CN112282949B - Method and device for optimizing control parameters of ignition working condition of electric control gasoline engine and vehicle - Google Patents
Method and device for optimizing control parameters of ignition working condition of electric control gasoline engine and vehicle Download PDFInfo
- Publication number
- CN112282949B CN112282949B CN202011013259.1A CN202011013259A CN112282949B CN 112282949 B CN112282949 B CN 112282949B CN 202011013259 A CN202011013259 A CN 202011013259A CN 112282949 B CN112282949 B CN 112282949B
- Authority
- CN
- China
- Prior art keywords
- parameter
- observation data
- control parameters
- control
- gasoline engine
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 71
- 238000005457 optimization Methods 0.000 claims abstract description 178
- 238000012360 testing method Methods 0.000 claims abstract description 165
- 238000005070 sampling Methods 0.000 claims abstract description 64
- 238000012549 training Methods 0.000 claims abstract description 57
- 230000008569 process Effects 0.000 claims abstract description 27
- 239000000446 fuel Substances 0.000 claims description 35
- 238000002347 injection Methods 0.000 claims description 30
- 239000007924 injection Substances 0.000 claims description 30
- MWUXSHHQAYIFBG-UHFFFAOYSA-N nitrogen oxide Inorganic materials O=[N] MWUXSHHQAYIFBG-UHFFFAOYSA-N 0.000 claims description 24
- 239000013618 particulate matter Substances 0.000 claims description 24
- 230000002159 abnormal effect Effects 0.000 claims description 11
- 238000007781 pre-processing Methods 0.000 claims description 8
- 238000005259 measurement Methods 0.000 claims description 7
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 7
- UGFAIRIUMAVXCW-UHFFFAOYSA-N Carbon monoxide Chemical compound [O+]#[C-] UGFAIRIUMAVXCW-UHFFFAOYSA-N 0.000 claims description 6
- 229910002091 carbon monoxide Inorganic materials 0.000 claims description 6
- 238000002485 combustion reaction Methods 0.000 claims description 6
- 239000004215 Carbon black (E152) Substances 0.000 claims description 5
- 229930195733 hydrocarbon Natural products 0.000 claims description 5
- 150000002430 hydrocarbons Chemical class 0.000 claims description 5
- 238000010586 diagram Methods 0.000 description 10
- 230000000694 effects Effects 0.000 description 5
- 239000002245 particle Substances 0.000 description 5
- 238000011156 evaluation Methods 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 230000008859 change Effects 0.000 description 3
- 239000000243 solution Substances 0.000 description 3
- 239000003054 catalyst Substances 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 238000000889 atomisation Methods 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 239000000498 cooling water Substances 0.000 description 1
- 238000010790 dilution Methods 0.000 description 1
- 239000012895 dilution Substances 0.000 description 1
- 238000004836 empirical method Methods 0.000 description 1
- 239000003344 environmental pollutant Substances 0.000 description 1
- 238000012854 evaluation process Methods 0.000 description 1
- 238000001704 evaporation Methods 0.000 description 1
- 230000008020 evaporation Effects 0.000 description 1
- 238000013401 experimental design Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 239000007789 gas Substances 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 231100000719 pollutant Toxicity 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000000717 retained effect Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000013522 software testing Methods 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
Images
Classifications
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F02—COMBUSTION ENGINES; HOT-GAS OR COMBUSTION-PRODUCT ENGINE PLANTS
- F02D—CONTROLLING COMBUSTION ENGINES
- F02D41/00—Electrical control of supply of combustible mixture or its constituents
- F02D41/02—Circuit arrangements for generating control signals
- F02D41/14—Introducing closed-loop corrections
- F02D41/1401—Introducing closed-loop corrections characterised by the control or regulation method
- F02D41/1406—Introducing closed-loop corrections characterised by the control or regulation method with use of a optimisation method, e.g. iteration
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F01—MACHINES OR ENGINES IN GENERAL; ENGINE PLANTS IN GENERAL; STEAM ENGINES
- F01N—GAS-FLOW SILENCERS OR EXHAUST APPARATUS FOR MACHINES OR ENGINES IN GENERAL; GAS-FLOW SILENCERS OR EXHAUST APPARATUS FOR INTERNAL COMBUSTION ENGINES
- F01N11/00—Monitoring or diagnostic devices for exhaust-gas treatment apparatus, e.g. for catalytic activity
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F02—COMBUSTION ENGINES; HOT-GAS OR COMBUSTION-PRODUCT ENGINE PLANTS
- F02D—CONTROLLING COMBUSTION ENGINES
- F02D45/00—Electrical control not provided for in groups F02D41/00 - F02D43/00
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F02—COMBUSTION ENGINES; HOT-GAS OR COMBUSTION-PRODUCT ENGINE PLANTS
- F02M—SUPPLYING COMBUSTION ENGINES IN GENERAL WITH COMBUSTIBLE MIXTURES OR CONSTITUENTS THEREOF
- F02M65/00—Testing fuel-injection apparatus, e.g. testing injection timing ; Cleaning of fuel-injection apparatus
- F02M65/001—Measuring fuel delivery of a fuel injector
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F02—COMBUSTION ENGINES; HOT-GAS OR COMBUSTION-PRODUCT ENGINE PLANTS
- F02M—SUPPLYING COMBUSTION ENGINES IN GENERAL WITH COMBUSTIBLE MIXTURES OR CONSTITUENTS THEREOF
- F02M65/00—Testing fuel-injection apparatus, e.g. testing injection timing ; Cleaning of fuel-injection apparatus
- F02M65/003—Measuring variation of fuel pressure in high pressure line
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F02—COMBUSTION ENGINES; HOT-GAS OR COMBUSTION-PRODUCT ENGINE PLANTS
- F02D—CONTROLLING COMBUSTION ENGINES
- F02D41/00—Electrical control of supply of combustible mixture or its constituents
- F02D41/02—Circuit arrangements for generating control signals
- F02D41/14—Introducing closed-loop corrections
- F02D41/1401—Introducing closed-loop corrections characterised by the control or regulation method
- F02D2041/1433—Introducing closed-loop corrections characterised by the control or regulation method using a model or simulation of the system
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/06—Multi-objective optimisation, e.g. Pareto optimisation using simulated annealing [SA], ant colony algorithms or genetic algorithms [GA]
-
- 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)
- General Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Chemical & Material Sciences (AREA)
- Data Mining & Analysis (AREA)
- Mechanical Engineering (AREA)
- Combustion & Propulsion (AREA)
- Business, Economics & Management (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Mathematical Optimization (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Computational Mathematics (AREA)
- Mathematical Analysis (AREA)
- Evolutionary Biology (AREA)
- Operations Research (AREA)
- Life Sciences & Earth Sciences (AREA)
- Pure & Applied Mathematics (AREA)
- Strategic Management (AREA)
- Mathematical Physics (AREA)
- Human Resources & Organizations (AREA)
- Software Systems (AREA)
- Economics (AREA)
- Probability & Statistics with Applications (AREA)
- Development Economics (AREA)
- Chemical Kinetics & Catalysis (AREA)
- Game Theory and Decision Science (AREA)
- Databases & Information Systems (AREA)
- Algebra (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Medical Informatics (AREA)
- Computer Hardware Design (AREA)
Abstract
The embodiment of the invention discloses a method and a device for optimizing control parameters of an ignition working condition of an electric control gasoline engine and a vehicle, and relates to the technical field of vehicles. The optimization method comprises the following steps: carrying out a bench sampling test according to the parameter values of a plurality of groups of control parameters obtained in advance, and determining a plurality of groups of observation data, wherein the observation data comprise the parameter values of each group of control parameters and the acquisition amount of the optimization target corresponding to the parameter values obtained through the bench sampling test; iteratively dividing the multiple groups of observation data into a training group and a testing group, wherein the training group is used for establishing a regression model, the testing group is used for testing the regression model, and a predicted value of the optimization target output by the regression model is obtained in the testing process; and determining an optimal predicted value from the predicted values of the optimization targets output by the plurality of regression models, and then determining an optimal parameter value of the control parameter.
Description
Technical Field
The invention relates to the technical field of vehicles, in particular to a method and a device for optimizing control parameters of an ignition working condition of an electric control gasoline engine and a vehicle.
Background
In order to improve the pollution of vehicles in China to the atmospheric environment, the emission of an electric control gasoline engine (also called an electric control gasoline engine) is more strictly limited by light national six-emission regulations, and the requirements of the emission of particulate matters are particularly increased. In order to meet the requirements of regulations, vehicle manufacturers commonly use electrically controlled Gasoline engine Particulate traps (GPF) to reduce the amount of Particulate Matter (PN) and the amount of Particulate Matter (PM) discharged. However, in the catalyst ignition stage (ignition stage for short) of the cold start of the vehicle, because the water temperature of the electric control gasoline engine is low, and the fuel evaporation and atomization capabilities are poor, the combustion state of the mixed gas in the cylinder of the electric control gasoline engine in the catalyst ignition stage is poor, and more particulate matters are emitted. Referring to fig. 1, the particulate matter PN emission of an electrically controlled gasoline engine vehicle in the light-off phase is shown, and it can be seen that the vehicle emits more particulate matter in the light-off phase.
Parameters such as intake and exhaust VVT (variable Valve timing) angle, fuel injection pressure, injection time and proportion are important factors influencing the performance of the electric control gasoline engine in the ignition stage, and particularly influence the particulate matter emission. By means of the traditional optimization method of point-by-point testing, the number of the optimized parameters is increased exponentially, a large amount of bench time is consumed, and finally the development period and the cost are hard to bear.
Disclosure of Invention
The embodiment of the invention provides a method for optimizing control parameters of an ignition working condition of an electric control gasoline engine, which can effectively shorten the test period of the optimal parameter values of the control parameters and improve the test accuracy so as to overcome the problems. Correspondingly, the invention also provides an electric control gasoline engine ignition working condition control parameter optimization device and a vehicle, so that the electric control gasoline engine ignition working condition control parameter optimization method is applied, the emission of particulate matter PN in the vehicle ignition stage can be greatly reduced, a better value of the performance index of the electric control gasoline engine is obtained, the driving experience of a user is improved, and the electric control gasoline engine is more environment-friendly.
In order to solve the above problems, an embodiment of the present invention discloses a method for optimizing control parameters of an ignition condition of an electronic control gasoline engine, including:
determining control parameters and an optimization target influenced by the control parameters in a light-off stage of the electric control gasoline engine;
carrying out a bench sampling test according to the parameter values of a plurality of groups of control parameters obtained in advance, and determining a plurality of groups of observation data, wherein the observation data comprise the parameter values of each group of control parameters and the acquisition amount of the optimization target corresponding to the parameter values obtained through the bench sampling test;
iteratively dividing the multiple groups of observation data into a training group and a testing group, wherein the training group is used for establishing a regression model, the testing group is used for testing the regression model, and a predicted value of the optimization target output by the regression model is obtained in the testing process;
and determining an optimal predicted value from the predicted values of the optimization targets output by the regression models, and determining an optimal parameter value of the control parameter according to the optimal predicted value.
Further, the method further comprises:
acquiring the load and the rotating speed of the electric control gasoline engine through at least one real vehicle test;
acquiring a working condition distribution area under the ignition stage of the electric control gasoline engine according to the acquired load and the acquired rotating speed;
according to the distribution density of the working condition points in the working condition distribution area, counting the characteristic working condition points of the ignition stage of the electric control gasoline engine;
according to the parameter values of a plurality of groups of control parameters obtained in advance, carrying out a bench sampling test, determining a plurality of groups of observation data, wherein the observation data comprise the parameter values of each group of control parameters and the collection quantity of the optimization target corresponding to the parameter values obtained through the bench sampling test, and the bench sampling test comprises the following steps:
and according to the parameter values of a plurality of groups of control parameters obtained in advance, developing a bench sampling test at the characteristic working condition points, and determining a plurality of groups of observation data, wherein the observation data comprise the parameter values of each group of control parameters and the acquisition amount of the optimization target corresponding to the parameter values obtained through the bench sampling test.
Further, different electric control gasoline engines have different control parameters; wherein,
when the electric control gasoline engine is an air inlet injection PFI electric control gasoline engine, the control parameters comprise an air inlet variable valve timing VVT angle, an exhaust variable valve timing VVT angle, single or multiple fuel injection time and fuel proportion distribution corresponding to multiple injections;
when the electric control gasoline engine is an in-cylinder direct injection type GDI electric control gasoline engine, the control parameters comprise fuel rail pressure, fuel injection starting time SOI and fuel injection stopping time EOIT;
the optimization objectives affected by the control parameters include: particulate matter PN, fuel consumption FUELCOSP, combustion stability Cov, Torque, nitrogen oxides NOx, carbon monoxide CO, hydrocarbon HC.
Further, the method further comprises:
setting a parameter range of the control parameter;
preprocessing the parameter range of the control parameters to obtain a plurality of groups of parameter values of the control parameters;
according to the parameter values of a plurality of groups of control parameters obtained in advance, carrying out a bench sampling test, determining a plurality of groups of observation data, wherein the observation data comprise the parameter values of each group of control parameters and the collection quantity of the optimization target corresponding to the parameter values obtained through the bench sampling test, and the bench sampling test comprises the following steps:
and respectively inputting the parameter values of the plurality of groups of control parameters into a rack, and acquiring the original emission of the electric control gasoline engine and/or testing the performance index of the electric control gasoline engine through the rack to obtain a plurality of groups of acquisition quantities of the optimization target corresponding to the parameter values.
Further, preprocessing the parameter range of the control parameter to obtain a plurality of groups of parameter values of the control parameter, including:
setting a sampling interval for a parameter range of the control parameter;
and in the parameter range of the control parameters, extracting the parameter values of the control parameters according to the sampling interval to obtain a plurality of groups of parameter values of the control parameters.
Further, iteratively dividing the multiple sets of observation data into a training set and a testing set, where the training set is used to establish a regression model, and the testing set is used to test the regression model, and obtain a predicted value of the optimization target output by the regression model in a testing process, includes:
dividing the multiple groups of observation data into a training group and a testing group according to a preset distribution proportion, wherein the training group is used for establishing the regression model, and the testing group is used for testing the regression model;
aiming at observation data in a test group, in the test process, taking the parameter value of the control parameter in the observation data as the input of the regression model, and outputting to obtain the predicted value of the optimization target;
extracting the acquisition amount of the optimization target corresponding to the parameter value of the control parameter in the observation data, and comparing the predicted value of the optimization target with the acquisition amount of the optimization target to determine whether the observation data is abnormal;
if the observation data are normal, reserving a predicted value of the optimization target obtained through the observation data;
and if the observation data are abnormal, removing the observation data from the multiple groups of observation data, dividing the remaining multiple groups of observation data into a training group and a testing group, and repeating the steps.
Further, the method further comprises:
counting the number of normal groups of observation data;
evaluating the prediction accuracy of the regression models according to the normal group number of the observation data, the collection amount of the optimization target in each group of the observation data and the predicted value of the optimization target obtained through the observation data;
determining an optimal predicted value from the predicted values of the optimization targets output by the regression models, and determining an optimal parameter value of the control parameter according to the optimal predicted value, wherein the determining comprises the following steps:
taking the predicted value of the optimization target output by the regression model with the highest prediction precision as an optimal predicted value;
and determining observation data corresponding to the optimal predicted value, and taking the parameter value of the control parameter in the observation data as the optimal parameter value of the control parameter.
Further, the method further comprises:
according to a preset boundary condition, the water temperature of the electric control gasoline engine is controlled through a rack, and the starting environment of the electric control gasoline engine in the ignition stage is simulated;
under the starting environment, inputting the optimal parameter value of the control parameter into a rack, and collecting the actual measurement quantity of the optimization target corresponding to the optimal parameter value through the rack;
and comparing the measured quantity of the optimization target with the acquisition quantity of the optimization target so as to verify the optimal parameter value of the control parameter.
From another aspect of the embodiments of the present invention, the embodiments of the present invention further disclose an optimization device for controlling parameters of the ignition condition of an electric control gasoline engine, comprising:
the test parameter determining module is used for determining control parameters in the ignition stage of the electric control gasoline engine and an optimization target influenced by the control parameters;
the bench sampling test module is used for developing a bench sampling test according to the pre-obtained parameter values of a plurality of groups of control parameters, and determining a plurality of groups of observation data, wherein the observation data comprise the parameter values of each group of control parameters and the acquisition amount of the optimization target corresponding to the parameter values obtained through the bench sampling test;
the regression model training module is used for iteratively dividing the multiple groups of observation data into a training group and a testing group, wherein the training group is used for establishing a regression model, the testing group is used for testing the regression model, and a predicted value of the optimization target output by the regression model is obtained in the testing process;
and the optimal parameter value determining module is used for determining an optimal predicted value from the predicted values of the optimization targets output by the regression models, and determining the optimal parameter value of the control parameter according to the optimal predicted value.
From another aspect of the embodiment of the invention, the embodiment of the invention discloses a vehicle which comprises an electric control gasoline engine, wherein the optimal parameter value of the control parameter of the electric control gasoline engine in the ignition stage is obtained by the method for optimizing the ignition working condition control parameter of the electric control gasoline engine.
The embodiment of the invention has the following advantages:
according to the embodiment of the invention, control parameters in the ignition stage of the electric control gasoline engine and an optimization target influenced by the control parameters are determined; carrying out a bench sampling test according to the parameter values of a plurality of groups of control parameters obtained in advance, and determining a plurality of groups of observation data, wherein the observation data comprise the parameter values of each group of control parameters and the acquisition amount of the optimization target corresponding to the parameter values obtained through the bench sampling test; iteratively dividing the multiple groups of observation data into a training group and a testing group, wherein the training group is used for establishing a regression model, the testing group is used for testing the regression model, and a predicted value of the optimization target output by the regression model is obtained in the testing process; and finally, determining an optimal predicted value from the predicted values of the optimization targets output by the regression models, and determining an optimal parameter value of the control parameter according to the optimal predicted value, so that the optimal parameter value of the control parameter in the ignition stage of the electric control gasoline engine can be rapidly determined, the obtained optimal parameter value has higher precision, the emission of particulate matter PN in the ignition stage of the vehicle can be greatly reduced, a better value of the performance index of the electric control gasoline engine can be obtained, the test period can be effectively shortened, the test cost is saved, and the method has important significance for improving the driving experience of users and protecting the environment.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a schematic diagram of particulate matter PN emission of an electrically controlled gasoline engine vehicle during a light-off phase;
FIG. 2 is a flowchart illustrating the steps of a method for optimizing control parameters for ignition conditions of an electronically controlled gasoline engine according to an embodiment of the present invention;
FIG. 3 is a partially schematic illustration of control parameters and optimization objectives determined in accordance with an alternative embodiment of the present invention;
FIG. 4 is a schematic diagram of a test protocol for a certain optimization objective;
FIG. 5 is a schematic diagram of a model evaluation criterion for an embodiment of the present invention;
FIG. 6 is a schematic diagram of a distribution of characteristic operating points of an embodiment of the present invention;
FIG. 7 is a schematic structural diagram of a vehicle according to an embodiment of the present invention;
FIG. 8 is a schematic structural diagram of an electric control gasoline engine ignition condition control parameter optimization device according to an embodiment of the present invention;
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
Aiming at the technical problems in the background art, the embodiment of the invention provides a method for optimizing control parameters of an ignition working condition of an electric control gasoline engine, wherein the temperature of cooling water of the electric control gasoline engine is controlled to be 15-35 ℃ through a rack, aiming at characteristic working condition points in the ignition stage, a test scheme is designed based on a space filling method, optimization targets (such as PN (positive-negative) emission results) under different control parameter combinations are collected, a regression model is established and trained by using a Gaussian process algorithm, and the optimal prediction optimization target and the corresponding control parameter values are given. And determining the optimal performance and the corresponding parameter setting value through bench retesting. The method can effectively shorten the test period and improve the test accuracy through model prediction, and has important practical significance.
Referring to fig. 2, a flowchart of steps of a method for optimizing control parameters of a light-off condition of an electric control gasoline engine according to an embodiment of the present invention is shown, and the method may include the following steps:
step S201, determining control parameters and an optimization target influenced by the control parameters in the ignition stage of the electric control gasoline engine;
the control parameters are important influencing factors influencing the emission and the performance of the electric control gasoline engine in the ignition stage, and are input of an electric control gasoline engine control system. When the electrically controlled gasoline engine is a port injection pfi (port Fuel injection) electrically controlled gasoline engine, the control parameters may include intake variable valve timing VVT angle (iVVT), exhaust variable valve timing VVT angle (eVVT), single or multiple Fuel injection timings, and Fuel proportion distribution corresponding to multiple injections; for example, when the electric gasoline engine is a Direct Injection (gdi) electric gasoline engine, the control parameters may include fuel rail pressure, start of Injection (soi), end of Injection (eoit), and the like.
The optimization target is a target to be improved in the ignition stage, and is the output (Outputs) of an electronic control gasoline engine control system and is influenced by a control parameter. The optimization target of the embodiment of the invention can be the emissions of the electric control gasoline engine, such as particulate matter PN, nitrogen oxide NOx (Nitrogen oxide), carbon monoxide CO (carbon monoxide), hydrocarbon HC (hydrocarbon); the fuel consumption may be a performance index Of an electrically controlled gasoline engine, such as fuel consumption fuelcosp (fuel consumption), combustion stability cov (combustion Of variance), and Torque. Referring to FIG. 3, a partially schematic illustration of the control parameters and optimization objectives determined in accordance with an alternative embodiment of the present invention is shown.
After determining the control parameters and the optimization targets influenced by the control parameters in the ignition stage of the electric control gasoline engine, designing a test scheme aiming at the control parameters and the optimization targets.
Step S202, developing a bench sampling test according to the parameter values of a plurality of groups of control parameters obtained in advance, and determining a plurality of groups of observation data, wherein the observation data comprise the parameter values of each group of control parameters and the acquisition amount of the optimization target corresponding to the parameter values obtained through the bench sampling test;
the different optimization objectives are influenced by one or more control parameters, e.g. the amount of pm PN emitted can be influenced by three control parameters, i vvt, e vvt, EOIT, simultaneously. In the experimental design scheme, the design of the experimental scheme can be carried out by taking the same optimization target as a design idea. The method comprises the steps of determining control parameters (one or more may be possible) influencing the optimization target, setting a parameter range of each control parameter, designing the control parameters into a plurality of test groups within the parameter range, wherein the control parameters and the optimization target in each test group are the same, and the specific parameter values of the control parameters are different among different test groups, so that the influence of the control parameters under different parameter values on the optimization target can be effectively compared.
Based on the above, the embodiment of the present invention provides the following steps:
setting a parameter range of the control parameter;
and preprocessing the parameter range of the control parameters to obtain a plurality of groups of parameter values of the control parameters.
