CN104675988A - Vehicle EHC (electrohydraulic control) fault diagnostic method - Google Patents
Vehicle EHC (electrohydraulic control) fault diagnostic method Download PDFInfo
- Publication number
- CN104675988A CN104675988A CN201410586448.6A CN201410586448A CN104675988A CN 104675988 A CN104675988 A CN 104675988A CN 201410586448 A CN201410586448 A CN 201410586448A CN 104675988 A CN104675988 A CN 104675988A
- Authority
- CN
- China
- Prior art keywords
- rvm
- ehc
- vehicle
- car load
- individuality
- 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
Classifications
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F16—ENGINEERING ELEMENTS AND UNITS; GENERAL MEASURES FOR PRODUCING AND MAINTAINING EFFECTIVE FUNCTIONING OF MACHINES OR INSTALLATIONS; THERMAL INSULATION IN GENERAL
- F16H—GEARING
- F16H61/00—Control functions within control units of change-speed- or reversing-gearings for conveying rotary motion ; Control of exclusively fluid gearing, friction gearing, gearings with endless flexible members or other particular types of gearing
- F16H61/12—Detecting malfunction or potential malfunction, e.g. fail safe; Circumventing or fixing failures
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F16—ENGINEERING ELEMENTS AND UNITS; GENERAL MEASURES FOR PRODUCING AND MAINTAINING EFFECTIVE FUNCTIONING OF MACHINES OR INSTALLATIONS; THERMAL INSULATION IN GENERAL
- F16H—GEARING
- F16H61/00—Control functions within control units of change-speed- or reversing-gearings for conveying rotary motion ; Control of exclusively fluid gearing, friction gearing, gearings with endless flexible members or other particular types of gearing
- F16H61/12—Detecting malfunction or potential malfunction, e.g. fail safe; Circumventing or fixing failures
- F16H2061/1256—Detecting malfunction or potential malfunction, e.g. fail safe; Circumventing or fixing failures characterised by the parts or units where malfunctioning was assumed or detected
- F16H2061/126—Detecting malfunction or potential malfunction, e.g. fail safe; Circumventing or fixing failures characterised by the parts or units where malfunctioning was assumed or detected the failing part is the controller
- F16H2061/1264—Hydraulic parts of the controller, e.g. a sticking valve or clogged channel
Landscapes
- Engineering & Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Mechanical Engineering (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
Abstract
The invention relates to a vehicle EHC (electrohydraulic control) fault diagnostic method and belongs to the technical field of vehicles. The vehicle EHC fault diagnostic method includes performing preprocessing, to be specific, subjecting prior data of vehicle EHC system parameters to normalization processing and establishing the corresponding relation between the system parameters and a vehicle EHC state; performing machine training, to be specific, selecting appropriate a kernel function, performing genetic algorithm optimization training on the hyper-parameters of the kernel function and setting an appropriate RVM (relevance vector machine) model; diagnosing faults, to be specific, adopting one-by-one RVM classifier to diagnose faults of samples to be detected and output results. According to the arrangement, the EHC system sample data normalized is trained by the relevance vector to establish the RVM model with the parameters optimized, and the vehicle EHC system fault detection test is achieved by the aid of the model. The vehicle EHC fault diagnostic method is good in robustness and higher in generalization ability.
Description
Technical field
The invention belongs to automobile technical field, be specifically related to a kind of car load electrichydraulic control method for diagnosing faults.
Background technique
Along with day by day enriching of automatic transmission case product, on market, automatic speed changing vehicle kind gets more and more, and obtains liking and affirmative of a lot of client.The after sales service technical support of automatic transmission case product, Measuring error and follow-up of quality have become the vital task that automobile 4S shop and automobile vendor will face.
Automatic transmission case electrichydraulic control loop is the important subtense angle of gearbox product, and its complex structure, specifically comprising the component such as solenoid valve, valve body, oil hydraulic circuit, control valve, is the crucial actuator that automatic transmission case completes various functional task.So the fault diagnosis in automatic transmission case electrichydraulic control loop is the Focal point and difficult point of gearbox product On-Board Diagnostics service, is badly in need of corresponding device and method support.
