing a neural network connection weight and a threshold, and optimizing the weight and the threshold by adopting a genetic algorithm;
and step 3: carrying out real number coding on the weight and the threshold of each BP neural network diagnosis model obtained by building, and randomly selecting 100 initial individuals of the weight and the threshold corresponding to the real number coding to form an initial population;
and 4, step 4: calculating a loss function expressed by the sum of squared errors; taking the reciprocal of the loss function as an individual fitness function;
and 5: selecting, crossing and mutating individuals in the current population to form a new population of the next generation;
and 6: judging whether the new population obtained in the step 5 reaches a convergence condition, and finishing weight and threshold optimization if the new population reaches the convergence condition; if the convergence condition is not reached, returning to the step 5 for recalculating;
and 7: taking the data of the optimal individuals in the population as the initial weight and the threshold of the optimized BP neural network model, starting iterative training on the BP neural network model until the loss function value is smaller than the preset threshold or the number of iterations is reached, and finishing the training of the BP neural network model;
and 8: inputting the verification set into a plurality of trained BP neural network models, and selecting the neural network model with the best performance as a fault state identification and fault positioning core API model;
and step 9: and obtaining the accuracy of the API model through the test set.
4. The intelligent auxiliary device for automobile rapid maintenance according to any one of claims 1-3, wherein during the database content updating process, data updating is performed according to the following principles:
(1) when the diagnosis result of the intelligent fault diagnosis embedded system is correct and the data of the same fault state identification and fault classification positioning in the database does not reach the capacity value, directly updating the vehicle operation information data, the fault state identification and fault classification positioning data into the database;
(2) when the intelligent fault diagnosis embedded system has correct fault diagnosis results and the data corresponding to fault state identification and fault classification positioning in the database reaches a capacity value, the database is not updated;
(3) when the intelligent fault diagnosis embedded system has wrong fault diagnosis results and data corresponding to fault state identification and fault classification positioning in the database does not reach a capacity value, directly updating vehicle operation information data, fault state identification and fault classification positioning data to the database;
(4) when the intelligent fault diagnosis embedded system has wrong fault diagnosis results and the data corresponding to fault state identification and fault classification positioning in the database reaches a capacity value, the vehicle operation information data and the fault state identification and fault classification positioning data during diagnosis are used for randomly replacing a group of data corresponding to the same fault state identification and fault classification positioning in the database.
5. The intelligent auxiliary device for automobile rapid maintenance according to claim 4, wherein the cloud data message queue is adopted for buffering: in order to ensure that the information transmission of the intelligent fault diagnosis embedded system is matched with the updating process of the cloud database, an information queue is additionally arranged in the middle to serve as a data buffer area, so that the access pressure of a cloud server is reduced; the cloud platform receives operation information simultaneously sent by the intelligent fault diagnosis embedded systems, and the operation information is pushed to a message queue through the data storage module; the message queue is designed into a circular queue data structure, and the queue stores vehicle operation history information according to a first-in first-out sequence; starting a resident process, monitoring the data storage condition of the message queue in real time, taking out the data for updating the database once finding that new data information arrives in the queue, and then deleting the processed information in the queue.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210402808.7A CN114821856B (en) | 2022-04-18 | 2022-04-18 | Intelligent auxiliary device connected in parallel to automobile traveling computer for rapid automobile maintenance |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210402808.7A CN114821856B (en) | 2022-04-18 | 2022-04-18 | Intelligent auxiliary device connected in parallel to automobile traveling computer for rapid automobile maintenance |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114821856A true CN114821856A (en) | 2022-07-29 |
| CN114821856B CN114821856B (en) | 2023-04-07 |
Family
ID=82535917
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210402808.7A Active CN114821856B (en) | 2022-04-18 | 2022-04-18 | Intelligent auxiliary device connected in parallel to automobile traveling computer for rapid automobile maintenance |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114821856B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116909496A (en) * | 2023-09-14 | 2023-10-20 | 山东索奇电子科技有限公司 | Vehicle fault data tracing method based on embedded high-speed data communication technology |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107851233A (en) * | 2015-06-19 | 2018-03-27 | 阿普泰克科技公司 | Local analytics at assets |
| GB201802284D0 (en) * | 2017-02-15 | 2018-03-28 | Ford Global Tech Llc | Feedback-based control model generation for an autonomous vehicle |
| CN108536123A (en) * | 2018-03-26 | 2018-09-14 | 北京交通大学 | The method for diagnosing faults of the train control on board equipment of the long neural network of memory network combination in short-term |
| US20190242936A1 (en) * | 2018-02-05 | 2019-08-08 | Wuhan University | Fault diagnosis method for series hybrid electric vehicle ac/dc converter |