In the embodiment of the invention, the preprocessing can be performed on the parameter range of the control parameter by a Latin hypercube sampling or uniform sampling method, so that a plurality of samples are extracted from the parameter range of the control parameter to obtain a plurality of groups of parameter values of the control parameter. Taking uniform sampling as an example, the parameter range of the control parameter is preprocessed, and the main steps are as follows: setting a sampling interval for a parameter range of the control parameter; and in the parameter range of the control parameters, extracting the parameter values of the control parameters according to the sampling interval to obtain a plurality of groups of parameter values of the control parameters.
The following are exemplified: the control parameter is iVVT, the parameter range is-10 to 30 degrees, the span of the range is 40 degrees, the sampling interval is set to be 2, and then a plurality of groups of parameter values obtained by uniform sampling are 20, and are respectively 20 points with equal intervals of-10, -8 and-6 … …. And carrying out the same pretreatment on a plurality of control parameters influencing the same optimization target to obtain a plurality of groups of parameter values of the control parameters. And regarding the same optimization target, taking each group of parameter values of the plurality of control parameters as an experiment group, so as to obtain a plurality of experiment groups aiming at the optimization target. Referring to fig. 4, a test scenario for a certain optimization objective is shown, in which there are 7 experimental groups, and the parameter values of the control parameters in the experimental groups are obtained by latin hypercube sampling. By preprocessing the parameter range of the control parameter, the embodiment of the invention reduces the test amount of the bench sampling test, reduces the development period and saves the cost compared with the traditional optimization method of point-by-point test.
The specific implementation steps of step S202 are as follows:
and respectively inputting the parameter values of the plurality of groups of control parameters into a rack, and acquiring the original emission of the electric control gasoline engine and/or testing the performance index of the electric control gasoline engine through the rack to obtain a plurality of groups of acquisition quantities of the optimization target corresponding to the parameter values.
To test the same optimization objective, a bench sampling test is performed for each test group of the optimization objective. Taking the experimental group 2 of fig. 4 as an example, the iVVT is set to 2, the eVVT is set to-29, and the EOIT is set to 277 on the rack, and the original emission of the electric control gasoline engine is collected or the performance index of the electric control gasoline engine is tested by the rack under the parameter value of the experimental group. In order to reduce the amount of particulate matter PN, GPF is provided on an existing vehicle, the original emission of an electronic control gasoline engine refers to the emission of the electronic control gasoline engine before entering the GPF, and then the particulate matter and pollutant emission in the original emission are measured in a dilution sampling manner and the like, so as to obtain the collection amount of an optimization target (such as the particulate matter PN in the original emission or the fuel consumption FUELCOSP in a performance index) corresponding to the parameter value. In practice, different optimization objectives may be affected by the same control parameter, in which case the acquisition of multiple optimization objectives may be quickly obtained. Certainly, during the test, also can carry out rack sampling test to a plurality of optimization targets simultaneously, further practice thrift test cost, improve efficiency of software testing.
Step S203, iteratively dividing the multiple groups of observation data into a training group and a testing group, wherein the training group is used for establishing a regression model, the testing group is used for testing the regression model, and a predicted value of the optimization target output by the regression model is obtained in the testing process; the specific implementation process of the step comprises the following steps:
substep S203-1, dividing the multiple groups of observation data into a training group and a testing group according to a preset distribution proportion, wherein the training group is used for establishing the regression model, and the testing group is used for testing the regression model;
substep S203-2, regarding observation data in a test group, in the test process, taking parameter values of the control parameters in the observation data as the input of the regression model, and outputting to obtain a predicted value of the optimization target;
substep S203-3, extracting the acquisition amount of the optimization target corresponding to the parameter value of the control parameter in the observation data, and comparing the predicted value of the optimization target with the acquisition amount of the optimization target to determine whether the observation data is abnormal;
substep S203-4, if the observation data are normal, reserving a predicted value of the optimization target obtained through the observation data;
and a substep S203-5, if the observation data are abnormal, removing the observation data from the plurality of groups of observation data, dividing the plurality of groups of remaining observation data into a training group and a testing group, and repeating the steps.
In the embodiment of the invention, the control parameter is iVVT, the optimization target is fuel consumption rate as an example, the parameter range of the iVVT is-10 to 30 degrees, 20 groups of parameter values are obtained, and the collection amount of the fuel consumption rate corresponding to the 20 groups of parameter values is also 20 groups. Each set of parameter values and corresponding collected quantities of fuel consumption is a set of observations, in this case 20 sets.
Because the number of groups of observation data is large, in order to improve the output accuracy of the regression model, in the embodiment of the invention, a plurality of groups of observation data are divided into a training group and a testing group according to a preset distribution proportion, and the observation data of the training group and the testing group are generally 9: the ratio of 1 is assigned, that is, 9 observation data in the training set are used for establishing the regression model, and one observation data in the testing set is used for testing the regression model.
During training, taking a Gaussian process regression function as an example, the Gaussian process regression function is trained by using the observation data in the training set, so as to establish a regression model. Specifically, the parameter value of the control parameter in the observation data is used as the input of the gaussian process regression function, the acquisition quantity of the optimization target corresponding to the parameter value is used as the output of the gaussian process regression function, and the gaussian process regression function is trained through a plurality of groups of observation data in the training set, so that the regression model capable of expressing the relationship between the control parameter and the optimization target can be established. Because the control parameters are set and continuous, the regression model, which varies with the control parameters, can output the predicted values of the corresponding optimization objectives on this basis.
Then, the embodiment of the invention adopts the observation data of the test group in the same distribution to verify a regression model established by the observation data of the training group, after the parameter values of the control parameters in the observation data in the test group are input into the regression model, the regression model can predict the predicted value of the optimization target, then the predicted value of the optimization target is compared with the acquisition quantity of the optimization target in the observation data in the test group, whether the difference value between the predicted value and the acquisition quantity is within a preset difference threshold value is judged, if so, the predicted value and the acquisition quantity are close to each other, namely, the group of observation data is normal, and the predicted value of the optimization target obtained by the observation data is retained; otherwise, if the difference value of the two is not within the preset difference threshold value, the observation data is required to be removed if the set of observation data is abnormal. And then dividing the remaining multiple groups of observation data into a training group and a testing group to establish a new regression model, and repeating the substep S203-1 to the substep S203-5. Therefore, the embodiment of the invention can optimize the observation data, thereby achieving the purpose of optimizing the newly established regression model.
Taking the 7 sets of data in fig. 4 as an example, the 7 sets of control parameters respectively obtained by the bench collection test have 7 sets of collection amounts of the optimization target, and therefore, there are 7 sets of observation data. Dividing the 7 groups of observation data into 6 training groups and 1 testing group, for example, firstly, taking the 6 groups of observation data corresponding to ID numbers 1-6 as the training groups, taking the observation data corresponding to ID number 7 as the testing groups, establishing and training a regression model by using the control parameters in the 6 training groups and the acquisition quantity of the optimization target, after obtaining the regression model, taking the control parameters in the testing groups as the input of the regression model, and outputting through the regression model to obtain the predicted value of the optimization target. If the predicted value of the optimization target is closer to the collection amount of the optimization target in the test group, the test group (observed data) is normal; otherwise, the observation data where the test group is located is removed if the test group is abnormal.
And if the observation data corresponding to the ID 7 number is normal, continuing optimization, taking 6 groups of observation data corresponding to the ID 2-7 numbers as training groups, taking the observation data corresponding to the ID 1 number as test groups, establishing and training regression models by using control parameters in the 6 training groups and the acquisition amount of an optimization target, repeating the steps, judging whether the observation data of the test groups are normal, if so, keeping the observation data normal, and if not, rejecting the observation data abnormal.
If the observation data corresponding to the ID 7 is abnormal, the observation data is removed, only 6 groups of observation data participate in the distribution of the training group and the testing group in the subsequent optimization process, for example, 5 groups of observation data corresponding to the ID 1-5 are used as the training group, the observation data corresponding to the ID 6 is used as the testing group, a regression model is established and trained by using the control parameters in the 5 training groups and the acquisition amount of the optimization target, and the optimization process is repeated.
