CN102337979A - Automatic calibration parameter optimization method of engine based on genetic algorithm - Google Patents
Automatic calibration parameter optimization method of engine based on genetic algorithm Download PDFInfo
- Publication number
- CN102337979A CN102337979A CN2011102301122A CN201110230112A CN102337979A CN 102337979 A CN102337979 A CN 102337979A CN 2011102301122 A CN2011102301122 A CN 2011102301122A CN 201110230112 A CN201110230112 A CN 201110230112A CN 102337979 A CN102337979 A CN 102337979A
- Authority
- CN
- China
- Prior art keywords
- optimization
- population
- genetic algorithm
- parameter
- map
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Abstract
The invention relates to an automatic calibration parameter optimization method of an engine based on genetic algorithm. The method comprises the following steps: introducing a genetic algorithm based on the original calibration technology; and establishing the corresponding original population, the fitness evaluation formula, the crossover rate, the mutation rate and the terminal conditions for each parameter by utilizing the genetic algorithm when the engine control parameters are optimized, and carrying out genetic algorithm optimization operation, thus the optimal gene is obtained as the control parameter in an MAP. The method has the beneficial effects that the method does not rely on any subjective factor, and has high objectivity, as long as engineers provide boundary conditions of the calibration objects and the final optimizing target, the optimization configuration of the selected parameters can be carried out automatically through the method; simultaneously the method has very high efficiency, multiple iterations can be carried out in very short time, thus the optimal solution can be obtained rapidly; and the method has high spreadability, and can spread the whole constrained range, thus the accuracy of optimization is improved.
Description
Technical field
The present invention relates to the Engine ECU Control Parameter and demarcate the technology of optimization automatically, the automatic calibrating parameters optimization method of especially a kind of motor based on genetic algorithm.
Background technique
The ECU Control Parameter is demarcated: according to the various performance requirements (like power character, Economy and discharging etc.) of car load, and the process of Operational Limits of adjustment, optimization and definite automatical control system (like rotating speed, air mass flow, throttle position, coolant water temperature etc.) and Control Parameter (like fuel injection pulsewidth, degrees of ignition advance, EGR valve opening etc.).Optimize in the rating test mainly to as if basic controlling output quantity MAP.MAP is the foundation that control output quantity (fuel injection pulsewidth, ignition advance angle etc.) is calculated; It has stored the basic output quantity size under different engine loads, the rotating speed; ECU utilizes interpolation inquiry MAP figure, through exporting control signal after the parameter correction according to engine operation condition.Can the meshed network of being made up of operating mode features such as load, rotating speeds among the MAP reflect each operating mode of motor comprehensively, and the accuracy of node upper domination number certificate all directly influences the performance of motor.
The demarcation of ECU is a very loaded down with trivial details job.The engineer need carry out Parameter Optimization to each operating point of motor.Suppose a Control Parameter with 50 operating points, the engineer could finally confirm its optimum value for the optimization repeatedly that each operating point need carry out repeatedly, and while optimization each time all needs great amount of time.This makes the engineer require a great deal of time, carry out a large amount of repetitive operations after, could accomplish demarcation to MAP figure in the motor.And often contain the MAP figure that much varies in size in the motor, and this just makes the cycle of whole calibrating work become very long, can cause working strength big simultaneously, cost drops into high.These factors have all limited the development that ECU demarcates.
A kind of method of typical ECU Control Parameter is following:
1. need to confirm the parameter or the MAP that demarcate to scheme by demarcating the engineer.These parameters that need demarcate can be confirmed according to the object of its required optimization; Because the factor of ECU control strategy; The engineer possibly choose a plurality of Control Parameter to reach the purpose that object is optimized, and demarcate the engineer and need understand the qualitative relationships between each Control Parameter and the optimised object this moment, for example when optimization objects is engine emission; The demarcation object of then choosing should be its factor of influence: ignition advance angle; Fuel injection pulsewidth etc., the engineer need understand under different working conditions simultaneously, and ignition advance angle and fuel injection pulsewidth are to the influence of discharging.
