CN109634107B - An optimization method of engine dynamic control law - Google Patents
An optimization method of engine dynamic control law Download PDFInfo
- Publication number
- CN109634107B CN109634107B CN201910055751.6A CN201910055751A CN109634107B CN 109634107 B CN109634107 B CN 109634107B CN 201910055751 A CN201910055751 A CN 201910055751A CN 109634107 B CN109634107 B CN 109634107B
- Authority
- CN
- China
- Prior art keywords
- sub
- optimization
- population
- control law
- point
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000005457 optimization Methods 0.000 title claims abstract description 123
- 238000000034 method Methods 0.000 title claims abstract description 58
- 230000001133 acceleration Effects 0.000 claims description 12
- 230000006835 compression Effects 0.000 claims description 11
- 238000007906 compression Methods 0.000 claims description 11
- 238000002485 combustion reaction Methods 0.000 claims description 9
- 238000000354 decomposition reaction Methods 0.000 claims description 9
- 230000035772 mutation Effects 0.000 claims description 4
- 238000012216 screening Methods 0.000 claims description 4
- 238000012163 sequencing technique Methods 0.000 claims description 4
- 230000007704 transition Effects 0.000 claims 2
- 238000012804 iterative process Methods 0.000 abstract 1
- 239000000446 fuel Substances 0.000 description 4
- 239000003921 oil Substances 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 3
- 238000012938 design process Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000012886 linear function Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 1
- 230000008014 freezing Effects 0.000 description 1
- 238000007710 freezing Methods 0.000 description 1
- 239000000295 fuel oil Substances 0.000 description 1
- 239000007921 spray Substances 0.000 description 1
Images
Classifications
-
- 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
- G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
- G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
- G05B13/0205—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric not using a model or a simulator of the controlled system
- G05B13/024—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric not using a model or a simulator of the controlled system in which a parameter or coefficient is automatically adjusted to optimise the performance
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Feedback Control In General (AREA)
- Combined Controls Of Internal Combustion Engines (AREA)
Abstract
本发明公开了一种发动机动态控制规律优化方法，按照控制规律自变量参数将发动机动态控制规律离散为数个子控制规律点。针对除起点和终点以外的每个子控制规律点建立相应的子种群进行优化，并在每步优化结束后选取最优个体作为该子种群的精英个体，冻结此子种群优化进程进行下一子种群优化，如此循环逐点优化直至优化进程满足终止条件。应用本发明的发动机动态控制规律优化方法，可以克服传统动态控制规律设计方法不适用于先进多可调部件的航空发动机动态控制规律设计问题，解决传统方法受人为因素影响大、需要往复迭代过程并且无法得到最优控制规律的问题。
The invention discloses an optimization method for an engine dynamic control law, which discretizes the engine dynamic control law into several sub-control law points according to independent variable parameters of the control law. For each sub-control rule point except the starting point and the end point, a corresponding sub-population is established for optimization, and after each step of optimization is completed, the optimal individual is selected as the elite individual of the sub-population, and the optimization process of this sub-population is frozen for the next sub-population. Optimization, and this cycle is optimized point by point until the optimization process satisfies the termination condition. The application of the engine dynamic control law optimization method of the present invention can overcome the problem that the traditional dynamic control law design method is not suitable for the design of the aero-engine dynamic control law of advanced multi-adjustable components, and solves the traditional method which is greatly affected by human factors, requires a reciprocating iterative process and The problem that the optimal control law cannot be obtained.
Description
Technical Field
The invention relates to the field of aircraft engines, in particular to an optimization method for a dynamic control rule of an engine.