Summary of the invention
The invention provides a kind of car load electrichydraulic control method for diagnosing faults, utilize and gather car load electrohydraulic control system parameter and set up RVM model, judge whether to there is fault.
Technological scheme of the present invention is: a kind of car load electrichydraulic control method for diagnosing faults, comprise the steps: step one: pretreatment: the priori data of car load electrohydraulic control system parameter is normalized, and set up the corresponding relation between described systematic parameter and car load electrichydraulic control state; Step 2: machine is trained: select suitable kernel function and carry out genetic algorithm optimization training to its hyper parameter, setting up suitable RVM model; Step 3: fault diagnosis: adopt " one to one " RVM classifier to carry out sample to be tested fault diagnosis and Output rusults.In described step 2, kernel function is gaussian radial basis function kernel function.In described step one, car load electrohydraulic control system parameter is retardation pressure, turns to oil pressure, operating oil pressure, fuel level.In described step 2, the step of genetic algorithm optimization RVM hyper parameter is: a. adopts real coding scheme to encode to initial weight and threshold value, obtains N code string, and each yard of string just correspond to the weights and threshold of one group of RVM algorithm; B. input training sample, calculate its error function value, evaluate the quality of connection weight (threshold value) with error sum of squares inverse, namely
, wherein
; C. population of future generation is entered by selection opertor by the individuality that fitness is large; D. with crossing-over rate and aberration rate, crossover and mutation operation is carried out to individuality, produce the individuality made new advances; E. calculate new individual fitness, new individuality is inserted in population simultaneously; If the individuality f. found reaches fitness criteria, then terminate algorithm, otherwise go to step b; G. the optimization initial value calculated using genetic algorithm as initial weight, with RVM Algorithm for Training network until draw diagnostic result.
The present invention has following good effect: utilize and gather car load electrohydraulic control system parameter and set up RVM model, judge whether to there is fault.The learning ability of the existing RVM algorithm of the inventive method and robustness, have again the very strong ability of searching optimum of genetic algorithm.
Accompanying drawing explanation
Fig. 1 is the method flow diagram of the specific embodiment of the invention.
Embodiment
Contrast accompanying drawing below, by the description to embodiment, the specific embodiment of the present invention is as the effect of the mutual alignment between the shape of involved each component, structure, each several part and annexation, each several part and working principle, manufacturing process and operation using method etc., be described in further detail, have more complete, accurate and deep understanding to help those skilled in the art to inventive concept of the present invention, technological scheme.
Diagnostic system of the present invention comprises brake-pressure sensor, steering oil pressure sensor, operating oil pressure sensor, fuel level sender and central processing unit (CPU), and brake-pressure sensor, steering oil pressure sensor, operating oil pressure sensor, fuel level sender are by the retardation pressure detected, the real time data turning to oil pressure, operating oil pressure, fuel level.Utilize inner setting thresholding to retardation pressure by central processing unit (CPU), turn to oil pressure, operating oil pressure, fuel level to contrast, due to detect retardation pressure, turn to oil pressure, operating oil pressure, between fuel level and car load electrichydraulic control fault, there is certain non-linear relation, therefore machine learning algorithm can be utilized to set up corresponding relation between them, thus realize fault diagnosis.But because the fluctuation of each machine training result is comparatively large, easily occurred study or owed study phenomenon, make judged result occur deviation, accuracy and the robustness of diagnosis and distinguish result still have much room for improvement.The car load electrichydraulic control diagnostic method that the present invention proposes is based on Method Using Relevance Vector Machine (relevance vector machine, RVM) Fault Identification technology, utilize the hyper parameter of genetic algorithm optimization RVM kernel function, set up the relation between each sensing data and engine operating state by " one to one " classifier, improve precision and the reliability of fault diagnosis.
According to automobile failure diagnosis knowledge, retardation pressure, turn to oil pressure, operating oil pressure, fuel level and car load electrichydraulic control fault diagnosis closely bound up, therefore car load electricity liquid fault diagnosis can be judged according to sensor parameters, theoretical according to this, machine learning method can be utilized to carry out learning training to priori data sample, the machinery diagnosis model trained is used for the analyzing and diagnosing of car load electrichydraulic control fault.