2. demarcate the engineer according to selected calibrating parameters in the step 1, MAP figure and optimization objects are carried out confirming of operating point.After reasonably having chosen operating point, through the numerical value of continuous adjustment parameter, make optimization objects reach optimization aim after, note the parameter current value, be written among the corresponding operating point MAP.Constantly repeat this operation, all operating points all are optimized.
3. demarcate the engineer according to the MAP figure that obtains in the step 2, carry out curve fitting.If can't fit to response curve, then utilize interpolation to carry out completion a little, to realize the continuous variation of parameter.
In above-mentioned method, demarcate the engineer and need carry out a large amount of surveying works and continuous proving operation, could get optimized parameter to the end.These work will spend great amount of time and drop into a large amount of costs.In addition, the definite of operating point determined by concrete demarcation demand.It is many more that operating point is chosen, and then the granularity of MAP figure is high more, and the validity of control is high more, and whole control process is then more near perfect condition, but operating point is many more, and then the required operation of calibration process is many more, the cycle of demarcation, and cost, loaded down with trivial details degree is high more.Otherwise, reduce operating point, then can reduce the cycle of demarcation, cost and loaded down with trivial details degree, granularity reduces but this can cause MAP figure, and the validity of control reduces, and it is far away more that whole control process and perfect condition depart from, even wrong Control Parameter occurs.This has just produced a contradiction, should guarantee that the granularity of MAP figure is high, and control accuracy is high, also will let the cycle of calibration process short simultaneously, and cost is low, and loaded down with trivial details degree is low, and in this time, this traditional method will produce significant limitation.
The topmost method that present stage overcomes the above problems is the experience that relies on constantly accumulation and engineer to demarcate.Through accumulation, the scope that the engineer can the constrained optimum parameter occurs, thus reduce the number of times of demarcating, improve the efficient of demarcation with this, but this method has very big binding character:
1. accumulation needs process, and this causes this method to be difficult for popularizing, and starting is difficult simultaneously.
2. with strong points.When target engine changed, the experience of accumulation may be with going its corresponding effect.
3. deficiency in objective property.Because all based on demarcating the engineer to the subjective observation of optimization aim and based on the parameter optimization of experience, this will inevitably cause Parameter Optimization to have the subjective factor of demarcating the engineer to whole calibrating procedure, thereby makes the result of demarcation lack certain objectivity.
Automatic calibrating parameters optimization method of the present invention is based upon on the basis of genetic algorithm; Genetic algorithm is the computation model of biological evolution process of natural selection and the genetics mechanism of simulation Darwin theory of biological evolution; Be a kind of optimized Algorithm that is based upon on the random device, it utilizes the primitive solution of random algorithm generation some, and is biological for getting iterative operation through each; These primitive solutions are intersected; Breeding and mutation operation eliminate ropy individuality then, make last separating slowly approach optimum separating.
In the process of using genetic algorithm, most important is exactly that the data that each stage produces are assessed, and the assessment of this part realizes through valuation functions of definition usually.For the calibration process that the front is described, its valuation functions is fairly obvious, promptly demarcates the required optimization aim that reaches.For an objective function, the value of in its boundary conditions, regulating its factor of influence makes itself and objective function optimization, and then its fitness (the optimization degree of Control Parameter just) is just high more.
Summary of the invention
The object of the invention will solve the deficiency that above-mentioned technology exists just, and provides a kind of motor based on genetic algorithm automatic calibrating parameters optimization method, is a kind of optimisation technique that can improve demarcation efficient significantly, improve the demarcation accuracy.
The objective of the invention is to obtain a high efficiency, highi degree of accuracy, adaptable automatic calibration method: this method is through given boundary conditions, and valuation functions and optimization aim are carried out parameter optimization automatically, thereby obtains the optimal control parameter under precondition.This method does not rely on any subjective factor, has the objectivity of height, and the engineer only need provide the boundary conditions of demarcation object and last optimization aim, just can carry out selected Parameter Optimization coupling automatically through this method.Simultaneously, because the essential concurrency of genetic algorithm makes this method have very high efficient, can carry out repeatedly iteration in the short period of time, thereby obtain optimal solution fast.And because the scope spreadability of genetic algorithm is carried out the while iteration with a plurality of points in the scope, therefore have the spreadability of height, can cover whole restriction range, thereby improve the accuracy of optimizing.
The present invention solves the technological scheme that its technical problem adopts: the automatic calibrating parameters optimization method of this motor of the present invention based on genetic algorithm, on the basis of original calibration technique, introduce genetic algorithm; When carrying out the optimization of control parameters of engine; Utilize genetic algorithm to set up its pairing initial population for each parameter, fitness is passed judgment on formula, crossing-over rate; Aberration rate and end condition; Carry out the genetic algorithm optimization operation, finally obtain optimal base because of the Control Parameter in scheming as MAP, its concrete steps are:
1) confirms the optimization objects (Optimizing Target is hereinafter to be referred as OT) of demarcation task.According to OT, need to confirm the MAP that demarcates to scheme and parameter, include 1-2 reference parameter among this MAP figure, be used for confirming of operating point, include at least 1 Control Parameter simultaneously, in order to demarcate optimization; Then each unit task Cell Task among the MAP figure is carried out genetic algorithm optimization;
2) according to the required control accuracy that reaches of OT, we specify its granularity, the distribution of design conditions point for MAP figure.Operating point generally evenly to be distributed as the master, if specific demand is arranged, then carries out local correction based on demand.To be divided into the operating point be a plurality of little task (Cell Task) of standard for whole M AP figure the most at last, and each Cell Task need carry out primary parameter optimization, and whole M AP figure is evolved to optimum state.
3) be the initialization factor that each Cell Task specifies optimized Algorithm, be GA=(M, F, c, m, p
c, p
m), wherein M is a group size, and F is an ideal adaptation degree evaluation function, and c is the interlace operation operator, and m is the mutation operation operator, p
cBe crossover probability, p
mBe the variation probability;
4) according to the specific requirement precision, for optimization objects is encoded, the span with optimization aim of confirming as of the formula of encoding is decided to be [a
i, b
i], given its required precision is δ, calculates its binary-coded length
(λ is an integer).Then, utilize method at random, from a
iTo b
iThe middle initial population S that produces
1, the size of population is decided to be M according to demand;
5) genetic optimization of entering genetic algorithm; Wherein select successively; Intersect; The three kinds of modes of inheritance that make a variation, and this process that constantly circulates obtains the optimized parameter of optimum value for this Cell Task at last up to reaching end condition (end condition can for obtaining optimum value or stopping genetic algebra for reaching).
6) the Cell Task to other repeats above-mentioned operation, finally accomplishes the parameter optimization of whole M AP figure.
Following as preferred step:
1) with the value of feedback of OT (Optimizing Target) fitness, through the decoding formula as population
Each individuality in the population of choosing is decoded, obtain s
i, and calculate its pairing fitness f (s
i);
2) after having confirmed all fitness, population is selected heredity.At first calculate each individual selection probability; Its formula is
after obtaining whole selection probability, through making
Select with gambling wheel back-and-forth method.This optimization method is simulated as follows:
A) at [0,1] interval interior equally distributed random numbers r that produces.
B) if r≤q
1, then individual s
1Selected.
C) if q
K-1＜r≤q
k(2≤k≤M), then individual s
kSelected.Q wherein
iBe called individual s
i(i=1,2 ..., accumulation probability M), its formula is:
When calculate each individual choose number of times after, promptly available new individuality replaces the original individual S of colony's (being sub-group) that forms a new generation
2, and algebraically counter t is set to t+1.
3) continue the population of a new generation is carried out crisscross inheritance.The crossing-over rate (crossover rate) of setting population is p
cAnd point of intersection x, the intersection number of individuals that calculates in the population does | Mp
c|, and from population, randomly draw the individuality of corresponding number, interlace operation is carried out in pairing, and (if intersect number is odd number; Then last intrasubject intersects); Promptly exchange the gene (binary number) on two individuals, 1 to the x position, and with new individual former individuality, the generation population S of replacing that produces
3
4) continue the population of a new generation is carried out mutation genetic.The aberration rate (Mutation rate) of setting population is P
mAnd mutant gene seat y, from population S
3In confirm at random | Mp
m| individuals, carry out mutation genetic, promptly change the gene on the corresponding mutant gene seat of individuality, and with new individual former individuality, the generation population S of replacing that produces
4
5) repeat top genetic manipulation,, get the maximum individuality of optimal solution or fitness this moment as asking result, the algorithm end up to reaching hereditary end condition (the maximum population generation of setting appears or reached in optimal solution).
As preferably, through using genetic algorithm GA, for each the unit task (CellTask) among the MAP figure of control parameters of engine is specified the initialization factor (M, F, c, m, p
c, p
m), carry out genetic manipulation (select, intersect, variation) then, population is constantly bred optimization, obtain optimal solution at last and write among the Cell Task, as final Control Parameter.
The present invention compares with existing method, has the following advantages: 1. the automation that has height.Demarcate the engineer does not need to carry out staking-out work always, and only need provide the boundary conditions of its parameter area and optimization objects, can accomplish the optimization to whole M AP figure.2. this method has the objectivity and the correctness of height.Optimization method carries out genetic iteration through genetic algorithm to the parameter on the whole coverage area, and this makes the last optimal solution that produces be applicable to gamut.3. versatility.Because whole optimization method is based on genetic algorithm, it does not use any specific aim particular value, and this makes that whole optimization method can be multiplexing by various optimization objects institute, increases spreadability of the present invention.4. Security.Whole optimization method is based upon on the ripe algorithm, uses safer.
Description of drawings
Fig. 1 genetic algorithm schematic flow sheet
The difference schematic representation of Fig. 2 traditional algorithm and genetic algorithm;
Fig. 3 generally demarcates schematic flow sheet;
Fig. 4 optimized Algorithm provided by the invention residing position view in the whole calibrating flow process;
Embodiment
Below in conjunction with accompanying drawing and embodiment the present invention is described further:
Embodiment:
1. the demarcation with advance angle of engine ignition is an example, carries out the explanation of practical implementation method: according to demarcating conceptual design, this example adopts the simplest four cylinder engine Tesis model to carry out last analogue simulation.In this model, the reference parameter of ignition advance angle is engine speed and load torque, therefore with engine speed and load torque as factor of influence, carry out the MAP figure formulation of ignition advance angle.Based on the demarcation demand, the MAP figure of the selected 10*8 of this example describes, and its granularity is 80.Its MAP figure is as shown in table 1.
Calibrating parameters MAP figure operating point is divided (MAP figure granularity depends on optimization aim) under table 1 genetic algorithm
2. each Cell Task among the MAP figure is carried out genetic algorithm optimization.With Cell Task11 is example, at first specifies the aberration rate of its genetic algorithm, might as well be 0.01, and crossing-over rate might as well be 0.5, and the pedestal that intersects is 6, and group size might as well be 50, and genetic algorithm stops evolutionary generation, might as well be 100, and the initial population number is 20.After accomplishing initialization, promptly begin ignition advance angle is encoded.The scope of ignition advance angle is generally 25 ° to 50 °; Suppose that its required precision is 0.01; So; Calculating it according to coding formula
is 12; Through random device, obtain 20 middle random numberss of 25-50, through code conversion 12 binary representation with it.As optimization objects, fitness function is the engine power of Tesis engine mockup under corresponding ignition advance angle to last this example with engine power.
3. successively 20 individuals are carried out fitness and calculate, obtain individual fitness set f (s at last
i) (i=1,2 ..., 20), according to this each individual selection probability of fitness set calculating be:
Next, using gambling wheel back-and-forth method to carry out genetic optimization selects.According to this method, choose the random numbers r of [0,1] for each individuality
i, according to criterion: if r
i≤q
1, then individual s
1Selected, if q
K-1＜r
i≤q
k(2≤k≤M), then individual s
kSelected (q wherein
iBe called individual s
i(i=1,2 ..., accumulation probability M), its formula is:
After selecting end,, produce population of future generation with the old individuality of the individual replacement of a new generation.
4. the group to new generation carries out crisscross inheritance.Because crossing-over rate is 0.5, the number of individuals that calculates intersection is 20*0.5=10, promptly has 10 individuals to have crossbreeding heredity, and it is divided into 5 groups, carries out crisscross inheritance.Because the base that intersects is 6, thus to thereafter six intersect, produce the old individuality of the individual replacement of a new generation, produce population of future generation.
5. the group to new generation carries out mutation genetic.Because aberration rate is 0.01; The gene figure place that calculates variation is 12*20*0.01=2.4; Rounding promptly has 2 genes to produce variation, and two genes randomly drawing in the new population carry out mutation operation, promptly to the gene negate on this; Produce the old individuality of the individual replacement of a new generation, produce population of future generation.
6. repeat the operation of 2-5, stop evolutionary generation 100 up to reaching, the highest individuality of obtaining of fitness is with its final optimal value as this Cell Task.
7. traversal is accomplished all Cell Task, has promptly obtained optimum ignition advance angle MAP figure.
Though describe the present invention through above-mentioned instantiation; But the present invention is not limited to above-mentioned instance; Related personnel for related domain; Can make other various changes and distortion, and all these should belong in the protection domain of claim of the present invention according to technological scheme of the present invention and thought.
Claims (3)
1. the automatic calibrating parameters optimization method of the motor based on genetic algorithm is characterized in that: on the basis of original calibration technique, introduce genetic algorithm; When carrying out the optimization of control parameters of engine, utilize genetic algorithm to set up its pairing initial population for each parameter, fitness is passed judgment on formula; Crossing-over rate, aberration rate and end condition carry out the genetic algorithm optimization operation; Finally obtain optimal base because of the Control Parameter in scheming as MAP, its concrete steps are:
Confirm the optimization objects of demarcation task, be called for short OT; According to OT, need to confirm the MAP that demarcates to scheme and parameter, include 1-2 reference parameter among this MAP figure, be used for confirming of operating point, include at least 1 Control Parameter simultaneously, in order to demarcate optimization; Then each unit task Cell Task among the MAP figure is carried out genetic algorithm optimization;
Based on the required control accuracy that reaches of OT; For MAP figure specifies its granularity; The distribution of design conditions point, operating point to be evenly to be distributed as the master, and to be divided into the operating point be a plurality of little task Cell Task of standard to whole M AP figure the most at last; Each Cell Task need carry out primary parameter optimization, and whole M AP figure is evolved to optimum state;
Be the initialization factor of each Cell Task appointment optimized Algorithm, for, wherein M is a group size, and F is an ideal adaptation degree evaluation function, and c is the interlace operation operator, and m is the mutation operation operator, is crossover probability, is the variation probability;
According to the specific requirement precision, for optimization objects is encoded, the span with optimization aim of confirming as of the formula of encoding is decided to be, given its required precision, for, calculate its binary-coded length:, be integer; Then, utilize method at random, to the generation initial population, the size of population is decided to be M according to demand;
Get into the genetic optimization of genetic algorithm, wherein select successively, intersect; Three kinds of modes of inheritance make a variation; And this process that constantly circulates is up to reaching end condition, and end condition is for obtaining optimum value or for reaching the termination genetic algebra, obtaining the optimized parameter of optimum value for this Cell Task at last.
2. according to the automatic calibrating parameters optimization method of the said motor of claim 1, it is characterized by based on genetic algorithm:
With the value of feedback of OT fitness as population, through the decoding formula each individuality in the population of choosing is decoded, obtain, and calculate its pairing fitness;
Calculate each individual selection probability; Its formula is after obtaining whole selection probability, to select through using gambling wheel back-and-forth method; When calculate each individual choose number of times after; Be that available new individuality replaces the original individual colony that forms a new generation, i.e. sub-group, and algebraically counter is set to;
Continuation is carried out crisscross inheritance to the population of a new generation, and the crossing-over rate of setting population is for reaching the point of intersection, and the intersection number of individuals that calculates in the population does; And from population, randomly draw the individuality of corresponding number, interlace operation is carried out in pairing, is odd number if intersect number; Then last intrasubject intersects; Promptly exchange the gene of two individuals 1 on putting in place, and new individually replace former individuality, produce population with what produce;
Continuation is carried out mutation genetic to the population of a new generation, and the aberration rate of setting population is confirmed individuals at random for reaching the mutant gene seat from population; Carry out mutation genetic; Promptly change the gene on the corresponding mutant gene seat of individuality, and replace former individuality with the new individuality that produces, the generation population;
Genetic manipulation above repeating, up to reaching hereditary end condition, promptly optimal solution occurs or has reached the maximum population generation of setting, and gets the maximum individuality of optimal solution or fitness this moment as asking result, and algorithm finishes.
