CN111337258A - Device and method for online calibration of engine control parameters by combining genetic algorithm and extremum search algorithm - Google Patents
Device and method for online calibration of engine control parameters by combining genetic algorithm and extremum search algorithm Download PDFInfo
- Publication number
- CN111337258A CN111337258A CN202010094829.8A CN202010094829A CN111337258A CN 111337258 A CN111337258 A CN 111337258A CN 202010094829 A CN202010094829 A CN 202010094829A CN 111337258 A CN111337258 A CN 111337258A
- Authority
- CN
- China
- Prior art keywords
- engine
- control parameters
- control
- algorithm
- optimization
- 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
- 238000004422 calculation algorithm Methods 0.000 title claims abstract description 100
- 238000000034 method Methods 0.000 title claims abstract description 72
- 230000002068 genetic effect Effects 0.000 title claims abstract description 44
- 238000010845 search algorithm Methods 0.000 title claims abstract description 23
- 238000005457 optimization Methods 0.000 claims abstract description 120
- 238000002485 combustion reaction Methods 0.000 claims abstract description 46
- 238000004891 communication Methods 0.000 claims abstract description 11
- 239000000446 fuel Substances 0.000 claims description 38
- 230000008569 process Effects 0.000 claims description 25
- 230000004044 response Effects 0.000 claims description 21
- 238000004364 calculation method Methods 0.000 claims description 13
- 238000012544 monitoring process Methods 0.000 claims description 12
- 238000002347 injection Methods 0.000 claims description 8
- 239000007924 injection Substances 0.000 claims description 8
- 239000000498 cooling water Substances 0.000 claims description 3
- 230000008878 coupling Effects 0.000 claims description 3
- 238000010168 coupling process Methods 0.000 claims description 3
- 238000005859 coupling reaction Methods 0.000 claims description 3
- 238000011156 evaluation Methods 0.000 claims description 2
- 238000009838 combustion analysis Methods 0.000 claims 1
- 238000012360 testing method Methods 0.000 description 33
- 230000008859 change Effects 0.000 description 10
- 238000013461 design Methods 0.000 description 8
- 238000005516 engineering process Methods 0.000 description 5
- 239000000243 solution Substances 0.000 description 5
- 125000004122 cyclic group Chemical group 0.000 description 4
- 238000013507 mapping Methods 0.000 description 4
- 238000003672 processing method Methods 0.000 description 3
- 108090000623 proteins and genes Proteins 0.000 description 3
- 238000004088 simulation Methods 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 230000001419 dependent effect Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000008303 genetic mechanism Effects 0.000 description 2
- 238000012804 iterative process Methods 0.000 description 2
- 238000012067 mathematical method Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000002146 bilateral effect Effects 0.000 description 1
- 230000008094 contradictory effect Effects 0.000 description 1
- 238000011217 control strategy Methods 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000013400 design of experiment Methods 0.000 description 1
- 238000013401 experimental design Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000012812 general test Methods 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M15/00—Testing of engines
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B23/00—Testing or monitoring of control systems or parts thereof
- G05B23/02—Electric testing or monitoring
- G05B23/0205—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults
- G05B23/0208—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterized by the configuration of the monitoring system
- G05B23/0213—Modular or universal configuration of the monitoring system, e.g. monitoring system having modules that may be combined to build monitoring program; monitoring system that can be applied to legacy systems; adaptable monitoring system; using different communication protocols
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/12—Computing arrangements based on biological models using genetic models
- G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biophysics (AREA)
- Theoretical Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Artificial Intelligence (AREA)
- General Health & Medical Sciences (AREA)
- Physiology (AREA)
- Biomedical Technology (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Genetics & Genomics (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Automation & Control Theory (AREA)
- Combined Controls Of Internal Combustion Engines (AREA)
Abstract
The invention relates to the technical field of automatic calibration methods of engine parameters, in particular to an online calibration device and method of engine control parameters by combining a genetic algorithm and an extremum search algorithm. The device comprises an engine, a dynamometer, an engine controller, a combustion analyzer, a Mirco AutoBox rapid control prototype, a dynamometer upper computer control system, an engine controller upper computer, a Mirco AutoBox rapid control prototype upper computer and a CAN bus communication system; the invention relates to an on-line automatic calibration device for engine control parameters, which continuously changes the control parameters through an optimization algorithm under the fixed working condition of an engine and solves an optimization objective function to realize the automatic optimization of the control parameters.
Description
Technical Field
The invention relates to the technical field of automatic calibration methods of engine parameters, in particular to an online calibration device and method of engine control parameters by combining a genetic algorithm and an extremum search algorithm.
Background
During the operation of the engine, an Engine Controller (ECU) sends execution instructions to various actuators on the engine, such as an oil injector, an ignition coil and the like, according to the operation conditions (rotation speed and throttle opening determining conditions) of the engine. Taking an injector as an example, the injector includes opening duration, opening time and the like, and the execution commands of the actuators are collectively referred to as control parameters. If the engine can stably operate under a certain working condition, the control parameters under the working condition must be reasonably set. The control parameters set for the purpose of operating stability may be a wide range, and engine performance, such as dynamics, economy, and emissions, may vary as the control parameters vary. As for the overall performance of the engine, it is desirable that the power is better (the higher the torque is), the economy is better (the lower the fuel consumption rate), and the emission characteristics are better (the lower the emission concentration is), but there is a tendency to trade off among these three performances, and thus there is a contradiction in setting the control parameters. Because the engine has a plurality of control parameters to be calibrated, the number of the control parameters is more and more along with the increase of the electric control degree of the engine, and the control parameters are mutually coupled, and for the same engine performance, the influence rules of different control parameters are different, so that the control parameters need to be calibrated by a corresponding method to realize the expectation of the engine performance.
The traditional calibration method needs to comprehensively consider the influence relationship between each working condition and each control parameter, and carry out a large number of control parameter adjustment tests under different working conditions so as to find out the optimal control parameter variable meeting the requirements under different working conditions. On one hand, because the change of the test condition is difficult to calibrate, a general test bench test system is difficult to provide a standard dynamic working condition; on the other hand, repeated dynamic testing has a significant impact on the bench, making the calibration process difficult to reproduce. The traditional calibration method has the advantages of low theoretical level, parameter trial and error, low working efficiency and almost no use.
At present, two common calibration methods are available, namely model-based calibration and online optimization calibration.
1. The calibration technology based on the model is developed on the basis of the traditional calibration technology, a mathematical optimization theory is introduced into the calibration of the control parameters of the engine controller, different forms of fitting are carried out on test data to establish the model, and meanwhile, the advanced method of optimizing the control parameters is carried out on the basis of the model. The model-based optimization calibration method can flexibly and conveniently simulate various running states of the engine and has high repeatability. In the development process of the electric control system, the simulation working condition can be conveniently utilized to test the control system software and the control strategy, and the number of bench tests is reduced. Because the simulation model is adopted to replace an actual system, the test times of variable parameters can be not limited, and the simulation result has repeatability; the test under the limit state can be carried out without destructiveness, low cost and danger. Therefore, the bench test work can be greatly reduced, the cost is reduced, the development and test time is obviously shortened, and the calibration efficiency is improved. The calibration method based on model control is provided, so that the calibration technology is greatly developed and becomes an important means for developing an electric control system.