After the optimization process of the substep S203-1 to the substep S203-5, the embodiment of the present invention may obtain a plurality of regression models, and then, evaluate the prediction accuracy of the obtained regression models to select the regression model with the highest prediction accuracy, and further determine the optimal prediction value of the optimization target according to the regression model with the highest prediction accuracy. The evaluation process may comprise the steps of:
counting the number of normal groups of observation data;
evaluating the prediction accuracy of the regression models according to the normal group number of the observation data, the collection amount of the optimization target in each group of the observation data and the predicted value of the optimization target obtained through the observation data;
specifically, according to the embodiments of the present invention, the root Mean Square error rmse (root Mean Square error) and the decision coefficient R of each regression model may be calculated according to the normal number of groups of the observation data, the collection amount of the optimization target in each group of the observation data, and the predicted value of the optimization target obtained through the observation data2And evaluating the prediction accuracy of the regression models.
By performing substep S203-1 through substep S203-5, embodiments of the present invention may count the number of sets of normal observed data. Then substituting the normal group number of the observation data, the collection amount of the optimization target in each group of the observation data and the predicted value of the optimization target obtained by the observation data into formulas (1) to (4) for calculation, wherein:
in the above formula, ssr (sum of Squared residuals) represents the sum of Squared differences of the true value (the collected amount of the optimization target) and the predicted value of the optimization target; SST (Total Sum of squares) represents the Sum of the squared differences of the true value (the collection of the optimization objective) and the mean value (the mean of the collection of the optimization objective), and N represents the number of normal groups of the observed data.
Taking the 7 sets of data in fig. 4 as an example, if all the observed data are normal, the predicted values of the 7 sets of optimization targets can be obtained, and if the optimization target is the particle PN, the 7 sets of predicted values of the particle PN can be obtained. Referring to formula (3), the SSR is the sum of the predicted values of 7 sets of the optimization objective minus the squared difference of the acquisition amount of the optimization objective. Referring to equation (4), the sum of the collected amounts of 7 sets of the optimization targets is divided by 7 to obtain the mean value of the collected amounts of the optimization targets, and SST is the sum of the collected amounts of the 7 sets of the optimization targets minus the squared difference of the mean value of the optimization targets. The smaller the SSR is, the closer the predicted value and the actual value (the collection amount of the optimization target) of the optimization target are shown to be, the R is2The closer to 1.
Referring to FIG. 5, a schematic diagram of a model evaluation criterion of an embodiment of the present invention is shown. In FIG. 5, RMSE and R2For evaluating the prediction accuracy of the regression model. R2Has a value range of 0 to 1, R2The closer to 1, the higher the prediction precision of the regression model is, that is, the more the predicted value of the optimization target output by the regression model is closer to the acquisition amount of the optimization target. When R is2When the prediction accuracy is higher than or equal to 0.95, the prediction effect of the regression model is good, and the prediction accuracy is high; when R is2When the value is more than or equal to 0.7, the prediction effect of the regression model is good, but the ratio is higher than R2The prediction precision is lower when the prediction precision is more than or equal to 0.95; when R is2If < 0.7, the prediction effect of the regression model is general, indicating that the prediction accuracy is not high. The smaller the value of RMSE, the more accurate the prediction of the regression model isHigh. If R is selected2The regression model closest to 1 and with the smallest value of RMSE is the regression model with the highest prediction accuracy.
Step S204, determining an optimal predicted value from the predicted values of the optimization targets output by the regression models, and determining an optimal parameter value of the control parameter according to the optimal predicted value.
From the above, by establishing a plurality of regression models, the predicted values of a plurality of optimization targets can be output, and for how to determine the optimal predicted value among the predicted values of the optimization targets, the embodiment of the present invention can be determined by an empirical method, for example, the optimal predicted value is relative to the optimization target, evaluation criteria of the optimal predicted values of different optimization targets are different, for example, the amount of the particulate matter PN and the carbon monoxide CO is the lowest optimal predicted value, and for example, the Torque is the largest optimal predicted value. Or the prediction accuracy of the regression model can be determined, that is, the prediction value of the optimization target output by the regression model with the highest prediction accuracy is used as the optimal prediction value.
Then determining an optimal predicted value according to the predicted values of the optimization target, and determining observation data corresponding to the optimal predicted value; and then taking the parameter value of the control parameter in the observation data as the optimal parameter value of the control parameter.
For example, the optimization target is the fuel consumption rate, the lowest fuel consumption rate of the plurality of predicted values is used as the optimal predicted value, and according to the fact that the fuel consumption rate predicted according to the regression model is the lowest when the intake angle (control parameter) is-7 °, the intake angle of the set of observation data corresponding to the lowest fuel consumption rate is-7 °, which is also the optimal parameter value of the control parameter (intake angle).
In summary, it can be known that, in the embodiment of the present invention, each time 1 outlier (observation data) is removed, due to the change of the observation data, RMSE and R2Will vary and eventually a smaller RMSE and R will be obtained2A regression model close to 1 is preferred, and the present invention actually optimizes the regression model by optimizing the observed data. Verification of regression model in embodiments of the inventionThe method is interactive and synchronous with the training process, abnormal observation data can be eliminated quickly, the model training precision can be improved, the model training time can be shortened, and the optimization efficiency of the optimal parameter value of the control parameter is accelerated.
In order to further improve the prediction effect of the model, in an optional embodiment of the present invention, the following method is further provided, including:
acquiring the load and the rotating speed of the electric control gasoline engine through at least one real vehicle test;
acquiring a working condition distribution area under the ignition stage of the electric control gasoline engine according to the acquired load and the acquired rotating speed;
according to the distribution density of the working condition points in the working condition distribution area, counting the characteristic working condition points of the ignition stage of the electric control gasoline engine;
at this time, the implementation step of step S202 includes:
and according to the parameter values of a plurality of groups of control parameters obtained in advance, developing a bench sampling test at the characteristic working condition points, and determining a plurality of groups of observation data, wherein the observation data comprise the parameter values of each group of control parameters and the acquisition amount of the optimization target corresponding to the parameter values obtained through the bench sampling test.
During specific implementation, the rotating speed and the load of the electric control gasoline engine in the ignition stage are collected according to a preset sampling frequency or a preset time interval, the rotating speed of the electric control gasoline engine is used as an abscissa, the load percentage of the electric control gasoline engine is used as an ordinate, an area formed by the abscissa and the ordinate is divided into a plurality of sub-areas, and the characteristic working condition point is determined according to the point distribution density of a certain sub-area. Namely, a certain coordinate value with high point distribution density is selected as a characteristic working condition point. In other words, the embodiment of the invention can establish a corresponding regression model according to the characteristic working condition points to obtain the optimal parameter value of the control parameter under each characteristic working condition point, which has practical operation significance for optimizing the ignition stage of the vehicle.
And then, a bench sampling test is carried out at the characteristic working condition point, namely, under the specific rotating speed-load condition, the acquisition quantity change of the optimized target is obtained through the bench sampling test corresponding to the parameter values of different control parameters, at the moment, the model is modeled by using the observation data obtained at the characteristic working condition point without being influenced by the interference caused by the rotating speed-load change, and the model prediction effect is better. Referring to fig. 6, a distribution diagram of characteristic operating points of the embodiment of the present invention is shown. If the rotation speed is 1500rpm and the load is 55% as the characteristic working point, the rotation speed is 1850rpm and the load is 35% as the characteristic working point. Fig. 3 shows the control parameters and optimization targets determined at the characteristic operating point of 1500rpm and 55% load.