3. according to the automatic calibrating parameters optimization method of the said motor of claim 1, it is characterized by: through using genetic algorithm GA, for each the unit task among the MAP figure of control parameters of engine is specified the initialization factor based on genetic algorithm; Select successively then; Intersect, the three kinds of modes of inheritance that make a variation are constantly bred optimization to population; Obtain optimal solution at last and write among the Cell Task, as final Control Parameter.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2011102301122A CN102337979A (en) | 2011-08-11 | 2011-08-11 | Automatic calibration parameter optimization method of engine based on genetic algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2011102301122A CN102337979A (en) | 2011-08-11 | 2011-08-11 | Automatic calibration parameter optimization method of engine based on genetic algorithm |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102337979A true CN102337979A (en) | 2012-02-01 |
Family
ID=45513893
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2011102301122A Pending CN102337979A (en) | 2011-08-11 | 2011-08-11 | Automatic calibration parameter optimization method of engine based on genetic algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102337979A (en) |
Cited By (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104408271A (en) * | 2014-12-20 | 2015-03-11 | 吉林大学 | Model-based gasoline engine calibration method |
| CN104884774A (en) * | 2012-09-17 | 2015-09-02 | Mtu腓特烈港有限责任公司 | Method for operating an internal combustion engine |
| CN105045286A (en) * | 2015-09-16 | 2015-11-11 | 北京中科遥数信息技术有限公司 | Automatic pilot and genetic algorithm-based method for monitoring hovering range of unmanned aerial vehicle |
| CN105545493A (en) * | 2015-12-29 | 2016-05-04 | 中国航空工业集团公司沈阳发动机设计研究所 | Engine control rule computing method based on genetic algorithm |
| CN106597020A (en) * | 2016-11-25 | 2017-04-26 | 中国船舶重工集团公司第七0五研究所 | Turntable-free calibration method for accelerometer based on genetic algorithm |
| CN106874503A (en) * | 2017-02-24 | 2017-06-20 | 珠海迈科智能科技股份有限公司 | The method and apparatus for obtaining recommending data |
| CN107170442A (en) * | 2017-05-11 | 2017-09-15 | 北京理工大学 | Multi-parameters optimization method based on self-adapted genetic algorithm |
| CN108509487A (en) * | 2018-02-08 | 2018-09-07 | 杨睿嘉 | Image search method, equipment and the storage medium of cortex model are provided based on pulse |
| CN109165783A (en) * | 2018-08-15 | 2019-01-08 | 中国石油化工股份有限公司江汉油田分公司勘探开发研究院 | Oilfield development program optimization method and device |
| CN109343342A (en) * | 2018-11-23 | 2019-02-15 | 江苏方天电力技术有限公司 | Electric precipitator energy conservation optimizing method and system based on genetic algorithm |
| CN109753712A (en) * | 2018-12-28 | 2019-05-14 | 西安交通大学 | A kind of optimum design method of Neutron Dose Equivalent Rate instrument |
| CN109765875A (en) * | 2018-11-30 | 2019-05-17 | 联合汽车电子有限公司 | Whole vehicle functions module automatic calibration system and method |
| CN110080896A (en) * | 2019-04-24 | 2019-08-02 | 河南省图天新能源科技有限公司 | A kind of methane fuelled engine air/fuel ratio control method based on genetic algorithm |
| CN110505634A (en) * | 2019-08-17 | 2019-11-26 | 温州大学 | One kind realizing wireless aps disposition optimization method based on genetic algorithm |
| CN112037289A (en) * | 2020-09-16 | 2020-12-04 | 安徽意欧斯物流机器人有限公司 | Off-line parameter calibration method based on genetic algorithm |
| CN112446109A (en) * | 2020-11-04 | 2021-03-05 | 潍柴动力股份有限公司 | Calibration method and device for fuel injection pressure of engine |
| CN112464395A (en) * | 2020-11-03 | 2021-03-09 | 潍柴动力股份有限公司 | Method and device for calibrating physical model parameters of engine |