The calibration technology based on the model is off-line optimization, proper control parameters cannot be obtained in the real-time running process of the engine, and the model of the engine is obtained by performing a large number of tests and then utilizing a mathematical method according to test data; and then, aiming at the engine model by using an optimization algorithm, establishing an optimization objective function to obtain the optimal control parameters. The calibration based on the model can be divided into three aspects, wherein the first aspect is to carry out a test to obtain corresponding data of control parameters and engine performance; secondly, establishing an engine model and fitting test data; and thirdly, establishing an optimization target and optimizing control parameters off line.
In the experimental field, in order to reduce the amount of experiments as much as possible and ensure the data of model fitting, a design of experiments method (DOE) was developed. The most traditional test design method is the full factorial method, which requires all parameters to be set as variables in turn, and tests are performed in all ranges where this parameter can vary, until all parameters have been tested. The method can be used when the control parameters are less, but the test task becomes more and more complicated and the test time is longer and longer as the control parameters are increased, so that the test is almost impossible to carry out. Other test designs scientifically and reasonably select test working conditions on the basis of the test working conditions selected by the full factor method, and research influence relations among various control parameters and between the control parameters and the engine performance. Common methods include optimal design, orthogonal experimental design, optimization, regression orthogonal design, uniform design, and the like.
In modeling, the goal is to abstract an engine from specified performance metrics and their associated constraints, and to describe the exact mathematical model between engine controllable parameters and engine performance response. To achieve accurate calibration, the response model must operate quickly and accurately with good predictive generalization capability. The engine modeling mainly comprises the step of establishing a mapping relation of performance parameters of the engine to control parameters. Because the mapping relation is a complex nonlinear relation involving a plurality of inputs, a simple processing method is to set up the mapping relation between the inputs and the outputs only for the data itself without considering the inherent physical relation between each input parameter and the response characteristic, which accurately reflects the different influences of the control parameters on the engine response, and another processing method is to properly distinguish the control parameters, thus obtaining two different modeling methods: single-order modeling and second-order modeling. The single-order modeling is mainly directed to local models, while the second-order modeling is directed to global models, and the mathematical fitting methods used are most commonly polynomial fitting, regression model fitting, neural network fitting, and the like.
In terms of optimization algorithm design, many of the engine performance indicators are contradictory, and the optimization algorithm can seek a compromise between the performance indicators. And because the engine is a complex system, the control parameters have different influences on different performance indexes, so that the optimization of the control parameters belongs to the optimization problem of multiple parameters and multiple targets. Because of off-line optimization, the engine performance response is real-time, so the optimization algorithm is selected in many ways, most commonly lagrangian multiplier method, least square method, genetic algorithm, artificial neural network method and the like.
The calibration technology based on the model is off-line optimization, and the optimal control parameters cannot be obtained on line in real time in the running process of the engine;
the control parameters obtained through the optimization algorithm are the optimal solutions of the fitting model, errors exist between the results of the fitting model and the real engine performance response, and whether the obtained optimal solution of the fitting model is the real engine optimal control parameters depends on the precision of the fitting model. On one hand, the fitting process of the model requires that data are comprehensive enough and can reflect the response of the engine in detail, and although the test is carried out according to a test design method, the obtained test data can well meet the modeling requirement, errors always exist between the obtained test data and the real engine; on the other hand, when a model is fitted by using a mathematical method, there is also an error between the fitted model and the test data. The error between the fitting model and the real engine can be reduced, but can not be eliminated all the time, which is an inevitable problem.
2. In the aspect of online calibration of engine control parameters, existing patents mainly refer to introduction of an automatic calibration system, and the automatic calibration system comprises a plurality of control modules such as a control unit, a collection unit and an optimization unit. The control unit outputs the control parameters of the actuator, the acquisition unit acquires the response results (torque, fuel consumption rate, emission and the like) of the engine performance, and the optimization unit compares the engine responses under different control parameters to obtain the optimal control parameters.
The automatic calibration system mainly introduces the system, does not introduce the optimization algorithm, does not provide a proper method for optimizing the control parameters, and does not provide an optimization objective function;
the method can not ensure that under a certain working condition, the optimal control parameter based on the optimization objective function can be found out in a full-automatic closed loop manner, and the engine can be operated according to the optimal control parameter.
In the aspect of designing an online calibration optimization algorithm, the more common optimization algorithms are a genetic algorithm and an extremum search algorithm. The genetic algorithm is implemented by giving a large number of control parameter combinations and respectively sending the control parameter combinations to an actuator for execution, collecting engine performance responses under different control parameter combinations, recording results of the control parameter combinations and the engine responses, solving an optimization objective function according to the engine responses, and obtaining the optimal control parameter combination of an optimization objective through a plurality of tests and solutions.
And the extremum searching algorithm collects the performance response of the engine, calculates to obtain an optimized objective function value and the gradient of the optimized objective function, determines the next control parameter according to the gradient, and converges the optimized objective function to the minimum value through multiple iterations to obtain the optimal control parameter.
The genetic algorithm can generate a large amount of control parameter combinations, so a large amount of tests are required to solve the optimization objective function, and because the tests are carried out on the engine rack in real time, the tests carried out on each group of control parameters consume a certain time, so that the calibration time is very long;
the extremum searching algorithm has fast convergence time and short calibration time, but the research object is generally a convex optimization problem, and the local optimum point is a global optimum point; and the problem of nonlinear and non-convex optimization between the input of the engine control parameters and the output of the performance response exists, a plurality of local optimal points exist, and the local optimal points are not necessarily global optimal points. Therefore, when the extremum search algorithm is used for solving the optimization objective function, local optimization rather than global optimization is obtained, and therefore, the obtained control parameters are not optimal control parameters.
Disclosure of Invention
In order to solve the problems of the prior art, the invention provides a device and a method for online calibrating engine control parameters by combining a genetic algorithm and an extremum search algorithm, and the scheme is directed at an online automatic calibration device for the engine control parameters, which can continuously change the control parameters by an optimization algorithm under the fixed working condition of an engine and solve an optimization objective function to realize automatic optimization of the control parameters. The two are combined, so that the optimization time is shortened, and the optimal control parameters are accurately found.