In an optional embodiment of the present invention, to verify the feasibility of the optimal parameter values obtained by the embodiment of the present invention, there is further provided a method comprising the steps of:
according to a preset boundary condition, the water temperature of the electric control gasoline engine is controlled through a rack, and the starting environment of the electric control gasoline engine in the ignition stage is simulated;
under the starting environment, inputting the optimal parameter value of the control parameter into a rack, and collecting the actual measurement quantity of the optimization target corresponding to the optimal parameter value through the rack;
and comparing the measured quantity of the optimization target with the acquisition quantity of the optimization target so as to verify the optimal parameter value of the control parameter.
The boundary conditions are the conditions of the water temperature of the electric control gasoline engine, such as the normal water temperature of a heat engine of the electric control gasoline engine is 90 ℃ and the normal temperature of a cold machine is 25 ℃.
For example, before optimization, the actual measurement of the particulate matter PN is 3^ e 7/cm under the conditions that the rotating speed of the electric control gasoline engine is 1500rpm and the torque is 50Nm under the specific working condition and the original control parameters (30 crank angle degrees at the opening time of an intake valve, 10 crank angle degrees below the opening time of an exhaust valve and 280 crank angle degrees at the oil injection cut-off time)3(ii) a Performing bench sampling test by covering 20 groups of different control parameters (an inlet valve range is minus 10-30 degrees of crank angle, an exhaust valve range is minus 20-20 degrees of crank angle, and an oil injection stop time is 40-360 degrees of crank angle), obtaining the collection amount of the corresponding particulate matter PN, and then obtaining the observation numberEstablishing and optimizing a regression model according to the data; based on the observation data of the optimal parameter values of the control parameters (10 degrees of an intake valve, 0 degree of an exhaust valve, 200 degrees of oil injection stop time and 1^ e7 particles/cm of particulate matter PN3) The actual measurement of the particulate matter PN actually measured by the bench is 1.2^ e7 particles/cm3It is clear that the measured amount and the collected amount of the particulate matter PN are 1^ e7 particles/cm3) The phase difference is not great, compared with that before optimization (3^ e7 pieces/cm)3) The reduction is obvious.
Based on the same inventive concept, an embodiment of the present invention provides a vehicle, and referring to fig. 7, a schematic structural diagram of the vehicle according to the embodiment of the present invention is shown, including an electronic control gasoline engine, where an optimal parameter value of a control parameter of the electronic control gasoline engine in a light-off phase is obtained by using the method for optimizing a light-off condition control parameter of the electronic control gasoline engine according to the embodiment of the present invention. Through the method of the embodiment of the invention, the optimal parameter value of the control parameter of the electric control gasoline engine in the ignition stage can be rapidly determined, and compared with the vehicle in the prior art, the vehicle of the embodiment of the invention can greatly reduce the emission of particulate matter PN in the ignition stage of the vehicle, obtain a better value of the performance index of the electric control gasoline engine, improve the driving experience of a user and is more environment-friendly.
It should be noted that, for simplicity of description, the method embodiments are described as a series of acts or combination of acts, but those skilled in the art will recognize that the present invention is not limited by the illustrated order of acts, as some steps may occur in other orders or concurrently in accordance with the embodiments of the present invention. Further, those skilled in the art will appreciate that the embodiments described in the specification are presently preferred and that no particular act is required to implement the invention.
Based on the same inventive concept, the embodiment of the invention also provides an ignition working condition control parameter optimization device of the electric control gasoline engine, and referring to fig. 8, a structural schematic diagram of the ignition working condition control parameter optimization device of the electric control gasoline engine is shown, and the device can comprise the following modules:
the test parameter determination module 801 is used for determining control parameters and optimization targets influenced by the control parameters in the ignition stage of the electric control gasoline engine;
a rack sampling test module 802, configured to perform a rack sampling test according to pre-obtained parameter values of multiple sets of the control parameters, and determine multiple sets of observation data, where the observation data includes the parameter value of each set of the control parameters and a collection amount of the optimization target corresponding to the parameter value obtained through the rack sampling test;
a regression model training module 803, configured to iteratively divide the multiple sets of observation data into a training set and a test set, where the training set is used to establish a regression model, and the test set is used to test the regression model, and obtain a predicted value of the optimization target output by the regression model in a test process;
an optimal parameter value determining module 804, configured to determine an optimal predicted value from the predicted values of the optimization targets output by the multiple regression models, and determine an optimal parameter value of the control parameter according to the optimal predicted value.
In an optional embodiment of the present invention, the apparatus further comprises:
the load and rotating speed acquisition module is used for acquiring the load and the rotating speed of the electric control gasoline engine through at least one real vehicle test;
the working condition distribution obtaining module is used for obtaining a working condition distribution area under the ignition stage of the electric control gasoline engine according to the acquired load and the acquired rotating speed;
the characteristic working condition point counting module is used for counting characteristic working condition points in the ignition stage of the electric control gasoline engine according to the distribution density of the working condition points in the working condition distribution area;
the bench sampling test module 802, comprising:
and the characteristic working condition point test submodule is used for developing a rack sampling test at the characteristic working condition points according to the pre-obtained parameter values of a plurality of groups of control parameters, and determining a plurality of groups of observation data, wherein the observation data comprise the parameter values of each group of control parameters and the acquisition quantity of the optimization target corresponding to the parameter values obtained through the rack sampling test.
In an optional embodiment of the invention, different electric control gasoline engines have different control parameters; wherein,
when the electric control gasoline engine is an air inlet injection PFI electric control gasoline engine, the control parameters comprise an air inlet variable valve timing VVT angle, an exhaust variable valve timing VVT angle, single or multiple fuel injection time and fuel proportion distribution corresponding to multiple injections;
when the electric control gasoline engine is an in-cylinder direct injection type GDI electric control gasoline engine, the control parameters comprise fuel rail pressure, fuel injection starting time SOI and fuel injection stopping time EOIT;
the optimization objectives affected by the control parameters include: particulate matter PN, fuel consumption FUELCOSP, combustion stability Cov, Torque, nitrogen oxides NOx, carbon monoxide CO, hydrocarbon HC.
In an optional embodiment of the present invention, the apparatus further comprises the following modules:
the parameter range setting module is used for setting the parameter range of the control parameter;
the parameter range processing module is used for preprocessing the parameter range of the control parameters to obtain a plurality of groups of parameter values of the control parameters;
the bench sampling test module 802 includes the following sub-modules:
and the acquisition quantity obtaining submodule is used for respectively inputting the parameter values of a plurality of groups of control parameters into the rack, acquiring the original emission of the electric control gasoline engine and/or testing the performance index of the electric control gasoline engine through the rack, and obtaining a plurality of groups of acquisition quantities of the optimization target corresponding to the parameter values.
In an optional embodiment of the present invention, the parameter range processing module includes:
the sampling interval setting submodule is used for setting a sampling interval according to the parameter range of the control parameter;
and the parameter value extraction submodule is used for extracting the parameter values of the control parameters according to the sampling interval in the parameter range of the control parameters to obtain a plurality of groups of parameter values of the control parameters.
In an optional embodiment of the present invention, the regression model training module 803 includes the following sub-modules:
the observation data grouping submodule is used for dividing the multiple groups of observation data into a training group and a testing group according to a preset distribution proportion, the training group is used for establishing the regression model, and the testing group is used for testing the regression model;
the model output submodule is used for outputting a predicted value of the optimization target by taking a parameter value of the control parameter in the observation data as the input of the regression model in the test process aiming at the observation data in the test group;
the observation data judgment submodule is used for extracting the acquisition amount of the optimization target corresponding to the parameter value of the control parameter in the observation data, and comparing the predicted value of the optimization target with the acquisition amount of the optimization target to determine whether the observation data is normal or not;
a first result execution sub-module, configured to retain a predicted value of the optimization target obtained through the observation data when the observation data is normal;
and the second result execution submodule is used for eliminating the observation data from the multiple groups of observation data when the observation data are abnormal, dividing the remaining multiple groups of observation data into a training group and a testing group, and repeating the steps.