| CN113449897A (en) * | 2020-03-25 | 2021-09-28 | 联合汽车电子有限公司 | Method for optimizing point sweeping of test parameters of engine bench |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101281551A (en) * | 2007-12-28 | 2008-10-08 | 奇瑞汽车股份有限公司 | Process for reducing vehicle fuel consume |
| CN101333961A (en) * | 2008-08-07 | 2008-12-31 | 清华大学 | Hydrogen gas natural gas mixed fuel engine optimizing method |
-
2011
- 2011-08-11 CN CN2011102301122A patent/CN102337979A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101281551A (en) * | 2007-12-28 | 2008-10-08 | 奇瑞汽车股份有限公司 | Process for reducing vehicle fuel consume |
| CN101333961A (en) * | 2008-08-07 | 2008-12-31 | 清华大学 | Hydrogen gas natural gas mixed fuel engine optimizing method |
Non-Patent Citations (3)
| Title |
|---|
| 吴锋,贾岱润,姚栋伟,杨志家: "电控汽油机点火提前角多目标优化方法研究", 《内燃机工程》, vol. 29, no. 1, 29 February 2008 (2008-02-29), pages 24 - 28 * |
| 张炳达: "《智能信息处理技术基础》", 31 October 2008, article "第四章 遗传算法", pages: 41-61 * |
| 黄钰,黄海波,陈绪平: "电控CNG发动机标定方法", 《西华大学学报》, vol. 24, no. 6, 30 November 2005 (2005-11-30), pages 6 - 9 * |
Cited By (25)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104884774A (en) * | 2012-09-17 | 2015-09-02 | Mtu腓特烈港有限责任公司 | Method for operating an internal combustion engine |
| CN104408271B (en) * | 2014-12-20 | 2017-07-07 | 吉林大学 | A kind of gasoline engine scaling method based on model |
| CN104408271A (en) * | 2014-12-20 | 2015-03-11 | 吉林大学 | Model-based gasoline engine calibration method |
| CN105045286A (en) * | 2015-09-16 | 2015-11-11 | 北京中科遥数信息技术有限公司 | Automatic pilot and genetic algorithm-based method for monitoring hovering range of unmanned aerial vehicle |
| CN105545493A (en) * | 2015-12-29 | 2016-05-04 | 中国航空工业集团公司沈阳发动机设计研究所 | Engine control rule computing method based on genetic algorithm |
| CN106597020A (en) * | 2016-11-25 | 2017-04-26 | 中国船舶重工集团公司第七0五研究所 | Turntable-free calibration method for accelerometer based on genetic algorithm |
| CN106597020B (en) * | 2016-11-25 | 2019-10-22 | 中国船舶重工集团公司第七0五研究所 | A kind of acceleration planned immunization turntable scaling method based on genetic algorithm |
| CN106874503A (en) * | 2017-02-24 | 2017-06-20 | 珠海迈科智能科技股份有限公司 | The method and apparatus for obtaining recommending data |
| CN106874503B (en) * | 2017-02-24 | 2020-03-20 | 珠海迈科智能科技股份有限公司 | Method and device for acquiring recommended data |
| CN107170442A (en) * | 2017-05-11 | 2017-09-15 | 北京理工大学 | Multi-parameters optimization method based on self-adapted genetic algorithm |
| CN108509487A (en) * | 2018-02-08 | 2018-09-07 | 杨睿嘉 | Image search method, equipment and the storage medium of cortex model are provided based on pulse |
| CN109165783A (en) * | 2018-08-15 | 2019-01-08 | 中国石油化工股份有限公司江汉油田分公司勘探开发研究院 | Oilfield development program optimization method and device |
| CN109343342A (en) * | 2018-11-23 | 2019-02-15 | 江苏方天电力技术有限公司 | Electric precipitator energy conservation optimizing method and system based on genetic algorithm |
| CN109765875A (en) * | 2018-11-30 | 2019-05-17 | 联合汽车电子有限公司 | Whole vehicle functions module automatic calibration system and method |
| CN109765875B (en) * | 2018-11-30 | 2021-07-16 | 联合汽车电子有限公司 | Automatic calibration system and method for whole vehicle function module |
| CN109753712A (en) * | 2018-12-28 | 2019-05-14 | 西安交通大学 | A kind of optimum design method of Neutron Dose Equivalent Rate instrument |
| CN110080896A (en) * | 2019-04-24 | 2019-08-02 | 河南省图天新能源科技有限公司 | A kind of methane fuelled engine air/fuel ratio control method based on genetic algorithm |
| CN110505634A (en) * | 2019-08-17 | 2019-11-26 | 温州大学 | One kind realizing wireless aps disposition optimization method based on genetic algorithm |
| CN110505634B (en) * | 2019-08-17 | 2020-06-02 | 温州大学 | Method for realizing wireless AP deployment optimization based on genetic algorithm |
| CN113449897A (en) * | 2020-03-25 | 2021-09-28 | 联合汽车电子有限公司 | Method for optimizing point sweeping of test parameters of engine bench |