The technical scheme adopted by the invention is as follows:
the device for calibrating the control parameters of the engine on line by combining a genetic algorithm and an extremum search algorithm comprises the engine, a dynamometer, an engine controller, a combustion analyzer, a Mirco AutoBox rapid control prototype, a dynamometer upper computer control system, an engine controller upper computer, a Mirco AutoBox rapid control prototype upper computer and a CAN bus communication system, wherein the engine controller upper computer is connected with the dynamometer and the dynamometer is connected with the engine controller through the CAN bus communication system:
the engine and the dynamometer are connected through a coupling, and the dynamometer is used for providing starting rotating speed for starting the engine, fixing the rotating speed of the engine and measuring the torque of the engine;
the engine controller is used for controlling an actuator on the engine, receiving signals collected by the sensor and communicating with an upper computer of the engine controller through a CAN bus;
the combustion analyzer acquires in-cylinder pressure and a crank angle through a cylinder pressure sensor arranged in an engine cylinder and a corner mark instrument on a crank shaft, further calculates average indicating pressure of each cycle, communicates with a Mirco AutoBox rapid control prototype through a CAN bus, and sends information to the Mirco AutoBox rapid control prototype;
the Mirco AutoBox rapid control prototype is used for building an optimization algorithm, and calculating the next control parameter by collecting input quantity;
the dynamometer upper computer control system is used for controlling the dynamometer and monitoring the state of the dynamometer and the measured engine torque;
the upper computer of the engine controller is used for sending a control command to the engine controller so as to control the actuator; monitoring the engine state collected by the sensor; when the control parameters are automatically calibrated, the control parameters calculated by the Mirco AutoBox rapid control prototype are received, and then the control parameters are sent to an engine controller;
the Mirco AutoBox rapid control prototype upper computer is used for monitoring input and output quantity during calculation of an optimization algorithm, and is also provided with an automatic calibration switch for controlling the starting and ending of an automatic calibration process;
the CAN bus communication system consists of two CAN buses, wherein one CAN bus is connected with a Mirco AutoBox rapid control prototype and a combustion analyzer, and the combustion analyzer sends the IMEP obtained by calculation to the Mirco AutoBox rapid control prototype; the other CAN bus is connected with the engine controller, an upper computer of the engine controller and the Mirco AutoBox rapid control prototype, and the engine controller sends signals acquired by the sensor to the upper computer of the engine controller through the CAN bus for monitoring the state of the engine; the upper computer of the engine controller issues the control parameters to the engine controller through a CAN bus, and the engine controller controls the actuator to execute according to the parameters; the Mirco AutoBox rapid control prototype monitors oil injection pulse width sent by an engine controller on a CAN bus to calculate fuel consumption rate, and simultaneously sends corresponding control parameters to the engine controller in the automatic online parameter calibration process to replace the same control parameters sent by an upper computer of the engine controller.
Meanwhile, the method for calibrating the engine control parameters on line by combining the genetic algorithm and the extremum search algorithm comprises the following steps:
A. setting an engine starting rotating speed by an upper computer control system of the dynamometer, and presetting a rotating speed of stable operation of the engine;
B. starting, the engine stably runs at a preset running rotating speed, and at the moment, a dynamometer upper computer control system monitors dynamic indexes such as dynamometer states, engine torque and the like in real time; then setting the opening of a throttle valve through an upper computer of an engine controller, determining the operation condition, and detecting the exhaust temperature and the cooling water temperature state of the engine in real time;
C. opening a switch for controlling automatic calibration of parameters on a Mirco AutoBox rapid control prototype upper computer to automatically calibrate the control parameters;
D. the combustion analyzer collects and calculates the average indicated pressure of the engine in real time in the whole process, communicates through a CAN bus, and continuously sends the information of the average indicated pressure to the Mirco AutoBox rapid control prototype;
E. after receiving an instruction for starting the automatic calibration process of the control parameters, the Mirco AutoBox rapid control prototype receives average indicated pressure sent by a combustion analyzer, and calculates to obtain average indicated torque as an engine dynamic index;
F. receiving an oil injection pulse width sent by an engine controller, and calculating the fuel consumption rate as an engine economic index;
G. after the performance index of the engine is obtained, an optimization objective function under the control parameter at the moment is solved, the control parameter is changed through an optimization algorithm, and a new control parameter is sent to an engine controller through a CAN bus;
H. the engine controller continuously monitors and receives the control parameters sent by the Mirco AutoBox rapid control prototype on the CAN bus to replace the control parameters sent by the upper computer of the engine controller, and after new control parameters are obtained, the engine controller executes the control actuator according to the control parameters, and the performance of the engine changes to carry out next optimization calculation.
The technical scheme provided by the invention has the beneficial effects that:
1. compared with the traditional calibration, the provided calibration process is completely automatically carried out on line without manual participation;
2. the design of the optimization algorithm ensures that the optimization objective function corresponding to the finally converged control parameters is globally optimal rather than locally optimal;
3. compared with the online calibration based on the genetic algorithm, the provided optimization algorithm has the advantages of high convergence rate and accurate search result.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
FIG. 1 is a plant layout of an apparatus for online calibration of engine control parameters incorporating a genetic algorithm and an extremum seeking algorithm in accordance with the present invention;
FIG. 2 is a diagram of a CAN bus communication system of an apparatus for on-line calibration of engine control parameters in combination with a genetic algorithm and an extremum search algorithm according to the present invention;
FIG. 3 is a flowchart of an online calibration method for online calibration of engine control parameters incorporating a genetic algorithm and an extremum seeking algorithm in accordance with the present invention;
FIG. 4 is a flowchart of an iterative process of a genetic algorithm of a method for online calibration of engine control parameters incorporating a genetic algorithm and an extremum seeking algorithm in accordance with the present invention;
FIG. 5 is a flowchart of an iterative process for an extremum seeking algorithm for a method for online calibration of engine control parameters incorporating a genetic algorithm and an extremum seeking algorithm in accordance with the present invention;
FIG. 6 is a flow chart illustrating the cycle number determination of a method for online calibration of engine control parameters incorporating a genetic algorithm and an extremum seeking algorithm in accordance with the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, embodiments of the present invention will be described in detail with reference to the accompanying drawings.
Example one
The embodiment provides an online calibration device for engine control parameters by combining a genetic algorithm and an extremum search algorithm, and as shown in the attached drawing 1, the device comprises an engine, a dynamometer, an engine controller, a combustion analyzer, a Mirco AutoBox rapid control prototype, a dynamometer upper computer control system, an engine controller upper computer and a Mirco AutoBox rapid control prototype upper computer.
The engine is connected with the dynamometer through a coupling, and the dynamometer has the main function of providing starting rotating speed for the starting of the engine; the rotating speed of the engine is fixed, and stable operation conditions are guaranteed; measuring the torque of the engine, etc.
The engine controller is used for controlling various actuators on the engine, receiving signals collected by various sensors and communicating with an upper computer of the engine controller through a CAN bus.