In an optional embodiment of the present invention, the apparatus further comprises:
the observation data counting module is used for counting the number of groups of the normal observation data;
the regression model evaluation module is used for evaluating the prediction accuracy of the regression models according to the normal group number of the observation data, the acquisition amount of the optimization target in each group of the observation data and the predicted value of the optimization target obtained through the observation data;
the optimal parameter value determining module 804 includes:
the optimal predicted value determining submodule is used for taking the predicted value of the optimization target output by the regression model with the highest prediction precision as the optimal predicted value;
and the parameter value selection submodule is used for determining observation data corresponding to the optimal predicted value and taking the parameter value of the control parameter in the observation data as the optimal parameter value of the control parameter.
In an optional embodiment of the present invention, the apparatus further comprises:
the starting environment simulation module is used for controlling the water temperature of the electric control gasoline engine through a rack according to a preset boundary condition and simulating the starting environment of the electric control gasoline engine in a light-off stage;
the real measurement acquisition module is used for inputting the optimal parameter value of the control parameter into a rack under the starting environment and acquiring the real measurement of the optimization target corresponding to the optimal parameter value through the rack;
and the parameter value verification module is used for comparing the measured quantity of the optimization target with the acquisition quantity of the optimization target so as to verify the optimal parameter value of the control parameter.
In conclusion, according to the embodiment of the invention, the optimal parameter value of the control parameter in the ignition stage of the electric control gasoline engine can be rapidly determined, the obtained optimal parameter value has higher precision, the emission of particulate matter PN in the ignition stage of the vehicle can be greatly reduced, the more optimal value of the performance index of the electric control gasoline engine can be obtained, the test period is effectively shortened, the test cost is saved, and the method has important significance for improving the driving experience of users and protecting the environment.
It should be noted that, in the present specification, the embodiments are all described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments may be referred to each other.
It should also be noted that, in this document, the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the drawings, and are only for convenience in describing the present invention and simplifying the description, but do not indicate or imply that the referred device or element must have a specific orientation, be constructed and operated in a specific orientation, and thus, should not be construed as limiting the present invention. Moreover, relational terms such as "first" and "second" are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions or should not be construed as indicating or implying relative importance. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or terminal that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or terminal. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in the process, method, article, or terminal equipment comprising the element.
The technical solutions provided by the present application are described in detail above, and the principles and embodiments of the present application are described herein by using specific examples, which are only used to help understanding the present application, and the content of the present description should not be construed as limiting the present application. While various modifications of the illustrative embodiments and applications will be apparent to those skilled in the art based upon this disclosure, it is not necessary or necessary to exhaustively enumerate all embodiments, and all obvious variations and modifications can be resorted to, falling within the scope of the disclosure.
Claims (10)
1. An electric control gasoline engine ignition working condition control parameter optimization method is characterized by comprising the following steps:
determining control parameters and an optimization target influenced by the control parameters in a light-off stage of the electric control gasoline engine;
carrying out a bench sampling test according to the parameter values of a plurality of groups of control parameters obtained in advance, and determining a plurality of groups of observation data, wherein the observation data comprise the parameter values of each group of control parameters and the acquisition amount of the optimization target corresponding to the parameter values obtained through the bench sampling test;
iteratively dividing the multiple groups of observation data into a training group and a testing group, wherein the training group is used for establishing a regression model, the testing group is used for testing the regression model, and a predicted value of the optimization target output by the regression model is obtained in the testing process;
and determining an optimal predicted value from the predicted values of the optimization targets output by the regression models, and determining an optimal parameter value of the control parameter according to the optimal predicted value.
2. The method of claim 1, further comprising:
acquiring the load and the rotating speed of the electric control gasoline engine through at least one real vehicle test;
acquiring a working condition distribution area under the ignition stage of the electric control gasoline engine according to the acquired load and the acquired rotating speed;
according to the distribution density of the working condition points in the working condition distribution area, counting the characteristic working condition points of the ignition stage of the electric control gasoline engine;
according to the parameter values of a plurality of groups of control parameters obtained in advance, carrying out a bench sampling test, determining a plurality of groups of observation data, wherein the observation data comprise the parameter values of each group of control parameters and the collection quantity of the optimization target corresponding to the parameter values obtained through the bench sampling test, and the bench sampling test comprises the following steps:
and according to the parameter values of a plurality of groups of control parameters obtained in advance, developing a bench sampling test at the characteristic working condition points, and determining a plurality of groups of observation data, wherein the observation data comprise the parameter values of each group of control parameters and the acquisition amount of the optimization target corresponding to the parameter values obtained through the bench sampling test.
3. The method according to claim 1, characterized in that different electrically controlled gasoline engines have different control parameters; wherein,
when the electric control gasoline engine is an air inlet injection PFI electric control gasoline engine, the control parameters comprise an air inlet variable valve timing VVT angle, an exhaust variable valve timing VVT angle, single or multiple fuel injection time and fuel proportion distribution corresponding to multiple injections;
when the electric control gasoline engine is an in-cylinder direct injection type GDI electric control gasoline engine, the control parameters comprise fuel rail pressure, fuel injection starting time SOI and fuel injection stopping time EOIT;
the optimization objectives affected by the control parameters include: particulate matter PN, fuel consumption FUELCOSP, combustion stability Cov, Torque, nitrogen oxides NOx, carbon monoxide CO, hydrocarbon HC.
4. The method according to claim 1 or 2, characterized in that the method further comprises:
setting a parameter range of the control parameter;
preprocessing the parameter range of the control parameters to obtain a plurality of groups of parameter values of the control parameters;
according to the parameter values of a plurality of groups of control parameters obtained in advance, carrying out a bench sampling test, determining a plurality of groups of observation data, wherein the observation data comprise the parameter values of each group of control parameters and the collection quantity of the optimization target corresponding to the parameter values obtained through the bench sampling test, and the bench sampling test comprises the following steps:
and respectively inputting the parameter values of the plurality of groups of control parameters into a rack, and acquiring the original emission of the electric control gasoline engine and/or testing the performance index of the electric control gasoline engine through the rack to obtain a plurality of groups of acquisition quantities of the optimization target corresponding to the parameter values.
5. The method of claim 4, wherein preprocessing the parameter ranges of the control parameters to obtain a plurality of sets of parameter values of the control parameters comprises:
setting a sampling interval for a parameter range of the control parameter;
and in the parameter range of the control parameters, extracting the parameter values of the control parameters according to the sampling interval to obtain a plurality of groups of parameter values of the control parameters.
6. The method of claim 1, wherein iteratively dividing the plurality of sets of observation data into a training set and a testing set, the training set being used to build a regression model, the testing set being used to test the regression model, wherein obtaining predicted values of the optimization objective output by the regression model during testing comprises:
dividing the multiple groups of observation data into a training group and a testing group according to a preset distribution proportion, wherein the training group is used for establishing the regression model, and the testing group is used for testing the regression model;
aiming at observation data in a test group, in the test process, taking the parameter value of the control parameter in the observation data as the input of the regression model, and outputting to obtain the predicted value of the optimization target;
extracting the acquisition amount of the optimization target corresponding to the parameter value of the control parameter in the observation data, and comparing the predicted value of the optimization target with the acquisition amount of the optimization target to determine whether the observation data is normal or not;
if the observation data are normal, reserving a predicted value of the optimization target obtained through the observation data;
and if the observation data are abnormal, removing the observation data from the multiple groups of observation data, dividing the remaining multiple groups of observation data into a training group and a testing group, and repeating the steps.
