CN110185791A - A kind of stepped automatic transmission Optimization about control parameter method - Google Patents
A kind of stepped automatic transmission Optimization about control parameter method Download PDFInfo
- Publication number
- CN110185791A CN110185791A CN201910460321.2A CN201910460321A CN110185791A CN 110185791 A CN110185791 A CN 110185791A CN 201910460321 A CN201910460321 A CN 201910460321A CN 110185791 A CN110185791 A CN 110185791A
- Authority
- CN
- China
- Prior art keywords
- parameter
- optimized
- optimization
- target response
- automatic transmission
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F16—ENGINEERING ELEMENTS AND UNITS; GENERAL MEASURES FOR PRODUCING AND MAINTAINING EFFECTIVE FUNCTIONING OF MACHINES OR INSTALLATIONS; THERMAL INSULATION IN GENERAL
- F16H—GEARING
- F16H61/00—Control functions within control units of change-speed- or reversing-gearings for conveying rotary motion ; Control of exclusively fluid gearing, friction gearing, gearings with endless flexible members or other particular types of gearing
- F16H61/04—Smoothing ratio shift
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/10—Geometric CAD
- G06F30/15—Vehicle, aircraft or watercraft design
-
- 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
-
- 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
- F16—ENGINEERING ELEMENTS AND UNITS; GENERAL MEASURES FOR PRODUCING AND MAINTAINING EFFECTIVE FUNCTIONING OF MACHINES OR INSTALLATIONS; THERMAL INSULATION IN GENERAL
- F16H—GEARING
- F16H61/00—Control functions within control units of change-speed- or reversing-gearings for conveying rotary motion ; Control of exclusively fluid gearing, friction gearing, gearings with endless flexible members or other particular types of gearing
- F16H2061/0075—Control functions within control units of change-speed- or reversing-gearings for conveying rotary motion ; Control of exclusively fluid gearing, friction gearing, gearings with endless flexible members or other particular types of gearing characterised by a particular control method
- F16H2061/009—Control functions within control units of change-speed- or reversing-gearings for conveying rotary motion ; Control of exclusively fluid gearing, friction gearing, gearings with endless flexible members or other particular types of gearing characterised by a particular control method using formulas or mathematic relations for calculating parameters
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/10—Numerical modelling
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Evolutionary Computation (AREA)
- Economics (AREA)
- Computer Hardware Design (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Aviation & Aerospace Engineering (AREA)
- Tourism & Hospitality (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Mechanical Engineering (AREA)
- Automation & Control Theory (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Computational Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Game Theory and Decision Science (AREA)
- Development Economics (AREA)
- Control Of Transmission Device (AREA)
Abstract
The invention discloses a kind of stepped automatic transmission Optimization about control parameter methods, comprising: Step 1: selecting parameter to be optimized and target response as optimization aim to shift gears comfort level；Wherein, the parameter to be optimized includes: speed, vehicle acceleration, engine speed, accelerator open degree, accelerator open degree change rate, vehicle mass and hydraulic control system main oil pressure；The target is corresponding are as follows: root mean square of weighed acceleration, maximum impact degree and maximum engine rotation speed variable quantity；Step 2: determining the value interval of the parameter to be optimized and the value interval of target response；Step 3: choosing parameter sample point to be optimized in the value interval of the parameter to be optimized, the corresponding target response value of each sample point is obtained；Step 4: establishing optimization model according to the parameter sample point to be optimized of selection and the corresponding target response value of each sample point；Step 5: carrying out parameter optimization according to the optimization model.The present invention can be improved the effect and optimization efficiency of parameter optimization.
Description
Technical field
The invention belongs to stepped automatic transmission control technology field, in particular to a kind of stepped automatic transmission control
Parameter optimization method processed.
Background technique
Stepped automatic transmission is developed on the basis of manual transmission, its electric-control system can be according to vehicle
The information such as speed, engine speed and accelerator open degree when driving automatically select suitable gear, to guarantee vehicle driving
Dynamic property, fuel economy and riding comfort.
When the control strategy exploitation of stepped automatic transmission electric-control system, need to carry out parameter optimization, i.e., according to needed for
The control target asked, is adjusted the parameter in control strategy, to realize the purpose of optimal control quality.The choosing of control parameter
Cut-off connects the characteristic for determining automatic transmission, has a great impact to the driving performance of vehicle.But current control parameter is excellent
Change the experience adjustments for all relying on engineer substantially and constantly test obtain, this method low efficiency and it is relatively difficult to achieve it is global most
It is excellent.And in the parameter optimisation procedure of stepped automatic transmission, usually in response with vehicle dynamic property or economic index
It optimizes, associated scaling method research is also more, but using comfort level of shifting gears as the Optimization about control parameter method of target
It studies less.
Summary of the invention
The present invention provides a kind of stepped automatic transmission Optimization about control parameter method, comfort level of shifting gears is optimization mesh
It is to establish parameter according to parameter to be optimized and target response that mark, which selects parameter and target response, an object of the present invention to be optimized,
Optimized model, and parameter optimization is carried out according to parameter optimization agent model, it can be improved the efficiency of parameter optimization.
The present invention provides a kind of stepped automatic transmission Optimization about control parameter methods, act on behalf of mould establishing parameter optimization
The confidence level of parameter optimization agent model is verified after type, and when confidence level is unsatisfactory for requiring, reselects optimization ginseng
Numerical example, using the validity chosen the second object of the present invention is to improve parameter sample, to improve the effect of optimization.
Technical solution provided by the invention are as follows:
A kind of stepped automatic transmission Optimization about control parameter method, comprising:
Step 1: selecting parameter to be optimized and target response；
Wherein, the parameter to be optimized includes: speed, vehicle acceleration, engine speed, accelerator open degree, accelerator open degree
Change rate, vehicle mass and hydraulic control system main oil pressure；And
The target response are as follows: root mean square of weighed acceleration, maximum impact degree and maximum engine rotation speed variable quantity；
Step 2: determining the value interval of the parameter to be optimized and the value interval of the target response；
Step 3: choosing parameter sample point to be optimized in the value interval of the parameter to be optimized, each sample point is obtained
Corresponding target response value；
Step 4: it is excellent to establish parameter according to the parameter sample point to be optimized of selection and the corresponding target response value of each sample point
Change model；
Step 5: carrying out parameter optimization according to the optimization model.
Preferably, in the step 2, after the value interval for determining the parameter to be optimized, according to following public affairs
Formula determines the value interval of target response:
Δωemax=[max (ωe)-min(ωe)]T；
In formula, awFor root mean square of weighed acceleration, jmaxFor maximum impact degree, Δ ωemaxFor maximum engine rotation speed change
Change amount；T is the shift process time, and a is vehicle acceleration, ωeFor engine speed, t is the time.
Preferably, in the step 3, the method for choosing parameter sample point to be optimized are as follows:
Step 1, by parameter x to be optimizediEquiprobability is divided into m subinterval, each subinterval note in its value interval
For
Step 2 carries out random sampling respectively in each subinterval, defines random number λ ∈ [0,1], then random sampling point
Are as follows:
Wherein, m is the number of samples extracted, and k is subinterval serial number.
It preferably,, will be described to be optimized using the parameter sample point to be optimized of selection as input in the step 4
The corresponding target response of parameter sample point establishes RBF neural network model, as the optimization model as output.
Preferably, before the step 4, further include simplified according to the related coefficient of the parameter to be optimized of selection to
The quantity of Optimal Parameters；
Wherein, the related coefficient of the parameter to be optimized are as follows:
Wherein,X and y are respectively indicated
Different parameters to be optimized；
Work as rxyWhen > 0.8, only retain a parameter to be optimized in x or y；
Work as rxyWhen < -0.8, retain the opposite number of x and x as parameter to be optimized, or retain y and y opposite number be used as to
Optimal Parameters.
It preferably, further include according to parameter to be optimized target corresponding with parameter to be optimized before the step 4
The related coefficient of response simplifies the quantity of parameter to be optimized；
Wherein, the related coefficient of the parameter to be optimized target response corresponding with parameter to be optimized are as follows:
Wherein,X indicates to be optimized
Parameter, z indicate the corresponding target response of parameter to be optimized；
Work as rxzWhen (- 0.1,0.1) ∈, corresponding parameter to be optimized is deleted.
It preferably, further include the prediction output valve and reality output of certificate parameter Optimized model in the step 4
Between the degree of correlation that is worth repeat step 3 when the degree of correlation is less than 0.9；
Wherein, the calculation method of the degree of correlation are as follows:
Wherein, ziFor real output value,For reality output average value,Output valve is predicted for optimization model.
Preferably, optimal for target with comfort level in the step 5, the control parameter to be optimized is carried out excellent
Change, includes the following steps:
Step 1, using binary coding mode, to speed v, vehicle acceleration a, engine speed ωe, accelerator open degree α,
Accelerator open degree change rateVehicle mass m and hydraulic control system main oil pressure p are encoded；
First generation population is randomly generated in step 2, and each of described first generation population individual includes v, a, ωe、α、M and p；Comfort level
The corresponding fitness of individual in step 3, calculating first generation population；Wherein, the fitness are as follows:
The individual is pressed ranking fitness by step 4, is selected the high individual of fitness, is intersected and made a variation, and generates the
Two generation populations；
The corresponding fitness of individual in step 5, calculating second generation population, circulation carries out step 4-5, until reaching setting
The number of iterations；The corresponding individual of maximum adaptation degree is selected as optimal solution；
Step 6 is decoded the optimal solution, obtains optimal v, a, ωe、α、The calibration value of m and p.
Preferably, in the step 5, the different parameters to be optimized is obtained according to the optimization model
Corresponding target response value, and adjustment is optimized to the parameter to be optimized according to the target response value.
The beneficial effects of the present invention are:
Stepped automatic transmission Optimization about control parameter method provided by the invention is selected using shifting gears comfort level as optimization aim
Parameter to be optimized and target response are selected, optimization model is established according to parameter to be optimized and target response, and excellent according to parameter
Change agent model and carry out parameter optimization, can be improved the efficiency of parameter optimization.
Stepped automatic transmission Optimization about control parameter method provided by the invention, after establishing parameter optimization agent model
The confidence level of parameter optimization agent model is verified, and when confidence level is unsatisfactory for requiring, reselects Optimal Parameters sample
This, until the confidence level of the parameter optimization agent model of foundation is met the requirements；The present invention can be improved having for parameter sample selection
Effect property, to improve the effect of optimization.
Detailed description of the invention
Fig. 1 is square to weighted acceleration for the accelerator open degree variation analyzed in the embodiment of the present invention 1 by main effect
The influence schematic diagram of root.
Fig. 2 is that the speed analyzed in the embodiment of the present invention 1 by main effect changes to root mean square of weighed acceleration
Influence schematic diagram.
Fig. 3 is the shadow changed by the accelerator open degree that main effect is analyzed in the embodiment of the present invention 1 to maximum impact degree
Ring schematic diagram.
Fig. 4 is that the influence that the speed analyzed in the embodiment of the present invention 1 by main effect changes to maximum impact degree is shown
It is intended to.
Fig. 5 is that the accelerator open degree analyzed in the embodiment of the present invention 1 by main effect changes to maximum engine rotation speed
The influence schematic diagram of variable quantity.
Fig. 6 is that the speed variation analyzed in the embodiment of the present invention 1 by main effect changes maximum engine rotation speed
The influence schematic diagram of amount.
Specific embodiment
Present invention will be described in further detail below with reference to the accompanying drawings, to enable those skilled in the art referring to specification text
Word can be implemented accordingly.
As shown in Figure 1, the present invention provides a kind of stepped automatic transmission Optimization about control parameter methods, including walk as follows
It is rapid:
Step 1: determining control parameter to be optimized and target response.
Stepped automatic transmission includes that electric control mechanical type automatic speed variator, hydraulic automatic speed variator and double clutches become automatically
Fast device, the control parameter in various automatic speed-variator electric control system control strategies is variant, in shift process, each speed changer
There are its respectively unique control parameter, such as control valve oil pressure, clutch principal and subordinate's Moving plate rotational speed difference etc..Different automatic transmission
Need that respectively structure feature and executing agency select its corresponding control parameter to be optimized according to its.
In the present invention using comfort level of shifting gears as optimization aim, the control parameter to be optimized of selection includes: speed v (km/h),
Vehicle acceleration a (m/s2), engine speed ω (rpm), accelerator open degree α (%), accelerator open degree change rateVehicle
(for car load when changing between unloaded and fully loaded, value directly affects vehicle inertia to quality m (kg), so that it is comfortable to change shift
Property) and hydraulic control system main oil pressure p (kpa).The shift of stepped automatic transmission of these parameters under various operating conditions is determined
In plan and shift implementation procedure, the ride comfort in shift process can be directly affected, to influence comfort level.
The present invention chooses root mean square of weighed acceleration aw, maximum impact degree jmaxWith maximum engine rotation speed variation delta
ωemaxAs target response.Using shifting comfort as target, the above response is the response on vehicle heading, i.e., longitudinally
Response.Under identical driving cycle, root mean square of weighed acceleration aw, maximum impact degree jmaxWith maximum engine rotation speed variable quantity
ΔωemaxSmaller, shift comfort level is better.Wherein, the expression formula of each target response are as follows:
T is the shift process time, is had in shift process time T:
Δωemax=[max (ωe)-min(ωe)]T (3)。
Step 2: primarily determining control parameter value interval to be optimized and response section.
In the control strategy exploitation of stepped automatic transmission electric-control system and optimization process, to reduce parameter value model
It encloses, convenient for efficiently carrying out parameter optimization, need to primarily determine control parameter value interval to be optimized and response section.For to excellent
The value interval for changing control parameter, can not be calculated by formula, can be according to the specified of each control parameter respective execution mechanisms
Variation range primarily determines value interval.The control parameter value interval to be optimized primarily determined is substituted into the expression of target response
Formula (1), (2) and (3), can primarily determine target response section.
Step 3: obtaining sample point by the methods of sampling.
The control parameter to be optimized of various stepped automatic transmissions is all more, and each parameter has its respective value model
It encloses, therefore experiment parameter dimension is high, valued space is big, sample point quantity is more.The present embodiment is sampled using Latin hypercube
Design obtains in value interval and is evenly distributed, full of the sample point for testing valued space and quantity is greatly reduced, reduces test
Guarantee the representativeness of sample point while number.
For n control parameters to be optimized, need to obtain m sample point, for each control parameter x to be optimizediAt it
Equiprobability is divided into the selection probability in each subinterval in m subinterval and is in valued spaceEach subinterval is denoted asCarry out independent sampling respectively in each subinterval, when independent sampling is randomly choosed, and is defined
Random number λ ∈ [0,1], then random sampling point are as follows:
After generating sample, by each parameter sample representation to be optimizedThe drawing to form that sample number is m is arranged according to random number
Fourth hypercube Sampling, as m × n LHD.
Step 4: obtaining target response according to selected sample point.
The sample point of selection is brought into the expression formula (1), (2) and (3) for substituting into target response, each sample point institute is calculated
Corresponding target response.
Step 5: carrying out correlation analysis, control parameter quantity to be optimized is simplified.
In the parameter optimization of stepped automatic transmission electric-control system, there are many control parameters to be optimized, some of them
The similarity between influence is smaller or some parameters of parameters on target response is very high, such control parameter without optimization,
It can directly be indicated with constant or similar control parameter.
In the present embodiment, using correlation coefficient rxyCorrelation between analysis and Control parameter to be optimized:
Wherein, sxyFor the covariance of sample x and sample y,sxFor the standard deviation of sample x,
Correlation coefficient rxyValue interval between 1 and -1,1 indicate perfect positive correlation, -1 indicate perfect negative correlation, 0 table
Show completely uncorrelated.Between parameter to be optimized, correlation coefficient rxyParameter greater than 0.8 (or being less than -0.8) can be joined with its correlation
(or opposite number of relevant parameter) is counted to indicate, reduces number of parameters to be optimized.
It in another embodiment, further include using correlation coefficient rxzAnalysis and Control parameter to be optimized is corresponding with parameter to be optimized
Target response between correlation:
Wherein, sxyFor the covariance of sample x and target response z X indicates parameter to be optimized, and z indicates that the corresponding target of parameter to be optimized is rung
It answers；
Correlation coefficient r between parameter to be optimized is corresponding to targetxzWhen (- 0.1,0.1) ∈, by corresponding parameter to be optimized
It deletes, to reduce number of parameters to be optimized.
Step 6: training forms optimization agent model.
Using the sample point of acquisition as input, the corresponding response of each sample point is as output, construction RBF neural mould
Type forms parameter optimization agent model.Model includes input layer, single layer hidden layer and output layer, is inputted as all parameters to be optimized,
Hidden neuron number is identical as sample points, exports as 3 target responses, and RBF neural network model is based on gaussian kernel function
Building, i.e.,
Preferably, further including the confidence level for verifying agent model after establishing optimization agent model.Parameter optimization acts on behalf of mould
Degree of correlation between the prediction output of type and reality output can characterize the confidence level of the calibration agent model of training formation, related journey
The calculating formula of degree isWherein, ziBy testing obtained real output value,To pass through examination
The reality output average value tested,Output valve is predicted for optimization agent model.Related journey R2Closer to 1, agent model is indicated
Prediction output confidence level it is higher, if degree of correlation R2Less than 0.9, then obtains new sample point and re-establish optimization and act on behalf of mould
Type, until degree of correlation R2Until meeting the requirements.Foundation optimization agent model can be treated excellent using the optimization agent model
Change parameter to optimize.
Embodiment 1
After establishing optimization agent model, sensitivity analysis is carried out to each parameter to be optimized using parameter optimization agent model.
In the present embodiment, it is analyzed using the method that single-factor main effect is analyzed, that is, it, should when analyzing a certain Parameters variation to be optimized
The influence that all combined situations that parameter to be optimized and other parameters generate generate target response, so that engineer is selected,
Adjustment and optimal control parameter provide guidance.The present embodiment is carried out by taking control parameter accelerator open degree α to be optimized and speed v as an example
The result of main effect analysis is as shown in figs. 1 to 6.
Fig. 1 and Fig. 2 is respectively the root mean square of weighed acceleration a that main effect is analyzedwTo accelerator open degree α and speed v
Response, as seen from the figure: in the value range of diagram, accelerator open degree α is bigger, root mean square of weighed acceleration awIt is smaller, then
It takes accelerator open degree α to be greater than 8 direction in calibration, root mean square of weighed acceleration a can be madewReduce, i.e., comfort level is got higher；Scheming
In the value range shown, speed v obtains root mean square of weighed acceleration a at one at 4.2wLocal minimum, and it is being greater than 8
Successively decrease on direction, then in calibration, desirable speed v is 4.2 and is greater than 8 directions, can make root mean square of weighed acceleration awSubtract
Small, i.e., comfort level is got higher.
Fig. 3 and Fig. 4 is respectively the maximum impact degree j that main effect is analyzedmaxResponse to accelerator open degree α and speed v,
In order to improve comfort level, the direction for taking accelerator open degree α to be greater than 20 in calibration, speed v takes less than 10 or is greater than 20 directions.
Fig. 5 and Fig. 6 is respectively the maximum engine rotation speed variation delta ω that main effect is analyzedemaxTo accelerator open degree α
With the response of speed v, in order to improve comfort level, the direction that when calibration takes accelerator open degree α to be greater than 20, speed v takes the side less than 10
To.
Embodiment 2
Using the parameter optimization agent model of building, Optimal Parameters are treated using Analysis of Genetic Algorithms, can acquire the overall situation most
Small value, i.e. maximal comfort in shift process.Steps are as follows:
(1) first rule of thumb, Population Size is set as 50, and setting maximum algebra, that is, Optimized Iterative time is 1000, setting
Crossing-over rate is 1 approximate optimal solution of sufficiently evolving to obtain to guarantee population, sets aberration rate as 0.1 because of Optimal Parameters hair in this patent
A possibility that different of changing, is smaller；
(2) by the n of each group of solution control parameter v, a, ω to be optimizede、α、It is compiled after m and p normalization by binary system
Code, obtains the coding of group solution in feasible zone, as chromosome；
(3) initial population is created at random；
(4) objective function, i.e. function of the comfort level about each parameter to be optimized are set, if comfort level is C, is then hadThe more big then comfort level of functional value is higher, then having objective function is max (C), i.e., comfort level refers to
Mark is maximum；
(5) fitness of judgement individual, fitness refer to the target response that each individual obtains and finally obtained best sound
The departure degree answered, departure degree is bigger, and it is poorer to respond；Departure degree is smaller, and it is better to respond, without departing from as optimal solution.
Fitness formulaIt characterizes, i.e. the bigger closer optimal solution of F, individual corresponding to maximum F
As optimal solution.
(6) it is ranked up according to the fitness of individual, the individual for selecting fitness high is as parent, the low individual of fitness
It is eliminated, is intersected with the chromosome of parent according to certain method, generate filial generation and make a variation to child chromosome.
(7) population of new generation is generated by intersecting and making a variation, the individual adaptation degree of population of new generation is calculated, until reaching iteration
Number, the corresponding comfort level of the maximum individual of fitness are max (C).
(8) it regard the maximum individual of fitness as optimal solution, optimal solution is decoded and renormalization, obtain optimization ginseng
Number.
Optimal Parameters are treated with this algorithm and are combined optimization, and the optimal shift control of synthesis under optimum reelability quality can be obtained
Parameter processed.
Although the embodiments of the present invention have been disclosed as above, but its is not only in the description and the implementation listed
With it can be fully applied to various fields suitable for the present invention, for those skilled in the art, can be easily
Realize other modification, therefore without departing from the general concept defined in the claims and the equivalent scope, the present invention is simultaneously unlimited
In specific details and legend shown and described herein.
Claims (9)
1. a kind of stepped automatic transmission Optimization about control parameter method, which comprises the steps of:
Step 1: selecting parameter to be optimized and target response；
Wherein, the parameter to be optimized includes: speed, vehicle acceleration, engine speed, accelerator open degree, accelerator open degree variation
Rate, vehicle mass and hydraulic control system main oil pressure；And
The target response are as follows: root mean square of weighed acceleration, maximum impact degree and maximum engine rotation speed variable quantity；
Step 2: determining the value interval of the parameter to be optimized and the value interval of the target response；
Step 3: choosing parameter sample point to be optimized in the value interval of the parameter to be optimized, it is corresponding to obtain each sample point
Target response value；
Step 4: establishing parameter optimization mould according to the parameter sample point to be optimized of selection and the corresponding target response value of each sample point
Type；
Step 5: carrying out parameter optimization according to the optimization model.
2. stepped automatic transmission Optimization about control parameter method according to claim 1, which is characterized in that in the step
In rapid two, after the value interval for determining the parameter to be optimized, the value interval of target response is determined according to the following formula:
Δωemax=[max (ωe)-min(ωe)]T；
In formula, awFor root mean square of weighed acceleration, jmaxFor maximum impact degree, Δ ωemaxFor maximum engine rotation speed variable quantity；
T is the shift process time, and a is vehicle acceleration, ωeFor engine speed, t is the time.
3. stepped automatic transmission Optimization about control parameter method according to claim 2, which is characterized in that in the step
In rapid three, the method for choosing parameter sample point to be optimized are as follows:
Step 1, by parameter x to be optimizediEquiprobability is divided into m subinterval in its value interval, and each subinterval is denoted as
Step 2 carries out random sampling respectively in each subinterval, defines random number λ ∈ [0,1], then random sampling point are as follows:
Wherein, m is the number of samples extracted, and k is subinterval serial number.