RVM algorithm by " kernel function " nonlinear problem of lower dimensional space is mapped to higher dimensional space and changes into linear problem, and the ingenious part of kernel function is that mapping process does not increase a lot of amounts of calculation.Compared with the SVM algorithm relatively more early proposed, the selection of RVM algorithm Kernel Function is not by the restriction of Mercer condition, due to its separate openness and probability, be the algorithm that in machine learning method, Generalization Capability is the most outstanding theoretically, especially effective to the machine learning of small sample.
The method obtains the outstanding sparse solution of Generalization Capability, is be based upon on the basis of suitable Selection of kernel function and optimum configurations.In order to obtain best setting, genetic algorithm be have employed to the optimization of RVM hyper parameter herein, and " genetic algorithm-RVM model " is applied among RVM engine diagnosis.
Fault diagnosis system composition based on " genetic algorithm-RVM model " can be divided into 3 main process:
(1) pretreatment: to retardation pressure, turn to the priori data of oil pressure, operating oil pressure, fuel level to be normalized, and set up retardation pressure, turn to oil pressure, corresponding relation between operating oil pressure, fuel level and car load electrichydraulic control state;
2) machine training: select suitable kernel function and genetic algorithm optimization training is carried out to its hyper parameter, setting up suitable RVM model;
3) fault diagnosis: adopt " one to one " RVM classifier to carry out sample to be tested fault diagnosis and Output rusults.
Conventional RVM kernel function has 4 kinds:
Linear kernel function:
K( x，z) = x·z ( 1)
Polynomial kernel function:
K( x，z) = [ s( x·z) + c]
q( 2)
Gaussian radial basis function (RBF) kernel function:
K( x，z ) = exp(-λ‖x － z‖
2) ( 3)
Sigmoid kernel function:
K (x，z) = tanh［s( x·z) + c］ ( 4)
Select suitable kernel function to be the key that the method can successfully use, trained by testing authentication, more respective Generalization Capability, select RBF kernel function as the RVM model of fault diagnosis herein.
Genetic algorithm optimization RVM algorithmic procedure is as follows: genetic algorithm optimization RVM algorithm mainly utilizes the search of genetic algorithm not rely on gradient information, only need the feasible solution of solved function under constraint conditio, very low to the applicable elements of objective function, no matter whether continuously linear is all in Applicable scope for objective function, and algorithm search is by force of overall importance, be easy to get optimal solution.Therefore genetic algorithm combines with RVM algorithm, then can make full use of both advantages, makes learning ability and the robustness of the existing RVM algorithm of new algorithm, has again the very strong ability of searching optimum of genetic algorithm.In general, the combination of genetic algorithm and RVM algorithm can be carried out in 3 aspects, and one is the optimization to network connection weight and threshold value; Two is the optimization to network structure; Three is the optimization to networking rule.
A. adopt real coding scheme to encode to initial weight and threshold value, obtain N code string, each yard of string just correspond to the weights and threshold of one group of RVM algorithm.
B. input training sample, calculate its error function value, evaluate the quality of connection weight (threshold value) with error sum of squares inverse, namely
, wherein
.Error is less, illustrates that fitness is larger.
C. population of future generation is entered by selection opertor by the individuality that fitness is large.
D. with crossing-over rate and aberration rate, crossover and mutation operation is carried out to individuality, produce the individuality made new advances.
E. calculate new individual fitness, new individuality is inserted in population simultaneously.
If the individuality f. found reaches fitness criteria, then terminate algorithm, otherwise go to step b.
G. the optimization initial value calculated using genetic algorithm as initial weight, with RVM Algorithm for Training network until draw diagnostic result.
Above by reference to the accompanying drawings to invention has been exemplary description; obvious specific implementation of the present invention is not subject to the restrictions described above; as long as have employed the improvement of the various unsubstantialities that method of the present invention is conceived and technological scheme is carried out; or design of the present invention and technological scheme directly applied to other occasion, all within protection scope of the present invention without to improve.