7. The method of claim 6, further comprising:
counting the number of normal groups of observation data;
evaluating the prediction accuracy of the regression models according to the normal group number of the observation data, the collection amount of the optimization target in each group of the observation data and the predicted value of the optimization target obtained through the observation data;
determining an optimal predicted value from the predicted values of the optimization targets output by the regression models, and determining an optimal parameter value of the control parameter according to the optimal predicted value, wherein the determining comprises the following steps:
taking the predicted value of the optimization target output by the regression model with the highest prediction precision as an optimal predicted value;
and determining observation data corresponding to the optimal predicted value, and taking the parameter value of the control parameter in the observation data as the optimal parameter value of the control parameter.
8. The method of claim 1, further comprising:
according to a preset boundary condition, the water temperature of the electric control gasoline engine is controlled through a rack, and the starting environment of the electric control gasoline engine in the ignition stage is simulated;
under the starting environment, inputting the optimal parameter value of the control parameter into a rack, and collecting the actual measurement quantity of the optimization target corresponding to the optimal parameter value through the rack;
and comparing the measured quantity of the optimization target with the acquisition quantity of the optimization target so as to verify the optimal parameter value of the control parameter.
9. The utility model provides an automatically controlled gasoline engine fires operating mode control parameter optimizing apparatus which characterized in that includes:
the test parameter determining module is used for determining control parameters in the ignition stage of the electric control gasoline engine and an optimization target influenced by the control parameters;
the bench sampling test module is used for developing a bench sampling test according to the pre-obtained parameter values of a plurality of groups of control parameters, and determining a plurality of groups of observation data, wherein the observation data comprise the parameter values of each group of control parameters and the acquisition amount of the optimization target corresponding to the parameter values obtained through the bench sampling test;
the regression model training module is used for iteratively dividing the multiple groups of observation data into a training group and a testing group, wherein the training group is used for establishing a regression model, the testing group is used for testing the regression model, and a predicted value of the optimization target output by the regression model is obtained in the testing process;
and the optimal parameter value determining module is used for determining an optimal predicted value from the predicted values of the optimization targets output by the regression models, and determining the optimal parameter value of the control parameter according to the optimal predicted value.
10. A vehicle comprising an electric control gasoline engine, characterized in that the optimal parameter value of the control parameter of the electric control gasoline engine in the ignition stage is obtained by the method for optimizing the ignition working condition control parameter of the electric control gasoline engine according to any one of claims 1 to 8.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011013259.1A CN112282949B (en) | 2020-09-23 | 2020-09-23 | Method and device for optimizing control parameters of ignition working condition of electric control gasoline engine and vehicle |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011013259.1A CN112282949B (en) | 2020-09-23 | 2020-09-23 | Method and device for optimizing control parameters of ignition working condition of electric control gasoline engine and vehicle |
Publications (2)
Publication Number | Publication Date |
---|---|
CN112282949A CN112282949A (en) | 2021-01-29 |
CN112282949B true CN112282949B (en) | 2021-07-16 |
Family
ID=74421294
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202011013259.1A Active CN112282949B (en) | 2020-09-23 | 2020-09-23 | Method and device for optimizing control parameters of ignition working condition of electric control gasoline engine and vehicle |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112282949B (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113266449B (en) * | 2021-05-19 | 2022-04-26 | 潍柴动力股份有限公司 | Method and system for predicting air leakage situation in front of aftertreatment system |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP1703110A1 (en) * | 2005-03-18 | 2006-09-20 | Ford Global Technologies, LLC, A subsidary of Ford Motor Company | Method for optimising the calibration of an internal combustion engine |
CN101464964A (en) * | 2007-12-18 | 2009-06-24 | 同济大学 | Pattern recognition method capable of holding vectorial machine for equipment fault diagnosis |
CN104573195A (en) * | 2014-12-18 | 2015-04-29 | 东风康明斯发动机有限公司 | Single-point working condition optimization method for electronic control diesel engine |
CN107002576A (en) * | 2014-11-17 | 2017-08-01 | 大众汽车有限公司 | Control device for internal combustion engine |
CN107082051A (en) * | 2016-02-16 | 2017-08-22 | 罗伯特·博世有限公司 | System and method based on existing calibrated predicted calibration value |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US9506414B2 (en) * | 2013-10-01 | 2016-11-29 | GM Global Technology Operations LLC | Cold start emissions reduction diagnostic system for an internal combustion engine |
-
2020
- 2020-09-23 CN CN202011013259.1A patent/CN112282949B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP1703110A1 (en) * | 2005-03-18 | 2006-09-20 | Ford Global Technologies, LLC, A subsidary of Ford Motor Company | Method for optimising the calibration of an internal combustion engine |
CN101464964A (en) * | 2007-12-18 | 2009-06-24 | 同济大学 | Pattern recognition method capable of holding vectorial machine for equipment fault diagnosis |
CN107002576A (en) * | 2014-11-17 | 2017-08-01 | 大众汽车有限公司 | Control device for internal combustion engine |
CN104573195A (en) * | 2014-12-18 | 2015-04-29 | 东风康明斯发动机有限公司 | Single-point working condition optimization method for electronic control diesel engine |
CN107082051A (en) * | 2016-02-16 | 2017-08-22 | 罗伯特·博世有限公司 | System and method based on existing calibrated predicted calibration value |
Also Published As
Publication number | Publication date |
---|---|
CN112282949A (en) | 2021-01-29 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US9869254B2 (en) | Method for determining fuel blend in a dual fuel mixture | |
JP2011106334A (en) | Method of estimating heat release rate of engine using wiebe function model | |
US7672773B2 (en) | Transient engine performance adaptation method and system | |
CN113504050B (en) | Carbon deposition and coking test method and device for EGR (exhaust gas Recirculation) system | |
Ericson et al. | Transient emission predictions with quasi stationary models | |
de Nola et al. | Volumetric efficiency estimation based on neural networks to reduce the experimental effort in engine base calibration | |
CN114970404B (en) | Engine oil consumption calculation and optimization method based on in-cylinder combustion CFD analysis | |
CN112345259A (en) | Gasoline engine virtual calibration method based on knock self-recognition | |
CN112282949B (en) | Method and device for optimizing control parameters of ignition working condition of electric control gasoline engine and vehicle | |
CN107476903A (en) | A kind of supercharged diesel engine EGR Performance Evaluations and Optimum EGR rate determine method | |
CN105606367B (en) | A kind of engine steady operation catches fire detection and self adaptation decision method and device | |
Grahn et al. | A structure and calibration method for data-driven modeling of NO x and soot emissions from a diesel engine | |
CN102042105A (en) | Method for biodiesel blending detection based on relative air-to-fuel ratio estimation | |
Cavina et al. | Statistical analysis of knock intensity probability distribution and development of 0-d predictive knock model for a si tc engine | |
Chadry et al. | Embedded system using field programmable gate array (Fpga) myrio and labview programming to obtain data patern emission of car engine combustion categories | |
CN104612845A (en) | Dual-fuel engine knock detecting and controlling system and method based on ionic current | |
CN115828776A (en) | Rapid evaluation method, device, equipment and medium based on tumble and turbulence in gasoline engine cylinder | |
Gurel et al. | Multi-objective optimization of transient air-fuel ratio limitation of a diesel engine using DoE based Pareto-optimal approach | |
CN109139330A (en) | Engine with supercharger ignition control method and engine with supercharger Iganition control system | |
Yang et al. | Nonlinear dynamics of cycle-to-cycle variations in a lean-burn natural gas engine with a non-uniform pre-mixture | |
US10055523B2 (en) | Method for analyzing oxidation in an internal combustion engine | |
CN106321265A (en) | Method and system for identifying content of biodiesel in mixed fuel oil | |
Cameretti et al. | Virtual calibration method for diesel engine by software in the loop techniques | |
Polat et al. | An estimation of incylinder pressure based on lambda and engine speed in hcci engine using artificial neural networks | |
CN112053039A (en) | Engine oil dilution risk assessment method, device, equipment and storage medium |
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 |