4. stepped automatic transmission Optimization about control parameter method according to claim 3, which is characterized in that in the step
In rapid four, using the parameter sample point to be optimized of selection as input, by the corresponding target response of the parameter sample point to be optimized
As output, RBF neural network model is established, as the optimization model.
5. stepped automatic transmission Optimization about control parameter method according to claim 3 or 4, which is characterized in that in institute
It further include the quantity that parameter to be optimized is simplified according to the related coefficient of the parameter to be optimized of selection before stating step 4；
Wherein, the related coefficient of the parameter to be optimized are as follows:
Wherein,X and y respectively indicate difference
Parameter to be optimized；
Work as rxyWhen > 0.8, only retain a parameter to be optimized in x or y；
Work as rxyWhen < -0.8, retain the opposite number of x and x as parameter to be optimized, or retain the opposite number of y and y as to be optimized
Parameter.
6. stepped automatic transmission Optimization about control parameter method according to claim 5, which is characterized in that in the step
Before rapid four, further include simplified according to the related coefficient of parameter to be optimized target response value corresponding with parameter to be optimized it is to be optimized
The quantity of parameter；
Wherein, the related coefficient of the parameter to be optimized target response corresponding with parameter to be optimized are as follows:
Wherein,X indicates parameter to be optimized, z
Indicate the corresponding target response of parameter to be optimized；
Work as rxzWhen (- 0.1,0.1) ∈, corresponding parameter to be optimized is deleted.
7. stepped automatic transmission Optimization about control parameter method according to claim 6, which is characterized in that in the step
It further include the degree of correlation being worth between the prediction output valve of certificate parameter Optimized model and reality output, when the correlation in rapid four
When degree is less than 0.9, repeat step 3；
Wherein, the calculation method of the degree of correlation are as follows:
Wherein, ziFor real output value,For reality output average value,Output valve is predicted for optimization model.
8. stepped automatic transmission Optimization about control parameter method according to claim 7, which is characterized in that in the step
It is optimal for target with comfort level in rapid five, the control parameter to be optimized is optimized, is included the following steps:
Step 1, using binary coding mode, to speed v, vehicle acceleration a, engine speed ωe, accelerator open degree α, throttle
Aperture change rateVehicle mass m and hydraulic control system main oil pressure p are encoded；
First generation population is randomly generated in step 2, and each of described first generation population individual includes v, a, ωe、α、M and
p；Comfort level
The corresponding fitness of individual in step 3, calculating first generation population；Wherein, the fitness are as follows:
The individual is pressed ranking fitness by step 4, and the individual for selecting fitness high is intersected and made a variation, and the second generation is generated
Population；
The corresponding fitness of individual in step 5, calculating second generation population, circulation carry out step 4-5, until reaching changing for setting
Generation number；The corresponding individual of maximum adaptation degree is selected as optimal solution；
Step 6 is decoded the optimal solution, obtains optimal v, a, ωe、α、The calibration value of m and p.