Claims (4)
1. a car load electrichydraulic control method for diagnosing faults, is characterized in that: comprise the steps:
Step one: pretreatment: the priori data of car load electrohydraulic control system parameter is normalized, and set up the corresponding relation between described systematic parameter and car load electrichydraulic control state;
Step 2: machine is trained: select suitable kernel function and carry out genetic algorithm optimization training to its hyper parameter, setting up suitable RVM model;
Step 3: fault diagnosis: adopt " one to one " RVM classifier to carry out sample to be tested fault diagnosis and Output rusults.
2. car load electrichydraulic control method for diagnosing faults according to claim 1, is characterized in that: in described step 2, and kernel function is gaussian radial basis function kernel function.
3. car load electrichydraulic control method for diagnosing faults according to claim 1, is characterized in that: in described step one, and car load electrohydraulic control system parameter is retardation pressure, turns to oil pressure, operating oil pressure, fuel level.
4. car load electrichydraulic control method for diagnosing faults according to claim 1, is characterized in that: in described step 2, and the step of genetic algorithm optimization RVM hyper parameter is:
A. adopt real coding scheme to encode to initial weight and threshold value, obtain N code string, each yard of string just correspond to the weights and threshold of one group of RVM algorithm;
B. input training sample, calculate its error function value, evaluate the quality of connection weight (threshold value) with error sum of squares inverse, namely
, wherein
;
C. population of future generation is entered by selection opertor by the individuality that fitness is large;
D. with crossing-over rate and aberration rate, crossover and mutation operation is carried out to individuality, produce the individuality made new advances;
E. calculate new individual fitness, new individuality is inserted in population simultaneously;
If the individuality f. found reaches fitness criteria, then terminate algorithm, otherwise go to step b;
G. the optimization initial value calculated using genetic algorithm as initial weight, with RVM Algorithm for Training network until draw diagnostic result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410586448.6A CN104675988A (en) | 2014-10-28 | 2014-10-28 | Vehicle EHC (electrohydraulic control) fault diagnostic method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410586448.6A CN104675988A (en) | 2014-10-28 | 2014-10-28 | Vehicle EHC (electrohydraulic control) fault diagnostic method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104675988A true CN104675988A (en) | 2015-06-03 |
Family
ID=53311416
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410586448.6A Pending CN104675988A (en) | 2014-10-28 | 2014-10-28 | Vehicle EHC (electrohydraulic control) fault diagnostic method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104675988A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105868777A (en) * | 2016-03-25 | 2016-08-17 | 北京理工大学 | Fault diagnosis method for power compartment of armored car of related vector machine based on optimization |
| CN107045739A (en) * | 2016-02-05 | 2017-08-15 | 福特全球技术公司 | Vehicle data adjustment diagnostic test based on collection |
| CN112085100A (en) * | 2020-09-09 | 2020-12-15 | 中国北方车辆研究所 | RVM-based fault diagnosis method suitable for AT product |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5204816A (en) * | 1990-03-29 | 1993-04-20 | Eaton Corporation | Throttle error detection logic |
| US5609067A (en) * | 1994-10-14 | 1997-03-11 | Caterpillar Inc. | Transmission control fault detection |
| US20040010361A1 (en) * | 2002-07-15 | 2004-01-15 | Georg Gierer | Device for emergency operation of motor vehicle with automatic transmission |
| CN102762901A (en) * | 2009-12-16 | 2012-10-31 | 艾里逊变速箱公司 | System and method for detecting clutch-related faults in an automatic transmission |
-
2014
- 2014-10-28 CN CN201410586448.6A patent/CN104675988A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5204816A (en) * | 1990-03-29 | 1993-04-20 | Eaton Corporation | Throttle error detection logic |
| US5609067A (en) * | 1994-10-14 | 1997-03-11 | Caterpillar Inc. | Transmission control fault detection |
| US20040010361A1 (en) * | 2002-07-15 | 2004-01-15 | Georg Gierer | Device for emergency operation of motor vehicle with automatic transmission |
| CN102762901A (en) * | 2009-12-16 | 2012-10-31 | 艾里逊变速箱公司 | System and method for detecting clutch-related faults in an automatic transmission |
Non-Patent Citations (2)
| Title |
|---|
| 李敏君: "基于遗传算法的故障诊断研究", 《微计算机信息》 * |
| 毕晓君，柳长源，卢迪: "基于PSO-RVM算法的发动机故障诊断", 《哈尔滨工程大学学报》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107045739A (en) * | 2016-02-05 | 2017-08-15 | 福特全球技术公司 | Vehicle data adjustment diagnostic test based on collection |
| CN105868777A (en) * | 2016-03-25 | 2016-08-17 | 北京理工大学 | Fault diagnosis method for power compartment of armored car of related vector machine based on optimization |
| CN105868777B (en) * | 2016-03-25 | 2019-05-24 | 北京理工大学 | A kind of Method Using Relevance Vector Machine panzer piggyback pod method for diagnosing faults based on optimization |
| CN112085100A (en) * | 2020-09-09 | 2020-12-15 | 中国北方车辆研究所 | RVM-based fault diagnosis method suitable for AT product |
| CN112085100B (en) * | 2020-09-09 | 2024-04-23 | 中国北方车辆研究所 | RVM-based fault diagnosis method suitable for AT products |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Shao et al. | Rolling bearing fault diagnosis using adaptive deep belief network with dual-tree complex wavelet packet | |
| CN104712542B (en) | A kind of reciprocating compressor sensitive features based on Internet of Things are extracted and method for diagnosing faults | |
| CN107884475A (en) | A kind of city gas pipeline method for diagnosing faults based on deep learning neutral net | |
| CN109299705A (en) | Rotary machinery fault diagnosis method based on one-dimensional depth residual error convolutional neural networks | |
| CN112033463B (en) | Nuclear power equipment state evaluation and prediction integrated method and system | |
| CN107842713B (en) | Submarine pipeline magnetic flux leakage data based on KNN-SVR lacks interpolating method | |
| Wang et al. | A combination of residual and long–short-term memory networks for bearing fault diagnosis based on time-series model analysis | |
| CN105758645A (en) | Probabilistic neural network based engine fault diagnosis system | |
| Vong et al. | Simultaneous-fault detection based on qualitative symptom descriptions for automotive engine diagnosis | |
| Guan et al. | Rolling bearing fault diagnosis based on information fusion and parallel lightweight convolutional network | |
| CN110991471B (en) | Fault diagnosis method for high-speed train traction system | |
| CN105487009A (en) | Motor fault diagnosis method based on k-means RBF neural network algorithm | |
| CN109964182A (en) | Method and system for vehicle analysis | |
| CN113538353B (en) | Five-phase asynchronous motor rolling bearing fault diagnosis method based on single-channel diagram data enhancement and migration training residual error network | |
| CN103559542A (en) | Extension neural network pattern recognition method based on priori knowledge | |
| CN115859077A (en) | Multi-feature fusion motor small sample fault diagnosis method under variable working conditions | |
| CN104675988A (en) | Vehicle EHC (electrohydraulic control) fault diagnostic method | |
| Ren et al. | A deep learning method for fault detection of autonomous vehicles | |
| CN112132102A (en) | Intelligent fault diagnosis method combining deep neural network with artificial bee colony optimization | |
| CN104635146B (en) | Analog circuit fault diagnosis method based on random sinusoidal signal test and HMM (Hidden Markov Model) | |
| Wang et al. | Predictability of vehicle fuel consumption using LSTM: Findings from field experiments | |
| CN104216397B (en) | Failure recognition and detection method for intelligent drive axle system | |
| Jia et al. | Multitask convolutional neural network for rolling element bearing fault identification | |
| CN116340817A (en) | Intelligent fault identification method for hydraulic piston pump | |
| CN110033105B (en) | Suspension system fault detection method for unbalanced data set under complex working condition |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20150603 |
|
| RJ01 | Rejection of invention patent application after publication |