The combustion analyzer collects in-cylinder pressure and crank angle through a cylinder pressure sensor installed in an engine cylinder and an angle gage on a crank shaft, further calculates average indicated pressure (IMEP) of each cycle, communicates with a Mirco AutoBox rapid control prototype through a CAN bus, and sends information such as the IMEP and the like to the Mirco AutoBox rapid control prototype. The calculation process of the combustion analyzer is automatically calculated in the device, IMEP can be directly obtained for users, and therefore the calculation process is not introduced. However, it should be noted that there may be variations between engine cycles, and therefore it is necessary to take an average over a number of cycles as the engine response. In this example, the average of the IMEPs for 100 cycles (which CAN be directly selected and output in the combustion analyzer) was collected and transmitted to a MircoAutoBox rapid control prototype via the CAN bus. Since 100 cycles of data are collected, whether the data collection time is sufficient or not is considered in designing the algorithm, for example, at 2000rpm, it takes 3s to collect 100 cycles, and after issuing control parameters to the engine controller again, the engine response data is collected after waiting 3 s.
The Mirco AutoBox rapid control prototype is used for building an optimization algorithm, and the control parameters of the next time are calculated by collecting necessary input quantity. The dynamometer upper computer control system is used for controlling the dynamometer and monitoring the state of the dynamometer and information such as measured engine torque.
The upper computer of the engine controller is engine calibration software based on LabView, and mainly has the function of sending a control instruction to the engine controller so as to control various actuators; monitoring various states of the engine collected by a sensor; and when the control parameters are automatically calibrated, receiving the control parameters calculated by the Mirco AutoBox rapid control prototype, and sending the control parameters to the engine controller.
The Mirco AutoBox rapid control prototype upper computer is used for controlling the automatic calibration process of parameters without manual participation, so that the Mirco AutoBox rapid control prototype upper computer has the main functions of monitoring input and output quantity during calculation of an optimization algorithm, namely engine performance indexes, related control parameters and the like, and is provided with an automatic calibration switch for controlling the starting and the ending of the automatic calibration process.
The CAN bus communication system will be described in detail. As shown in fig. 2, the communication system is composed of two CAN buses, one of which is connected with a mirco autobox rapid control prototype and a combustion analyzer, and the main function is that the combustion analyzer sends the calculated IMEP to the mirco autobox rapid control prototype; the other CAN bus is connected with the engine controller, the upper computer of the engine controller and the Mirco AutoBox rapid control prototype, and the engine controller sends signals acquired by the sensor to the upper computer of the engine controller through the bus for monitoring the state of the engine. The upper computer of the engine controller issues the control parameters to the engine controller through a bus, and the engine controller controls the actuator to execute according to the parameters. The Mirco AutoBox rapid control prototype monitors oil injection pulse width sent by an engine controller on a bus to calculate fuel consumption rate, and meanwhile, the Mirco AutoBox rapid control prototype sends corresponding control parameters to the engine controller in the automatic online parameter calibration process to replace the same control parameters sent by an upper computer of the engine controller.
Example two
As shown in fig. 3, the present embodiment provides a method for online calibrating engine control parameters by combining a genetic algorithm and an extremum search algorithm, which mainly comprises the steps of setting an engine starting rotation speed in a dynamometer host computer control system, and presetting a rotation speed at which an engine stably runs; then starting, the engine stably runs at a preset running rotating speed, and at the moment, a dynamometer upper computer control system can monitor dynamic indexes such as dynamometer states, engine torque and the like in real time; then setting the opening of a throttle valve through an upper computer of an engine controller, determining the operation condition, and detecting various states of the exhaust temperature, the cooling water temperature and the like of the engine in real time; then, a switch for controlling the automatic calibration of the parameters is opened on the upper computer of the Mirco AutoBox rapid control prototype, and the control parameters are automatically calibrated; the combustion analyzer collects and calculates IMEP of the engine in real time in the whole process, and continuously sends IMEP information to the Mirco AutoBox rapid control prototype through CAN bus communication; after receiving an instruction for starting the automatic calibration process of the control parameters, the Mirco AutoBox rapid control prototype receives IMEP sent by the combustion analyzer, and calculates to obtain average indicated torque as an engine dynamic index. And receiving the oil injection pulse width sent by the engine controller, and calculating the fuel consumption rate as an engine economic index. After the performance index of the engine is obtained, an optimization objective function under the control parameter at the moment is solved, the control parameter is changed through an optimization algorithm, and a new control parameter is sent to an engine controller through a CAN bus; the engine controller continuously monitors and receives the control parameters sent by the Mirco AutoBox rapid control prototype on the CAN bus to replace the control parameters sent by the upper computer of the engine controller, and after new control parameters are obtained, the engine controller executes the control actuator according to the control parameters, and the performance of the engine changes to carry out next optimization calculation.
In the method of this embodiment, a specific method for solving the optimization objective function under the control parameters and changing the control parameters through the optimization algorithm is as follows: the established optimization algorithm is 'genetic algorithm + extremum search algorithm', the genetic algorithm is used for searching the optimal area, and after the optimal area is searched, the extremum search algorithm is used for searching the optimal point.
Firstly, a genetic algorithm is simply introduced, the genetic algorithm is a random search algorithm based on biological natural selection and a genetic mechanism, and excellent genes in a population are stored and evolved by adopting a natural selection rule and the genetic mechanism through three operations of selection, intersection and variation, so that the optimal population is gradually obtained through repeated iteration of multiple generations.
The basic operation process of the genetic algorithm is shown in fig. 4, for the problem of online calibration of control parameters, any combination of working parameter combinations (ignition advance angle, air-fuel ratio) to be optimized in a parameter feasible region can be called an individual, N individuals are randomly selected to form a population with the scale of N, each parameter (such as the ignition advance angle) in the individual can be called a gene, the process of parameter optimization is one or more of gene values (ignition advance angle, air-fuel ratio) of valve parameters which are continuously crossed and varied, and the individual with improved target performance is continuously selected to be stored for next crossing, variation and selection until the target performance reaches the optimal convergence condition.
As shown in FIG. 5, extremum search algorithm extremum search is a type of adaptive control method for online steady state optimization of unknown dynamic systems. Most are based on estimating the gradient of the input-output mapping and using this estimate to move the objective function towards a minimum by controlling the input quantities. The extremum searching algorithm is suitable for solving the problem that the minimum value point is the global optimal point.
The optimization algorithm steps for the "genetic algorithm + extremum search algorithm" proposed in this embodiment are generally as follows:
1. the control parameters to be optimized are first determined, which are relevant to the study object and the study content. As can be seen from the bus communication diagram and the flow chart, the optimized control parameters in the embodiment are the ignition advance angle and the air-fuel ratio;
2. determining an optimization target and establishing an optimization target function;
3. solving an optimization objective function by using an optimization algorithm to obtain optimal control parameters;
4. and issuing the optimal control parameters to an engine controller, and operating according to the optimal control parameters under the working condition.
The optimization target determined in the embodiment is to optimize the engine dynamic performance index and the economic performance index, wherein the dynamic performance index is indicated torque, and the economic performance index is fuel consumption rate, and the importance degrees of the engine dynamic performance index and the economic performance index are considered to be equivalent. An optimization objective function is determined according to the optimization objective, which is expressed by the following formula:
wherein J (X) is an optimization objective function value; x is a control parameter group, in this embodiment, [ ignition advance angle, air-fuel ratio ]; be is the fuel consumption rate, unit g/(kWh); te is indicated torque in Nm; bemax is the set minimum fuel consumption rate and is a constant, and the unit g/(kWh) is set according to the actual economic performance of the engine; temax is the set maximum indicated torque, is a constant, and is set in Nm from the actual engine performance.
By the optimization objective function shown in the formula (1), the indicated torque and the fuel consumption rate of the engine can simultaneously reach an optimization interval, the minimum value of the weighted and integrated optimization function of the indicated torque and the fuel consumption rate of the engine is expected to be searched, and the corresponding control parameter is the minimum value. Meanwhile, the fuel consumption rate and the indicated torque are normalized, and the fact that the actual weights of the fuel consumption rate and the indicated torque are equivalent is guaranteed. Therefore, as can be seen from the above equation, the value of the optimization objective function becomes smaller as the indicated torque becomes larger and the fuel consumption rate becomes lower.
The specific fuel consumption and the indicated torque of equation (1) are calculated, and are not directly collected by a sensor, which will be described below.
Specific fuel consumption is the mass of fuel consumed in 1 hour per 1kw of available power delivered by the engine. However, in the actual calibration process, the average effective pressure (BMEP) and the effective torque of the engine cannot be directly acquired or obtained, and only the average indicated pressure (IMEP) and the indicated torque can be acquired or obtained, so that the effective power cannot be obtained, and therefore, the fuel consumption rate be is changed into the fuel quality consumed in 1 hour when the engine sends out 1kw of indicated power. Therefore, the fuel consumption be is calculated as follows:
wherein be is the fuel consumption rate and the unit g/(kWh); be is the actual fuel consumption of the engine in 1 hour under fixed working conditions and control parameters, and the unit is g/(kWh); pe is the indicated power in kW.
The actual fuel consumption Be of the engine is obtained by calibrating the fuel injection rule of the fuel injector, and is shown as the following formula:
Be=1.32n·t-0.68n (3)
in the formula, n is the engine speed and the unit rpm; and t is the oil injection pulse width of the oil injector in unit ms.
The indicated power Pe is calculated as follows:
wherein Pe is the indicated power in kW; pe is the average indicated pressure in MPa; vs is the working volume of a single cylinder, in L; n is the rotation speed, unit rpm; i is the number of cylinders; τ is the number of strokes. Vs, n, i and τ in equation (1) are all constants for an engine operating at a fixed operating condition. Therefore, the indicated power can be calculated by acquiring the average indicated pressure. While the collection of the average indicated pressure is dependent on the combustion analyzer system, as will be described later.
The indicated torque can be calculated by the indicated power in the formula (4), and both can be used as the engine dynamic index to describe the engine dynamic.
Wherein Te is engine indicated torque in Nm; pe is the indicated power, in kW; n is the engine speed in rpm.
The average indicated pressure is collected by depending on a combustion analyzer system, and the cycle number of the collection is also adapted to the working condition, so that a multi-cycle processing method of the average indicated pressure is provided, and the following description is provided.
The ignition engine has cyclic variation during combustion, the result of each cycle is different, in order to obtain representative data, according to experience, the average value of 40-100 cycles can be taken for small combustion cyclic variation and good combustion process repeatability, and when the combustion cyclic variation is large, the average value of hundreds of cycles can be taken. Whereas the evaluation of combustion stability is generally characterized by a mean indicated pressure coefficient of variation (CoV):
where, σ is the standard deviation of the mean indicated pressure,is the average of the average indicated pressure over several cycles.
Generally, when the CoV is not more than 10%, the combustion is considered to be stable and the cycle fluctuation is small. As can be seen from equation 6, σ is ANDed when the fixed CoV is 10%Is related toRepresents a change in the engine operating conditions and is therefore dependent onFrom the above, table 1 can be obtained.
TABLE 110% CoV Standard deviation and variance of mean indicated pressures for different IMEP means
For a combustion analyzer, the average indicated pressure for a single cycle, and the variance of the average indicated pressure, may be calculated. It can be seen from the above table that for example an IMEP with an average value of 2bar corresponds to a variance of 0.04 for 10% CoV. If the variance of the obtained average indicated pressure is below 0.04 under the working condition that the IMEP is about 2bar, the combustion process is considered to be stable, the cyclic variation is small, and the number of the collected cycles is small; if it exceeds 0.04, the combustion process cycle varies greatly, and more cycles need to be collected. The more stable the combustion process, the fewer the number of cycles collected, and conversely, the more unstable the number of cycles collected.
When the CoV is 10%, the combustion process is stable; multiple tests show that when the CoV is 20%, the combustion process is unstable, the cycle variation is large, the engine has obvious surge phenomenon, and control parameters are changed in time or the engine is stopped to ensure the stable operation of the engine and the safety of equipment, so that the CoV of 20% is considered as the lowest boundary of unstable combustion. The combustion analyzer can continuously collect IMEP of at most 400 cycles once, and simultaneously, the combustion analyzer can collect at least 40 cycles even if the combustion is stable, so that the collection cycle number is determined to be 40-400 cycles. However, since the CoV is not equal to 0, it is considered that the CoV of 1% is the highest boundary of combustion stability. When the variance of IMEP calculated by the combustion analyzer is just the variance corresponding to 1% CoV, the collection cycle number at the moment is given to be 40; when the variance of IMEP calculated by the combustion analyzer is just the variance corresponding to 10% CoV, the collection cycle number at the moment is given as 100; when the variance of IMEP calculated by the combustion analyzer was exactly the variance corresponding to 20% CoV, the number of acquisition cycles at that time was given as 400. When other IMEP variances are acquired, linear interpolation should be performed between the two intervals to obtain the number of cycles that it needs to acquire.
As shown in fig. 6, the number of cycles of IMEP required to be collected must be completed before the optimization algorithm can proceed, and it is determined how many cycles of IMEP averages to collect in the combustion analyzer. Therefore, the acquired IMEP average value can accurately represent the performance of the engine, and meanwhile, the number of acquired cycles is reduced as much as possible, and the calculation time is reduced.
After the optimization objective is determined, an optimization algorithm is introduced. The genetic algorithm is used for searching an optimal area in a feasible area of the control parameter, and after the optimal area is found, an extremum searching algorithm is used for searching an optimal point of an objective function to determine the optimal control parameter. The steps of the genetic algorithm are as follows:
1.1, determining the availability domain of the control parameter and dividing the step length. The optimized control parameters are the ignition advance angle and the air-fuel ratio, so that the feasible regions of the two parameters under a certain fixed working condition are determined firstly, and then the change step length of the control parameters is determined, for example, the change step length of the ignition advance angle is 5 ℃ A;
and 1.2, generating an initial population. The genetic algorithm aims to quickly find the optimal region, and the accuracy is low, so that the population number N and the optimization algebra i do not need to be too many. And simultaneously, combining the change step length of the control parameters set in the last step, randomly generating N individuals (control parameter combinations) and generating an initial population.
And 1.3, solving an optimization objective function. And sequentially sending the N control parameter combinations to an engine controller through a CAN bus, collecting relevant parameters such as engine response at the moment, and calculating the optimization objective function values under different control parameters. After the optimization objective functions of all individuals are solved, the optimization objective function values are sorted from low to high, and the optimal control parameters of the optimization objective and the optimization objective function values are recorded. The individual ones at 1/2 are taken and randomly divided into three groups (related to the number of control parameters to be optimized), one group for varying the first control parameter (spark advance), one group for varying the second control parameter (air-fuel ratio), and the last group for varying both control parameters simultaneously. Combining the varied individuals with the original individuals to generate a new population, and ensuring that the number of individuals of the population is N;
and 1.4, performing convergence judgment. After a new population is generated, solving the optimization objective function again, wherein the optimization objective function is consistent with the step 1.3, after the optimal control parameter and the optimal objective function value of the population optimization objective are obtained, the optimal control parameter and the optimal objective function value are compared with the optimal objective function value of the previous generation, the optimization of the genetic algorithm is introduced when the convergence condition is met, otherwise, a new population is generated according to the step 1.3, and the optimization is continued.
After the genetic algorithm is optimized and converged, the optimal individual (control parameter combination) meeting the convergence condition is obtained, the control parameter combination is used as the center and is expanded forwards and backwards, and the expanded region is used as a feasible region for extremum search to optimize the extremum search algorithm. It should also be emphasized that the extremum search algorithm is used to solve convex optimization problems generally, whereas non-convex optimization problems are easily converged to local optima using the extremum search algorithm. For the optimization, the optimization problem is changed into a non-convex optimization problem by simultaneously changing two control parameters, but in an optimal region, only one control parameter is changed at a time, the other control parameter is kept unchanged, and the two control parameters are alternately changed based on corresponding gradients, so that the optimization problem is changed into a convex optimization problem, and therefore, the optimization is carried out according to an alternate search mode. The steps of the extremum searching algorithm are as follows:
and 2.1, initializing algorithm parameters. The extremum search algorithm needs to calculate the gradient of the optimized objective function in the optimization process, the gradient calculation is calculated in the optimization by adopting a bilateral finite difference mode, and the calculation mode is shown as the following formula:
wherein gk is the gradient of the optimization objective function; f is an optimization objective function; xi is one of the control parameters; λ is the disturbance, where a constant is chosen; v is the direction of the disturbance and is randomly selected from the direction pool. In the last time, before optimization, a disturbance value lambda and a direction pool are required to be given, and initial control parameters are set according to an optimal region obtained by genetic algorithm optimization;
and 2.2, solving an optimized objective function under disturbance, and calculating a gradient. Firstly, keeping one of the control parameters unchanged, randomly selecting a disturbance direction from a vector pool, enabling the other control parameter to change along the disturbance direction, obtaining the engine response, and calculating an optimization objective function. And then the control parameters before change are changed along the reverse direction of the previous disturbance direction to obtain the engine response at the moment, and an optimization objective function is calculated. Finally, the obtained forward and backward optimization objective functions are used for calculating gradients;
and 2.3, updating the control parameters. After the gradient generated due to the change of the control parameter is obtained, the control parameter changed in the previous step is updated according to the following formula:
xi=xi+ai·v·gk(8)
in the formula, x-iIs a new control parameter of the previous step change; ai is the variation gain to adjust the step size; v is the perturbation direction of the previous step; gk is the gradient of the optimization objective function.
And 2.4, changing other control parameters, and repeating the steps 2.2 and 2.3. On the basis of the updated control parameters in the step 2.3, the changed control parameters in the previous two steps are kept unchanged, the gradient of the other control parameter is solved according to the method in the step 2.2, and the control parameters are updated according to the step 2.3.
And 2.5, performing convergence judgment. And obtaining the corresponding optimized objective function after the two control parameters are alternately updated, and judging the optimized objective function corresponding to the previous generation. And (4) meeting the convergence condition, outputting the control parameter when the current control parameter is the globally optimal control parameter, and otherwise, continuously performing iterative computation according to the step 2.2-2.5.
The extremum searching algorithm is high in searching speed, can quickly search the global optimal point in the optimal area, and completes automatic online parameter calibration.
After the extremum searching algorithm is converged, the Mirco AutoBox rapid control prototype building the optimization algorithm CAN continuously send the optimal control parameters to the engine controller through the CAN bus, and the controller CAN record the optimal control parameters and the current working condition. The rotating speed of the engine is changed through the dynamometer upper computer control system, the opening of a throttle valve is changed through an engine controller, the operation working condition of the engine is updated, and the control parameters are continuously subjected to automatic online optimization calibration under the current working condition.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.
Claims (11)
1. The utility model provides a device that combines genetic algorithm and extreme value search algorithm's online demarcation of engine control parameter, includes engine, dynamometer, engine controller, combustion analysis appearance, mirco AutoBox quick control prototype and dynamometer host computer control system, engine controller host computer, mirco AutoBox quick control prototype host computer and CAN bus communication system:
the engine and the dynamometer are connected through a coupling, and the dynamometer is used for providing starting rotating speed for starting the engine, fixing the rotating speed of the engine and measuring the torque of the engine;
the engine controller is used for controlling an actuator on the engine, receiving signals collected by the sensor and communicating with an upper computer of the engine controller through a CAN bus;
the combustion analyzer acquires in-cylinder pressure and a crank angle through a cylinder pressure sensor arranged in an engine cylinder and a corner mark instrument on a crank shaft, further calculates average indicating pressure of each cycle, communicates with a Mirco AutoBox rapid control prototype through a CAN bus, and sends information to the Mirco AutoBox rapid control prototype;
the Mirco AutoBox rapid control prototype is used for building an optimization algorithm, and calculating the next control parameter by collecting input quantity;
the dynamometer upper computer control system is used for controlling the dynamometer and monitoring the state of the dynamometer and the measured engine torque;
the upper computer of the engine controller is used for sending a control command to the engine controller so as to control the actuator; monitoring the engine state collected by the sensor; when the control parameters are automatically calibrated, the control parameters calculated by the Mirco AutoBox rapid control prototype are received, and then the control parameters are sent to an engine controller;
the Mirco AutoBox rapid control prototype upper computer is used for monitoring input and output quantity during calculation of an optimization algorithm, and is also provided with an automatic calibration switch for controlling the starting and ending of an automatic calibration process;
the CAN bus communication system consists of two CAN buses, wherein one CAN bus is connected with a Mirco AutoBox rapid control prototype and a combustion analyzer, and the combustion analyzer sends the IMEP obtained by calculation to the Mirco AutoBox rapid control prototype; the other CAN bus is connected with the engine controller, an upper computer of the engine controller and the Mirco AutoBox rapid control prototype, and the engine controller sends signals acquired by the sensor to the upper computer of the engine controller through the CAN bus for monitoring the state of the engine; the upper computer of the engine controller issues the control parameters to the engine controller through a CAN bus, and the engine controller controls the actuator to execute according to the parameters; the Mirco AutoBox rapid control prototype monitors oil injection pulse width sent by an engine controller on a CAN bus to calculate fuel consumption rate, and simultaneously sends corresponding control parameters to the engine controller in the automatic online parameter calibration process to replace the same control parameters sent by an upper computer of the engine controller.