9. stepped automatic transmission Optimization about control parameter method according to claim 8, which is characterized in that in the step
In rapid five, according to the optimization model obtain the different parameters to be optimized corresponding to target response value, and root
Adjustment is optimized to the parameter to be optimized according to the target response value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910460321.2A CN110185791B (en) | 2019-05-30 | 2019-05-30 | Control parameter optimization method for stepped automatic transmission |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910460321.2A CN110185791B (en) | 2019-05-30 | 2019-05-30 | Control parameter optimization method for stepped automatic transmission |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110185791A true CN110185791A (en) | 2019-08-30 |
| CN110185791B CN110185791B (en) | 2020-03-20 |
Family
ID=67718746
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910460321.2A Active CN110185791B (en) | 2019-05-30 | 2019-05-30 | Control parameter optimization method for stepped automatic transmission |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110185791B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114321360A (en) * | 2022-01-04 | 2022-04-12 | 一汽解放汽车有限公司 | Gear identification method and device for manual transmission gearbox and computer equipment |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101118620A (en) * | 2007-09-18 | 2008-02-06 | 吉林大学 | Vehicle gear shifting quality evaluation method based on nerval net |
| CN108087541A (en) * | 2017-01-19 | 2018-05-29 | 西华大学 | The optimal Shift Strategy system of the more performance synthesises of automobile geared automatic transmission |
| JP2018135933A (en) * | 2017-02-21 | 2018-08-30 | トヨタ自動車株式会社 | Control device of vehicle |
-
2019
- 2019-05-30 CN CN201910460321.2A patent/CN110185791B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101118620A (en) * | 2007-09-18 | 2008-02-06 | 吉林大学 | Vehicle gear shifting quality evaluation method based on nerval net |
| CN108087541A (en) * | 2017-01-19 | 2018-05-29 | 西华大学 | The optimal Shift Strategy system of the more performance synthesises of automobile geared automatic transmission |
| JP2018135933A (en) * | 2017-02-21 | 2018-08-30 | トヨタ自動車株式会社 | Control device of vehicle |
Non-Patent Citations (2)
| Title |
|---|
| 张建国等: "换挡品质评价在AMT轿车开发中的应用", 《2010中国汽车工程学会年会论文集》 * |
| 陈宁等: "工程车辆自动变速挡位决策的遗传径向基神经网络方法", 《吉林大学学报（工学版）》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114321360A (en) * | 2022-01-04 | 2022-04-12 | 一汽解放汽车有限公司 | Gear identification method and device for manual transmission gearbox and computer equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110185791B (en) | 2020-03-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106021961B (en) | City standard state of cyclic operation construction method based on genetic algorithm optimization | |
| CN112101489A (en) | Equipment fault diagnosis method driven by united learning and deep learning fusion | |
| CN112429005B (en) | Pure electric vehicle personalized gear shifting rule optimization method considering transmission efficiency and application | |
| CN109747654B (en) | A kind of hybrid vehicle control parameter scaling method towards operating condition | |
| EP0969385A2 (en) | Evaluation method for a hereditary algorithm | |
| CN105488297A (en) | Method for establishing complex product optimization design agent model based on small sample | |
| CN112765902B (en) | Soft measurement modeling method for COD concentration in rural domestic sewage treatment process based on TentFWA-GD RBF neural network | |
| CN112614552B (en) | BP neural network-based soil heavy metal content prediction method and system | |
| CN113324026A (en) | Automatic gear shifting control method based on fuzzy neural network | |
| CN110185791A (en) | A kind of stepped automatic transmission Optimization about control parameter method | |
| CN116011110A (en) | Automatic driving automobile safety grade assessment method and system | |
| CN114004153A (en) | Penetration depth prediction method based on multi-source data fusion | |
| CN109408896B (en) | Multi-element intelligent real-time monitoring method for anaerobic sewage treatment gas production | |
| CN110007598A (en) | A kind of pre- scaling method of automatic transmission control parameter based on agent model | |
| CN113762602A (en) | Short-term wind speed prediction method for wind power plant | |
| CN113049259A (en) | Fuzzy control method of rack control system, storage medium and equipment | |
| Triase et al. | Implementation of Electre Method in Determining Tourism Places in North Sumatera | |
| CN116010291A (en) | Multipath coverage test method based on equalization optimization theory and gray prediction model | |
| Zhong et al. | An implementation of genetic-based learning classifier system on a wet clutch system | |
| CN113469370B (en) | Industrial Internet of things data sharing method based on federal incremental learning | |
| CN109871953B (en) | Wavelet neural network modeling method for heavy oil cracking process of fpRNA genetic algorithm | |
| JP2000339005A (en) | Method and device for controlling optimization of controlled target | |
| CN113470732A (en) | Microbial metabolism network model multi-optimization target determination method and application thereof | |
| Yin et al. | Multi-performance optimal gearshift schedule of stepped automatic transmissions adaptive to road slope | |
| CN112686366A (en) | Bearing fault diagnosis method based on random search and convolutional neural network |
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 |