| CN113449897B (en) * | 2020-03-25 | 2024-04-16 | 联合汽车电子有限公司 | Method for optimizing engine bench test parameters and sweeping points |
| CN112037289A (en) * | 2020-09-16 | 2020-12-04 | 安徽意欧斯物流机器人有限公司 | Off-line parameter calibration method based on genetic algorithm |
| CN112464395A (en) * | 2020-11-03 | 2021-03-09 | 潍柴动力股份有限公司 | Method and device for calibrating physical model parameters of engine |
| CN112446109A (en) * | 2020-11-04 | 2021-03-05 | 潍柴动力股份有限公司 | Calibration method and device for fuel injection pressure of engine |
| CN112446109B (en) * | 2020-11-04 | 2022-10-28 | 潍柴动力股份有限公司 | Calibration method and device for fuel injection pressure of engine |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102337979A (en) | Automatic calibration parameter optimization method of engine based on genetic algorithm | |
| Nicolini et al. | Optimal location and control of pressure reducing valves in water networks | |
| Hassanzadeh et al. | Minimizing total resource consumption and total tardiness penalty in a resource allocation supply chain scheduling and vehicle routing problem | |
| CN111337258B (en) | Device and method for online calibration of engine control parameters by combining genetic algorithm and extremum search algorithm | |
| CN104700160A (en) | Vehicle route optimization method | |
| CN102682348B (en) | Complex equipment parts for maintenance level optimization system and method for building up thereof | |
| CN110929913B (en) | Multi-target power generation plan decomposition coordination calculation method for direct-current cross-district interconnected power grid | |
| CN102184287A (en) | Modelling method for combustion optimization of waste plastics oil refining | |
| CN110460038A (en) | It is a kind of to be related to more scene method for expansion planning of power transmission network of new-energy grid-connected | |
| CN107657349B (en) | Method for extracting scheduling rules of staged power generation of reservoir | |
| CN113361761A (en) | Short-term wind power integration prediction method and system based on error correction | |
| CN111342456A (en) | Method and system for modeling energy system of transformer area | |
| CN103714226A (en) | Automatic generating method and automatic generating device for optimized orderly-power-consumption scheme | |
| Viganò et al. | Creation of the Italian Distribution System Scenario by Using Synthetic Artificial Networks | |
| CN106647242A (en) | Multivariable PID controller parameter setting method | |
| CN104200271A (en) | Multi-objective optimization algorithm for engine | |
| CN107038489B (en) | Multi-target unit combination optimization method based on improved NBI method | |
| CN115764870A (en) | Multivariable photovoltaic power generation power prediction method and device based on automatic machine learning | |
| CN106094572B (en) | A kind of source relates to net pilot production closed-loop simulation identification application method | |
| CN114997630A (en) | Multi-region environment economic scheduling method based on competitive learning constraint multi-target particle swarm algorithm | |
| CN111563699B (en) | Power system distribution robust real-time scheduling method and system considering flexibility requirement | |
| CN112949177A (en) | Multi-optimization target weighting method and system in construction of comprehensive energy system | |
| CN110080896A (en) | A kind of methane fuelled engine air/fuel ratio control method based on genetic algorithm | |
| Jankovic et al. | Taking a Passivhaus certified retrofit system onto scaled-up zero carbon trajectory | |
| KR101778599B1 (en) | Optimization design method of solar thermal system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C12 | Rejection of a patent application after its publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20120201 |