| CN111221247A (en) * | 2018-11-23 | 2020-06-02 | 西门子股份公司 | Method and system for fault diagnosis, training method of model, and medium |
| WO2020244134A1 (en) * | 2019-06-05 | 2020-12-10 | 华南理工大学 | Multi-task feature sharing neural network-based intelligent fault diagnosis method |
| CN112785016A (en) * | 2021-02-20 | 2021-05-11 | 南京领行科技股份有限公司 | New energy automobile maintenance and fault monitoring and diagnosis method based on machine learning |
| CN112858917A (en) * | 2021-01-15 | 2021-05-28 | 哈尔滨工业大学（威海） | Battery system multi-fault diagnosis method based on genetic algorithm optimization neural network |
| CN113392572A (en) * | 2020-03-11 | 2021-09-14 | 福特全球技术公司 | Vehicle health calibration |
| US20210327163A1 (en) * | 2020-04-20 | 2021-10-21 | Innova Electronics Corporation | Router for vehicle diagnostic system |
| US20210335062A1 (en) * | 2020-04-23 | 2021-10-28 | Zoox, Inc. | Predicting vehicle health |
| US20210350636A1 (en) * | 2020-05-07 | 2021-11-11 | Nec Laboratories America, Inc. | Deep learning of fault detection in onboard automobile systems |
| CN113888856A (en) * | 2021-09-30 | 2022-01-04 | 江苏久智环境科技服务有限公司 | Monitoring system for providing operation for sprinkler based on road traffic fault judgment model |
-
2022
- 2022-04-18 CN CN202210402808.7A patent/CN114821856B/en active Active
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107851233A (en) * | 2015-06-19 | 2018-03-27 | 阿普泰克科技公司 | Local analytics at assets |
| GB201802284D0 (en) * | 2017-02-15 | 2018-03-28 | Ford Global Tech Llc | Feedback-based control model generation for an autonomous vehicle |
| US20190242936A1 (en) * | 2018-02-05 | 2019-08-08 | Wuhan University | Fault diagnosis method for series hybrid electric vehicle ac/dc converter |
| CN108536123A (en) * | 2018-03-26 | 2018-09-14 | 北京交通大学 | The method for diagnosing faults of the train control on board equipment of the long neural network of memory network combination in short-term |
| CN111221247A (en) * | 2018-11-23 | 2020-06-02 | 西门子股份公司 | Method and system for fault diagnosis, training method of model, and medium |
| WO2020244134A1 (en) * | 2019-06-05 | 2020-12-10 | 华南理工大学 | Multi-task feature sharing neural network-based intelligent fault diagnosis method |
| CN113392572A (en) * | 2020-03-11 | 2021-09-14 | 福特全球技术公司 | Vehicle health calibration |
| US20210327163A1 (en) * | 2020-04-20 | 2021-10-21 | Innova Electronics Corporation | Router for vehicle diagnostic system |
| US20210335062A1 (en) * | 2020-04-23 | 2021-10-28 | Zoox, Inc. | Predicting vehicle health |
| US20210350636A1 (en) * | 2020-05-07 | 2021-11-11 | Nec Laboratories America, Inc. | Deep learning of fault detection in onboard automobile systems |
| CN112858917A (en) * | 2021-01-15 | 2021-05-28 | 哈尔滨工业大学（威海） | Battery system multi-fault diagnosis method based on genetic algorithm optimization neural network |
| CN112785016A (en) * | 2021-02-20 | 2021-05-11 | 南京领行科技股份有限公司 | New energy automobile maintenance and fault monitoring and diagnosis method based on machine learning |
| CN113888856A (en) * | 2021-09-30 | 2022-01-04 | 江苏久智环境科技服务有限公司 | Monitoring system for providing operation for sprinkler based on road traffic fault judgment model |
Non-Patent Citations (1)
| Title |
|---|
| 张玉立: "《基于BBO 改进神经网络的汽车点火系统故障诊断研究》", 《信息与电脑》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116909496A (en) * | 2023-09-14 | 2023-10-20 | 山东索奇电子科技有限公司 | Vehicle fault data tracing method based on embedded high-speed data communication technology |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114821856B (en) | 2023-04-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108563214B (en) | Vehicle diagnosis method, device and equipment | |
| CN111950627B (en) | Multi-source information fusion method and application thereof | |
| CN107862763B (en) | Train safety early warning evaluation model training method, module and monitoring evaluation system | |
| CN102801552A (en) | System and methods for fault-isolation and fault-mitigation based on network modeling | |
| CN114821856B (en) | Intelligent auxiliary device connected in parallel to automobile traveling computer for rapid automobile maintenance | |
| CN115718802A (en) | Fault diagnosis method, system, equipment and storage medium | |
| CN113173104B (en) | New energy vehicle power battery early warning method and system | |
| CN109886433A (en) | The method of intelligent recognition city gas pipeline defect | |
| CN116610092A (en) | Method and system for vehicle analysis | |
| CN114185760A (en) | System risk assessment method and device and charging equipment operation and maintenance detection method | |
| CN108304567A (en) | High-tension transformer regime mode identifies and data classification method and system | |
| CN112668526A (en) | Bolt group loosening positioning monitoring method based on deep learning and piezoelectric active sensing | |
| CN114298521A (en) | Reliability analysis method, device, equipment and storage medium for urban rail equipment | |
| CN114839492A (en) | Method and device for identifying GIS partial discharge type based on MOBILE NETV3 | |
| CN114995346A (en) | Engine electrical diagnosis calibration method, device, equipment and medium | |
| CN112329341B (en) | Fault diagnosis system and method based on AR and random forest model | |
| CN114036647A (en) | Power battery safety risk assessment method based on real vehicle data | |
| CN111381165A (en) | Vehicle power battery monitoring method, device and platform | |
| CN116400672A (en) | Remote vehicle diagnosis method and system based on knowledge graph and rule engine | |
| CN115373366A (en) | Interactive diagnosis system, diagnosis method and storage medium | |
| CN114060132B (en) | NO based on emission remote monitoring x Sensor cheating discrimination method | |
| CN115913891A (en) | Big data analysis-based advanced operation and maintenance system and operation and maintenance method | |
| CN114118137A (en) | Power metering production equipment monitoring method and device, terminal and storage medium | |
| CN112199776A (en) | Locomotive full life cycle evaluation method, evaluation system and optimization method | |
| CN109946069A (en) | A kind of numerical control equipment drag chain reliability accelerated test method based on loading spectrum |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |