CN104568344A - EGR (Exhaust Gas Recirculation) pipeline fault detection diagnosis method - Google Patents
EGR (Exhaust Gas Recirculation) pipeline fault detection diagnosis method Download PDFInfo
- Publication number
- CN104568344A CN104568344A CN201410527573.XA CN201410527573A CN104568344A CN 104568344 A CN104568344 A CN 104568344A CN 201410527573 A CN201410527573 A CN 201410527573A CN 104568344 A CN104568344 A CN 104568344A
- Authority
- CN
- China
- Prior art keywords
- data
- egr
- rvm
- rvm model
- diagnosis method
- 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
Landscapes
- Exhaust-Gas Circulating Devices (AREA)
Abstract
The invention relates to an EGR (Exhaust Gas Recirculation) pipeline fault detection diagnosis method and belongs to the field of engines. The method comprises the following steps: 1, creating a database by utilizing known pipeline data and performing data preprocessing; 2, creating an RVM (Relevance Vector Machine) model and performing optimal training on RVM model parameters by utilizing particle swarm optimization; 3, training the created RVM model by utilizing the known database; 4, performing diagnostic analysis on EGR pipeline detection data by adopting the RVM model; 5, creating a corresponding relationship between a detection data result and an EGR pipeline state. According to the method, an angle sensor and a pressure sensor are arranged on the front sides of an EGR valve and a turbocharger respectively, and a corresponding relationship between the opening change amplitude and the pressure value change amplitude of the EGR valve in the opening change control process of the EGR valve in normal working and fault states can be detected, so that the accuracy of the detection result is improved and the function deficiency during EGR pipeline detection in the prior art is overcome.
Description
Technical field
The present invention relates to engine art, be specifically related to a kind of EGR pipeline fault checkout and diagnosis method.
Background technology
Engine is as the heart of engineering machinery, and the quality of its performance is directly connected to dynamic property, economy, reliability, the feature of environmental protection and the security of the operation of engineering machinery whole system.Along with the raising of engine reinforcing degree, the structure of engine also becomes very complicated, and condition of work is also very severe, and the possibility broken down increases greatly.
Exhaust gas recirculatioon (ExhaustGasRecirculation, EGR) in engine refers to that a part for Exhaust Gas is imported the technology that suction side makes its air-breathing once again by engine after combustion.Its fundamental purpose is that the oxides of nitrogen (NO x) that can reduce in Exhaust Gas can improve rate of fuel consumption rate with when sharing sub-load.
For EGR engine, need that corresponding detection is carried out to the fault of egr system and ensure the security that engine runs.Wherein, the fault diagnosis of EGR valve itself can be carried out electric fault by EGR valve self-sensor device and judged clamping stagnation or electrical damage; In addition, also need to judge the gas leakage on EGR pipe road or obstruction; To this, mode of the prior art is, adopt intake flow sensor charge flow rate is detected, thus can according to charge flow rate whether exceed default detected value judge EGR pipe road whether exist gas leakage or block.But the voltage on EGR pipe road or resistance problems cause engine to there will be the situations such as difficulty in starting, Idling wobble, bad Acceleration, are the problems that intake flow sensor can't detect, affect the normal use of engine.
Summary of the invention
The deficiency of function when detecting EGR pipe road to overcome in prior art, the invention provides a kind of EGR pipeline fault checkout and diagnosis method.
Technical scheme of the present invention is: a kind of EGR pipeline fault checkout and diagnosis method, and its step comprises:
Step one, utilize known tubes circuit-switched data building database, line number of going forward side by side Data preprocess;
Step 2, set up RVM model, utilize particle cluster algorithm optimization to train RVM model parameter;
Step 3, the RVM model utilizing the training of given data storehouse to set up;
Step 4, employing RVM model are treated EGR pipe drive test data and are carried out diagnostic analysis;
Step 5, foundation detect the corresponding relation between data result and EGR pipe line state.
Data prediction in described step one is normalized data.
Particle cluster algorithm step in described step 2 is:
A. initialization population: the scale determining population, initial position and speed, according to the value of constraint condition to each particle initialization Lagrange factor a;
B. the target function value of each particle is calculated, i.e. the value of wanted majorized function;
C. the position local optimum Pbest and global optimum Gbest of each particle a is upgraded;
D. flying speed and the position of each particle a is upgraded;
E. judge whether data reach RVM model criteria, and the standard of reaching jumps out circulation, and calculate related coefficient, otherwise the step 2 returned), until meet the number of times of iteration;
F. the value of optimum a is returned, and by optimized Parameter transfer to RVM model.
G. EGR pipeline fault checkout and diagnosis method according to claim 1, it is characterized in that, the real-time pressure data that the testing data in described step 4 comprises the acquisition of the pressure transducer before being located at turbosupercharger and the angle delta data that the angular transducer being located at EGR valve is measured.
The RVM model that described step 4 uses is " one to one " RVM sorter.
The present invention has following good effect: be employed herein before angular transducer and pressure transducer be arranged at EGR valve and turbosupercharger respectively, the aperture amplitude of variation of EGR valve and the corresponding relation of force value amplitude of variation in the process controlled with the EGR valve aperture change under malfunction under normal operating conditions can be detected, RVM model through utilizing particle cluster algorithm optimised carries out Data Comparison and analysis, increase the accuracy rate of testing result, simultaneously the method is except detecting EGR pipe road whether fault, can also detect EGR pipe road is principal fault or plugging fault, specify that the result of fault detect, facilitate checking and keeping in repair of staff.
Accompanying drawing explanation
Fig. 1 is the workflow diagram of EGR pipeline fault checkout and diagnosis method in the present invention;
Fig. 2 is the workflow diagram of particle cluster algorithm in the present invention;
Fig. 3 is the work structuring figure of EGR pipeline fault checkout and diagnosis method in the present 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 principle of work, 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, technical scheme.
As shown in Figure 1-Figure 3, a kind of EGR pipeline fault checkout and diagnosis method, its step comprises:
S01 step one, utilize given data building database, line number of going forward side by side Data preprocess.Data in given data storehouse be pressure transducer before being located at turbosupercharger under normal operating conditions measure according to the angular transducer being located at EGR valve angle change time the real-time pressure data that records, pressure data corresponding when being the opening angle change of EGR valve, these data are that data when being non-faulting state under normal operating conditions form given data storehouse.Pre-service is normalized data, and normalization can accelerate the convergence of training network, and normalized concrete effect is the statistical distribution concluding unified samples.No matter be in order to modeling or in order to calculate, first basic measuring unit is same, particle cluster algorithm be with the statistics of sample in event respectively probability carry out training and predicting, normalization is same statistical probability distribution between 0-1; SVM classifies with linear partition distance after dimensionality reduction and emulates, and therefore the normalization of space-time dimensionality reduction is the statistics coordinate distribution be unified between-1--+1.
S02 step 2, set up RVM model, utilize particle cluster algorithm optimization to train RVM model parameter.Set up RVM model and first select suitable function, and carry out PSO optimization training to its hyper parameter, set up suitable RVM model, utilize particle cluster algorithm optimization to train RVM model parameter, allow model more easily restrain, arithmetic speed is faster.When setting up RVM model, first utilize known sample database to carry out particle cluster algorithm optimization and train successful Modling model, the foundation of known sample database obtains preservation when EGR pipe road normal condition.
RVM kernel function conventional during the selection of kernel function has 4 kinds:
Linear kernel function:
Polynomial kernel function:
Gaussian radial basis function (RBF) kernel function:
Sigmoid kernel function:
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.
Particle swarm optimization algorithm (particle swarm optimization, PSO) is a kind of optimizing algorithm based on iteration ［ 8 ］ proposed first in 1995 by Kennedy and Eberhart.This algorithm is the simulation to flock of birds social action, PSO algorithm and genetic algorithm similar, be a kind of optimized algorithm based on colony (population), each particle is by carrying out information interaction with other particles, adjust the Evolutionary direction of oneself, and avoid being absorbed in local optimum; Meanwhile, PSO algorithm adopts the random searching strategy being different from genetic algorithm, operates than genetic algorithm easy too much, therefore demonstrates more remarkable performance when solving some optimization problem.
Utilize the Lagrange multiplier in particle swarm optimization algorithm optimization Method Using Relevance Vector Machine herein, the optimal value that this vector of Lagrange multiplier meets each component of constraint condition in RVM is found by utilizing PSO, make the spacing distance between two classification maximum, thus construct optimal hyperlane.During initialization population, should constantly judge until the random initial value of each particle meets the constraint condition in optimized Method Using Relevance Vector Machine.The Lagrange multiplier a that PSO optimizes Method Using Relevance Vector Machine realizes by calling the M file subroutine pso.m write.Each component of each particle a is by self study and learn to other particles, constantly updates self speed and position, reaches global optimum.
The step of particle cluster algorithm is:
A. initialization population: the scale determining population, initial position and speed, according to the value of constraint condition to each particle initialization Lagrange factor a;
B. the target function value of each particle is calculated, i.e. the value of wanted majorized function;
C. the position local optimum Pbest and global optimum Gbest of each particle a is upgraded;
D. flying speed and the position of each particle a is upgraded;
E. judge whether data reach RVM model criteria, and the standard of reaching jumps out circulation, and calculate related coefficient, otherwise the step 2 returned), until meet the number of times of iteration;
F. the value of optimum a is returned, and by optimized Parameter transfer to RVM model.
The RVM model obtained after hyperparameter optimization training, namely can be used for classification and the process of data.PSO to the parameter optimisation procedure of RVM algorithm as shown in Fig. 2.
In RVM algorithm, the classification accuracy of selection to RVM algorithm of hyper parameter plays conclusive effect, in the past conventional parameter optimization method many employings people is for enumerating the mode such as optimizing, cross validation parameters, but this class methods required time is long, also there is the problem being easily absorbed in local optimum simultaneously.Particle cluster algorithm is a kind of global optimizing algorithm efficiently, and the parameter optimization that can be used for machine learning algorithm is arranged.Adopt the hyper parameter of PSO algorithm optimization RVM algorithm to arrange herein, thus set up the machine mould of fault diagnosis.
S03 step 3, the RVM model utilizing the training of given data storehouse to set up.Known sample database is set up, and is to utilize particle cluster algorithm to be optimized training to RVM model parameter in step 2, obtains a suitable RVM model, can be used for the classification process of data.Step 3 carries out machine training for utilizing given data storehouse to RVM model, namely records the data under normal steady state, can ensure the differentiation of fault mode and the use of model.
Under stationary conditions, the aperture change of EGR valve can make the pressure before turbosupercharger produce corresponding change, and the force value that the situation of change of the pressure before turbosupercharger can be obtained by pressure transducer herein obtain.When steady working condition is approximate, particularly under same steady working condition, when EGR pipeline is non-faulting, the aperture of EGR valve carries out the change of certain amplitude, and the force value that the pressure transducer before being located at turbosupercharger can be made to obtain also carries out the change of corresponding amplitude thereupon; And when the blocking of EGR pipe road or gas leakage, if still the aperture of EGR valve to be carried out the change of original amplitude, the amplitude of variation being now located at the force value that the pressure transducer before turbosupercharger obtains will have larger difference with original amplitude of variation.RVM model is after given data training, and under recording normal non-faulting state, the force value that angle change is corresponding, preserves and contrast with the data under malfunction.
S04 step 4, employing RVM model carry out diagnostic analysis to testing data.The angle delta data that the angular transducer that the real-time pressure data that testing data comprises the acquisition of the pressure transducer before being located at turbosupercharger is located at EGR valve is measured, pressure transducer instead of traditional inlet air flow sensor, more can measure the failure problems on EGR pipe road in conjunction with the angular transducer being arranged at EGR valve, reduce erroneous judgement rate in prior art.
Under malfunction, angle changes the Data Comparison under corresponding force value and non-faulting state, is vicissitudinous certainly, so, adopt RVM model to carry out diagnostic analysis to testing data.If EGR line clogging, now the aperture of EGR valve is carried out the change of original amplitude, the amplitude of variation being now located at the force value that the pressure transducer before turbosupercharger obtains is just much larger than original amplitude of variation.In addition, if the gas leakage of EGR pipe road, now the aperture of EGR valve is carried out the change of original amplitude, the amplitude of variation being now located at the force value that the pressure transducer before turbosupercharger obtains is just much little than original amplitude of variation.
During fault diagnosis, the RVM model of employing is " one to one " RVM sorter, and in order to carry out multicategory discriminant, be employed herein " one to one " RVM sorter, this sorter is most widely used general in the many sorting techniques of current RVM, and nicety of grading is relatively high.
S05 step 5, foundation detect the corresponding relation between data result and EGR pipe line state.Shown in Fig. 3 a 3 classification ABC represents the state on 3 kinds of EGR pipe roads respectively: non-faulting, principal fault and plugging fault.
The data of the pressure transducer before the turbosupercharger that when aperture of the EGR valve of preserving when testing data is exactly EGR pipeline fault detection changes, angular transducer is corresponding, testing data is after RVM Model Diagnosis is analyzed, and the result of output is EGR pipe road principal fault, EGR pipe road plugging fault or non-faulting state.
When the difference of the pressure data under the pressure data under normal condition and state to be measured belongs to pre-set interval value, judge that EGR pipe road is as non-faulting.When the difference of the pressure data under the pressure data under normal condition and state to be measured is less than the minimum value of described pre-set interval value, judge that EGR pipe road is as principal fault.When the difference of the pressure data under the pressure data under normal condition and state to be measured is greater than the maximal value of described pre-set interval value, judge that described EGR pipeline is as plugging fault.
RVM model has carried out classifying process to data, other data of Similarity Class in testing data and given data storehouse can be compared, it not the comparison of the independent of conventional processors or standard value, the comparative analysis of mass data can make the accuracy of diagnostic result higher, rate of false alarm reduces greatly, improve the accuracy rate of testing, alleviate the workload of staff.
As shown in Figure 3, after fault detect exports fault type, this detection method is also in operation.Whether after fault type exports, detect engine simultaneously and work, if engine quits work, then detection system can power cut-off; If engine continues in operation, then angle and pressure transducer continue Real-Time Monitoring task, wait in the detection circulation system of RVM model prediction diagnosis as testing data after the data that sensor exports carry out pre-service, continue the running status detecting EGR pipe road, ensure the normal work of engine.
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 technical scheme is carried out; or design of the present invention and technical scheme directly applied to other occasion, all within protection scope of the present invention without to improve.
Claims (5)
1. an EGR pipeline fault checkout and diagnosis method, is characterized in that, its step comprises:
Step one, utilize known tubes circuit-switched data building database, line number of going forward side by side Data preprocess;
Step 2, set up RVM model, utilize particle cluster algorithm optimization to train RVM model parameter;
Step 3, the RVM model utilizing the training of given data storehouse to set up;
Step 4, employing RVM model carry out diagnostic analysis to EGR pipe circuit-switched data to be measured;
Step 5, foundation detect the corresponding relation between data result and EGR pipe line state.
2. EGR pipeline fault checkout and diagnosis method according to claim 1, is characterized in that, the data prediction in described step one is normalized data.
3. EGR pipeline fault checkout and diagnosis method according to claim 1, is characterized in that, the particle cluster algorithm step in described step 2 is:
Initialization population: the scale determining population, initial position and speed, according to the value of constraint condition to each particle initialization Lagrange factor a;
Calculate the target function value of each particle, i.e. the value of wanted majorized function;
Upgrade the position local optimum Pbest and global optimum Gbest of each particle a;
Upgrade flying speed and the position of each particle a;
Judge whether data reach RVM model criteria, and the standard of reaching jumps out circulation, and calculate related coefficient, otherwise the step 2 returned), until meet the number of times of iteration;
Return the value of optimum a, and by optimized Parameter transfer to RVM model.
4. EGR pipeline fault checkout and diagnosis method according to claim 1, it is characterized in that, the real-time pressure data that the testing data in described step 4 comprises the acquisition of the pressure transducer before being located at turbosupercharger and the angle delta data that the angular transducer being located at EGR valve is measured.
5. EGR pipeline fault checkout and diagnosis method according to claim 1, is characterized in that, the RVM model that described step 4 uses is " one to one " RVM sorter.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410527573.XA CN104568344A (en) | 2014-10-09 | 2014-10-09 | EGR (Exhaust Gas Recirculation) pipeline fault detection diagnosis method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410527573.XA CN104568344A (en) | 2014-10-09 | 2014-10-09 | EGR (Exhaust Gas Recirculation) pipeline fault detection diagnosis method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104568344A true CN104568344A (en) | 2015-04-29 |
Family
ID=53084925
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410527573.XA Pending CN104568344A (en) | 2014-10-09 | 2014-10-09 | EGR (Exhaust Gas Recirculation) pipeline fault detection diagnosis method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104568344A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106768071A (en) * | 2017-01-24 | 2017-05-31 | 桂林师范高等专科学校 | Sewage monitoring system |
| CN107941436A (en) * | 2016-10-13 | 2018-04-20 | 福建宁德核电有限公司 | A kind of method and device for calculating non-return valve slip |
| CN111780933A (en) * | 2020-07-01 | 2020-10-16 | 华能国际电力股份有限公司大连电厂 | Method and system for diagnosing leakage fault of high-pressure heater based on neural network and thermodynamic modeling |
| CN113412419A (en) * | 2018-12-07 | 2021-09-17 | 阿特拉斯·科普柯空气动力股份有限公司 | Gas network and method for detecting leaks in a gas network under pressure or under vacuum |
| CN113887770A (en) * | 2020-07-01 | 2022-01-04 | 哈尔滨工业大学（威海） | Aero-engine full-life maintenance decision optimization algorithm based on problem decoupling |
| CN114252213A (en) * | 2021-12-22 | 2022-03-29 | 北京金茂人居环境科技有限公司 | Heating and ventilation circulating pipeline water leakage monitoring system and control method |
| CN115096596A (en) * | 2022-06-27 | 2022-09-23 | 中国第一汽车股份有限公司 | Method, device, equipment and medium for diagnosing low-pressure EGR (exhaust gas Recirculation) system of engine |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040206074A1 (en) * | 2002-09-06 | 2004-10-21 | Yoshiyuki Akao | Failure detection apparatus for an internal combustion engine |
| WO2011132289A1 (en) * | 2010-04-22 | 2011-10-27 | トヨタ自動車株式会社 | Abnormality detection device and abnormality detection method for egr system |
| CN102893011A (en) * | 2010-05-11 | 2013-01-23 | 丰田自动车株式会社 | Method for specifying egr rate in internal combustion engine, and control device for internal combustion engine |
| CN103726930A (en) * | 2013-12-09 | 2014-04-16 | 潍柴动力股份有限公司 | Fault detection device and method of exhaust gas recirculation engine |
| CN103759907A (en) * | 2014-01-23 | 2014-04-30 | 潍柴动力股份有限公司 | EGR pipeline fault detection method |
-
2014
- 2014-10-09 CN CN201410527573.XA patent/CN104568344A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040206074A1 (en) * | 2002-09-06 | 2004-10-21 | Yoshiyuki Akao | Failure detection apparatus for an internal combustion engine |
| WO2011132289A1 (en) * | 2010-04-22 | 2011-10-27 | トヨタ自動車株式会社 | Abnormality detection device and abnormality detection method for egr system |
| CN102893011A (en) * | 2010-05-11 | 2013-01-23 | 丰田自动车株式会社 | Method for specifying egr rate in internal combustion engine, and control device for internal combustion engine |
| CN103726930A (en) * | 2013-12-09 | 2014-04-16 | 潍柴动力股份有限公司 | Fault detection device and method of exhaust gas recirculation engine |
| CN103759907A (en) * | 2014-01-23 | 2014-04-30 | 潍柴动力股份有限公司 | EGR pipeline fault detection method |
Non-Patent Citations (3)
| Title |
|---|
| 宋进桂: "废气再循环系统的故障分析与诊断", 《拖拉机与农用运输车》 * |
| 毕晓君 等: "基于PSO-RVM算法的发动机故障诊断", 《哈尔滨工程大学学报》 * |
| 胡明江 等: "应用回声状态网络的EGR阀电路在线诊断", 《振动、测试与诊断》 * |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107941436A (en) * | 2016-10-13 | 2018-04-20 | 福建宁德核电有限公司 | A kind of method and device for calculating non-return valve slip |
| CN107941436B (en) * | 2016-10-13 | 2020-01-31 | 福建宁德核电有限公司 | method and device for calculating leakage rate of check valve |
| CN106768071A (en) * | 2017-01-24 | 2017-05-31 | 桂林师范高等专科学校 | Sewage monitoring system |
| CN113412419A (en) * | 2018-12-07 | 2021-09-17 | 阿特拉斯·科普柯空气动力股份有限公司 | Gas network and method for detecting leaks in a gas network under pressure or under vacuum |
| CN113412419B (en) * | 2018-12-07 | 2024-05-28 | 阿特拉斯·科普柯空气动力股份有限公司 | Gas network and method for detecting leaks in a gas network |
| CN111780933A (en) * | 2020-07-01 | 2020-10-16 | 华能国际电力股份有限公司大连电厂 | Method and system for diagnosing leakage fault of high-pressure heater based on neural network and thermodynamic modeling |
| CN113887770A (en) * | 2020-07-01 | 2022-01-04 | 哈尔滨工业大学（威海） | Aero-engine full-life maintenance decision optimization algorithm based on problem decoupling |
| CN111780933B (en) * | 2020-07-01 | 2022-04-15 | 华能国际电力股份有限公司大连电厂 | Method and system for diagnosing leakage fault of high-pressure heater based on neural network and thermodynamic modeling |
| CN113887770B (en) * | 2020-07-01 | 2024-04-16 | 哈尔滨工业大学（威海） | Aero-engine life-span maintenance decision optimization algorithm based on problem decoupling |
| CN114252213A (en) * | 2021-12-22 | 2022-03-29 | 北京金茂人居环境科技有限公司 | Heating and ventilation circulating pipeline water leakage monitoring system and control method |
| CN115096596A (en) * | 2022-06-27 | 2022-09-23 | 中国第一汽车股份有限公司 | Method, device, equipment and medium for diagnosing low-pressure EGR (exhaust gas Recirculation) system of engine |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104568344A (en) | EGR (Exhaust Gas Recirculation) pipeline fault detection diagnosis method | |
| CN104598813B (en) | Computer intrusion detection method based on integrated study and semi-supervised SVM | |
| Xu et al. | Predicting fan blade icing by using particle swarm optimization and support vector machine algorithm | |
| Sun et al. | Condition monitoring of wind turbine blades based on self-supervised health representation learning: A conducive technique to effective and reliable utilization of wind energy | |
| Sharma et al. | Modeling and analytics for cyber-physical systems in the age of big data | |
| US20120323343A1 (en) | Virtual sensor system and method | |
| CN109447236A (en) | A kind of method for diagnosing faults of hybrid vehicle heat management system | |
| Ruiz et al. | A new criterion to validate and improve the classification process of LAMDA algorithm applied to diesel engines | |
| Zheng et al. | Fault diagnosis method based on supervised particle swarm optimization classification algorithm | |
| CN113807418A (en) | Injection molding machine energy consumption abnormity detection method and system based on Gaussian mixture model | |
| Henningsson et al. | A virtual sensor for predicting diesel engine emissions from cylinder pressure data | |
| Shahpouri et al. | Hybrid machine learning approaches and a systematic model selection process for predicting soot emissions in compression ignition engines | |
| CN116842459A (en) | Electric energy metering fault diagnosis method and diagnosis terminal based on small sample learning | |
| Liao et al. | A comparative investigation of advanced machine learning methods for predicting transient emission characteristic of diesel engine | |
| Sun et al. | Self-updating continual learning classification method based on artificial immune system | |
| Boccardo et al. | Development of a virtual methodology based on physical and data-driven models to optimize engine calibration | |
| CN104697765A (en) | Method and system for detecting faults of automotive air conditioner | |
| Liao et al. | Prediction of the transient emission characteristics from diesel engine using temporal convolutional networks | |
| CN104594969A (en) | Oil pressure fault detection method | |
| He et al. | Robust key parameter identification of dedicated hybrid engine performance indicators via K-fold filter collaborated feature selection | |
| CN204027820U (en) | Air conditioning for automobiles fault detection system | |
| Chen et al. | ActGraph: prioritization of test cases based on deep neural network activation graph | |
| Zong et al. | Application of artificial fish swarm optimization semi-supervised kernel fuzzy clustering algorithm in network intrusion | |
| Wan et al. | Engine modelling architecture study for hybrid electric vehicle diagnosis application | |
| Zhou et al. | Deep learning approach for super-knock event prediction of petrol engine with sample imbalance |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20150429 |