2. The apparatus of claim 1, wherein the combustion analyzer calculates the average value of a plurality of cycles to be collected as the engine response.
3. The device for online calibration of the engine control parameters by combining the genetic algorithm and the extremum searching algorithm as claimed in claim 2, wherein the number of the collected cycles is 40-400 cycles.
4. A method for online calibration of engine control parameters by combining a genetic algorithm and an extremum search algorithm comprises the following steps:
A. setting an engine starting rotating speed by an upper computer control system of the dynamometer, and presetting a rotating speed of stable operation of the engine;
B. starting, the engine stably runs at a preset running rotating speed, and at the moment, a dynamometer upper computer control system monitors dynamic indexes such as dynamometer states, engine torque and the like in real time; then setting the opening of a throttle valve through an upper computer of an engine controller, determining the operation condition, and detecting the exhaust temperature and the cooling water temperature state of the engine in real time;
C. opening a switch for controlling automatic calibration of parameters on a Mirco AutoBox rapid control prototype upper computer to automatically calibrate the control parameters;
D. the combustion analyzer collects and calculates the average indicated pressure of the engine in real time in the whole process, communicates through a CAN bus, and continuously sends the information of the average indicated pressure to the Mirco AutoBox rapid control prototype;
E. after receiving an instruction for starting the automatic calibration process of the control parameters, the Mirco AutoBox rapid control prototype receives average indicated pressure sent by a combustion analyzer, and calculates to obtain average indicated torque as an engine dynamic index;
F. receiving an oil injection pulse width sent by an engine controller, and calculating the fuel consumption rate as an engine economic index;
G. after the performance index of the engine is obtained, an optimization objective function under the control parameter at the moment is solved, the control parameter is changed through an optimization algorithm, and a new control parameter is sent to an engine controller through a CAN bus;
H. the engine controller continuously monitors and receives the control parameters sent by the Mirco AutoBox rapid control prototype on the CAN bus to replace the control parameters sent by the upper computer of the engine controller, and after new control parameters are obtained, the engine controller executes the control actuator according to the control parameters, and the performance of the engine changes to carry out next optimization calculation.
5. The method for on-line calibration of engine control parameters in combination with genetic algorithm and extremum seeking algorithm as claimed in claim 4, wherein said step G, the specific method for changing control parameters by optimizing algorithm comprises:
g1, first determining the control parameters to be optimized: ignition advance angle and air-fuel ratio;
g2, determining an optimization target and establishing an optimization target function;
g3, solving the optimization objective function by using an optimization algorithm to obtain the optimal control parameters, wherein the optimization algorithm specifically comprises the following steps: searching an optimal region in a feasible region of the control parameter by using a genetic algorithm, and searching an optimal point of an objective function by using an extremum searching algorithm after the optimal region is found to determine an optimal control parameter;
g4, issuing the optimal control parameters to an engine controller, and operating according to the optimal control parameters under the working condition.
6. The method for online calibration of engine control parameters combining genetic algorithm and extremum seeking algorithm of claim 5, wherein said step G2 comprises the steps of:
determining an optimization target to optimize engine dynamic performance and economic performance indexes, and simultaneously considering that the engine dynamic performance and the economic performance indexes have the same importance degree, wherein the dynamic performance indexes are indication torque, and the economic performance indexes are fuel consumption rate; an optimization objective function shown by the following formula is determined according to the optimization objective:
wherein J (X) is an optimization objective function value; x is a control parameter group; be is the fuel consumption rate, unit g/(kWh); te is indicated torque in Nm; bemax is the set minimum fuel consumption rate and is a constant, and the unit g/(kWh) is set according to the actual economic performance of the engine; temax is the set maximum indicated torque, is a constant, and is set in Nm from the actual engine performance.
7. The method for on-line calibration of engine control parameters combining genetic algorithm and extremum seeking algorithm of claim 6 wherein X in said formula is [ spark advance, air-fuel ratio ].
8. The method for online calibration of engine control parameters combining genetic algorithm and extremum seeking algorithm of claim 5, wherein in step G3, the specific steps of the genetic algorithm include:
g311, determining the availability domain of the control parameter and dividing the step length;
g312, generating an initial population;
g313, solving an optimization objective function;
g314, convergence judgment is carried out.
9. The method for on-line calibration of engine control parameters according to claim 5, wherein in step G3, the specific steps of the extremum searching algorithm include:
g321, initializing algorithm parameters;
g322, solving the optimization objective function under disturbance, and calculating the gradient.
G323, updating the control parameters.
G324, changing other control parameters, and repeating the steps G322 and G323;
g325, convergence judgment is carried out.
10. The method for online calibration of engine control parameters combining genetic algorithm and extremum seeking algorithm of claim 4, wherein in step D, the combustion analyzer collects and calculates the average indicated pressure of the engine, and it needs to collect the average value of N cycles as the engine response, wherein the evaluation for determining the combustion stability is characterized by the average indicated pressure variation coefficient (CoV) and the following formula:
11. The method for online calibration of engine control parameters combining genetic algorithm and extremum seeking algorithm of claim 10, wherein the CoV ranges from 1% to 20%, and the number of cycles N collected ranges from 40 to 400.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010094829.8A CN111337258B (en) | 2020-02-14 | 2020-02-14 | Device and method for online calibration of engine control parameters by combining genetic algorithm and extremum search algorithm |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010094829.8A CN111337258B (en) | 2020-02-14 | 2020-02-14 | Device and method for online calibration of engine control parameters by combining genetic algorithm and extremum search algorithm |
Publications (2)
Publication Number | Publication Date |
---|---|
CN111337258A true CN111337258A (en) | 2020-06-26 |
CN111337258B CN111337258B (en) | 2021-03-23 |
Family
ID=71180605
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010094829.8A Expired - Fee Related CN111337258B (en) | 2020-02-14 | 2020-02-14 | Device and method for online calibration of engine control parameters by combining genetic algorithm and extremum search algorithm |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111337258B (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111650171A (en) * | 2020-07-03 | 2020-09-11 | 中国科学院工程热物理研究所 | Quantitative measurement and correction method for high-temperature and high-pressure fuel concentration field of fuel nozzle |
CN112182779A (en) * | 2020-10-29 | 2021-01-05 | 同济大学 | Gasoline engine online calibration method based on simulated annealing 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 |
CN113687643A (en) * | 2021-08-31 | 2021-11-23 | 重庆长安新能源汽车科技有限公司 | Motor active short circuit function test system and method for electric automobile |
CN113790838A (en) * | 2021-09-02 | 2021-12-14 | 厦门大学 | Method for measuring torque of rotating shaft of engineering machinery |
CN115114785A (en) * | 2022-06-28 | 2022-09-27 | 北京理工大学 | Engine performance index parameterization rapid prediction method |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102507195A (en) * | 2011-12-01 | 2012-06-20 | 北京理工大学 | Portable engine data standardization recording method and device |
CN103994005A (en) * | 2014-04-13 | 2014-08-20 | 北京理工大学 | Fuel economy calibrating method of electronic control monoblock pump diesel engine model |
US20160131068A1 (en) * | 2014-11-10 | 2016-05-12 | Caterpillar Inc. | Engine system utilizing modal weighted engine optimization |
CN106089541A (en) * | 2016-06-14 | 2016-11-09 | 吉林大学 | Free-piston engine control system ignition location optimization method based on extremum search |
CN108647389A (en) * | 2018-06-12 | 2018-10-12 | 中国北方发动机研究所(天津) | A kind of determination method of control parameters of engine optimization MAP |
CN108918148A (en) * | 2018-07-20 | 2018-11-30 | 上汽通用五菱汽车股份有限公司 | Scaling method, system and the computer readable storage medium of engine charge model |
CN110348136A (en) * | 2019-07-15 | 2019-10-18 | 吉林师范大学 | A kind of engine torque and discharge modeling method based on parameter calibration optimization |
-
2020
- 2020-02-14 CN CN202010094829.8A patent/CN111337258B/en not_active Expired - Fee Related
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102507195A (en) * | 2011-12-01 | 2012-06-20 | 北京理工大学 | Portable engine data standardization recording method and device |
CN103994005A (en) * | 2014-04-13 | 2014-08-20 | 北京理工大学 | Fuel economy calibrating method of electronic control monoblock pump diesel engine model |
US20160131068A1 (en) * | 2014-11-10 | 2016-05-12 | Caterpillar Inc. | Engine system utilizing modal weighted engine optimization |
CN106089541A (en) * | 2016-06-14 | 2016-11-09 | 吉林大学 | Free-piston engine control system ignition location optimization method based on extremum search |
CN108647389A (en) * | 2018-06-12 | 2018-10-12 | 中国北方发动机研究所(天津) | A kind of determination method of control parameters of engine optimization MAP |
CN108918148A (en) * | 2018-07-20 | 2018-11-30 | 上汽通用五菱汽车股份有限公司 | Scaling method, system and the computer readable storage medium of engine charge model |
CN110348136A (en) * | 2019-07-15 | 2019-10-18 | 吉林师范大学 | A kind of engine torque and discharge modeling method based on parameter calibration optimization |
Non-Patent Citations (1)
Title |
---|
甘俊英: ""基于遗传算法的非线性控制***参数优化方法"", 《工业仪表与自动化装置》 * |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111650171A (en) * | 2020-07-03 | 2020-09-11 | 中国科学院工程热物理研究所 | Quantitative measurement and correction method for high-temperature and high-pressure fuel concentration field of fuel nozzle |
CN111650171B (en) * | 2020-07-03 | 2023-08-18 | 中国科学院工程热物理研究所 | Quantitative measurement and correction method for high-temperature high-pressure fuel concentration field of fuel nozzle |
CN112182779A (en) * | 2020-10-29 | 2021-01-05 | 同济大学 | Gasoline engine online calibration method based on simulated annealing algorithm |
CN112182779B (en) * | 2020-10-29 | 2023-08-29 | 同济大学 | Gasoline engine online calibration method based on simulated annealing 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 |
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 |
CN113687643A (en) * | 2021-08-31 | 2021-11-23 | 重庆长安新能源汽车科技有限公司 | Motor active short circuit function test system and method for electric automobile |
CN113790838A (en) * | 2021-09-02 | 2021-12-14 | 厦门大学 | Method for measuring torque of rotating shaft of engineering machinery |
CN115114785A (en) * | 2022-06-28 | 2022-09-27 | 北京理工大学 | Engine performance index parameterization rapid prediction method |
CN115114785B (en) * | 2022-06-28 | 2024-01-23 | 北京理工大学 | Engine performance index parameterization rapid prediction method |
Also Published As
Publication number | Publication date |
---|---|
CN111337258B (en) | 2021-03-23 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111337258B (en) | Device and method for online calibration of engine control parameters by combining genetic algorithm and extremum search algorithm | |
US11436395B2 (en) | Method for prediction of key performance parameter of an aero-engine transition state acceleration process based on space reconstruction | |
US20190019096A1 (en) | Estimator, estimation method, program and storage medium where program stored for model parameter estimation and model parameter estimation system | |
CN113110367B (en) | Engine hardware in-loop test method and system | |
KR102206324B1 (en) | Methods for ascertaining a model of a starting variable of a technical system | |
CN104408271A (en) | Model-based gasoline engine calibration method | |
CN114462319B (en) | Active regulation and control method for combustion performance of aero-engine and intelligent prediction model | |
JP2003013794A (en) | Adaptation method of control parameter for vehicle and adaptation device | |
CN107989708A (en) | Self study engine torque control system and its method based on disturbance observation | |
Vogt et al. | On-line adaptation of grid-based look-up tables using a fast linear regression technique | |
CN112146888A (en) | Parameter calibration method, device, equipment and storage medium | |
EP3244043A1 (en) | Apparatus and method for controlling egr valve | |
CN113049259B (en) | Fuzzy control method of rack control system, storage medium and equipment | |
CN113759709A (en) | Method and device for training strategy model, electronic equipment and storage medium | |
CN110909492B (en) | Sewage treatment process soft measurement method based on extreme gradient lifting algorithm | |
CN112163632A (en) | Application of semi-supervised extreme learning machine based on bat algorithm in industrial detection | |
CN114934848B (en) | Fuzzy neural network modeling method for optimizing control of combustion performance of diesel engine | |
CN112182779B (en) | Gasoline engine online calibration method based on simulated annealing algorithm | |
CN112883653B (en) | Artificial intelligence-based modeling method for real-time engine model | |
CN111255557B (en) | Automobile engine cooling detection system and control method thereof | |
CN111210877A (en) | Method and device for deducing physical property parameters | |
CN110080896A (en) | A kind of methane fuelled engine air/fuel ratio control method based on genetic algorithm | |
CN109358511A (en) | A kind of system core performance index adaptive regulation method of data-driven | |
CN114511138A (en) | Steam quantity prediction method and system based on width learning and regression model | |
CN112699600B (en) | Thermal power operating parameters and NOxMethod for analyzing partial return between emission concentrations |
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 | ||
CF01 | Termination of patent right due to non-payment of annual fee | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20210323 |