Background
For an aircraft engine, optimization aiming at an engine dynamic control rule has important significance on final performance of the engine. And as the number of adjustable components of the engine is increased, the control law of the engine is not the traditional change law of pure fuel or rotating speed along with time or other independent variables, but a plurality of adjustable components are often adjusted together. Taking an engine acceleration control law optimization as an example, in the process of engine acceleration, not only the change law of fuel oil along with time or other independent variables of the rotating speed needs to be given, but also parameters such as the minimum cross-sectional area of a spray pipe, the guide vane angle of a fan guider, the critical cross-sectional area of a turbine guider and the like need to be adjusted, namely the change law of a plurality of control parameters needs to be given at the same time. The traditional design method of the engine control law is usually designed point by engineering personnel from a starting point to an end point, and the method cannot carefully consider the influence between all points of the control law, for example, aiming at the design problem of the acceleration state control law, the starting point and the end point steady state points of the control law can generate constraint on the design of the control law. On the one hand, the design objective of a general control law requires that the engine starts from the start of the control law to boost the rotor acceleration power as fast as possible; on the other hand, the presence of the end point requires that the accelerating power be reduced early so that it does not overrun at the end point. The constraints of the starting point and the end point cannot be simultaneously and accurately considered in the point-by-point design process, the lifting amount of the rotor acceleration power compared with the previous point on each sub-control rule point and the time when the acceleration power starts to decrease are solved by most of the problems according to self experience of engineering personnel, and reciprocating design is often needed to enable the control rule to meet various limits, and the designed control rule cannot be guaranteed to be optimal. However, the traditional engine control rule optimization method can only optimize a single-point steady-state control rule and cannot be directly applied to the design of a dynamic control rule. With the increasing of controllable parameters of the engine at present, a new optimization method for the dynamic control law of the aero-engine needs to be developed, so that the optimal dynamic control law of the engine is obtained on the premise of ensuring the calculation convergence and the calculation speed.
Disclosure of Invention
The technical problem solved by the invention is as follows: the invention relates to an engine control rule optimization method, which aims to solve the problems that in the prior art, the design process of an engine dynamic control rule is complex, the design problem of the engine control rule for multiple adjustable components is poor in feasibility, and the obtained control rule is not optimal.
The technical scheme of the invention is as follows: an engine control law optimization method comprises the following steps:
the method comprises the following steps: assuming that the initial steady state point of the engine in the dynamic process is A and the final steady state point is B, the optimization objects are controllable parameters and the change rule of the adjustable component adjustment quantity from the A to the B dynamic state, wherein a plurality of adjustment parameters and engine performance parameters comprise high and low pressure rotor rotating speed, the section area of the throat part of a tail nozzle, the inlet guide vane angle of a compression component, the section area of an inlet guider of a turbine component, the oil supply quantity of a combustion chamber and the temperature before the turbine;
step two: determining the design target from A to B and the constraint condition required to be ensured by the engine in the dynamic process part or the complete machine parameter; the design target is that the time required by the dynamic process from A to B is shortest, the constraint parameters are that the surge margin of a compression part is not higher than a threshold value, the high-low pressure rotating speed acceleration rate is not higher than a threshold value, the change rate of the fuel-air ratio of a combustion chamber is not higher than a threshold value, and the threshold values of the three are different and unequal.
Step three: establishing an initial control law, wherein the initial control law can be set according to the following formula:
X＝(XB-XA)×t/T
wherein T is the independent variable time in the dynamic control rule, T is the total time of the dynamic control rule, X is the vector formed by the parameters in the step one, and X is the vector formed by the parameters in the step oneBVector of control parameters, X, for the final steady-state pointAA control parameter vector which is a starting steady-state point;
step four: dispersing the control law established in the third step into m sub-control law points according to the independent variable, wherein the m sub-control law points do not comprise a point A and a point B;
step five, aiming at the m sub-control rule points obtained by decomposition in the step four, establishing corresponding m evolutionary optimization sub-populations with the size of n;
and step six, establishing an elite individual set. Randomly selecting an individual from each evolution optimization sub-population as an elite individual of the sub-population, wherein the elite individual set is a set of elite individuals of the m evolution optimization sub-populations. It should be noted that any individual in each evolutionary optimization sub-population corresponds to a sub-control law point in a dynamic control law, and a set formed by these individuals corresponds to a dynamic control law, that is, an elite individual set also corresponds to a dynamic control law.
And step seven, initializing a control rule overall sub-population point-by-point optimization process. Taking i as 1, wherein i represents the evolutionarily optimized sub-population number to be operated in the current and subsequent steps.
And step eight, performing one-step optimization iteration on the ith evolutionary optimization sub-population. And (3) operating the current n individuals in the ith evolutionary optimization sub-population, namely parent individuals, by using mutation and crossover operators of a differential evolution algorithm to generate n sub-population.
And step nine, calculating the fitness of each individual in the ith evolutionary optimization sub-population.
Step ten, screening the ith evolutionary optimization sub-population. And D, according to the fitness calculated in the ninth step, carrying out fitness sequencing from small to large on n individuals in the ith evolutionary optimization sub-population, only reserving the first n individuals in the current evolutionary optimization sub-population, and discarding the rest individuals.
Step eleven, updating the elite individual set. And updating the elite individuals of the ith evolutionary optimization sub-population into the individuals with the minimum fitness in the sub-population.
And step twelve, judging whether the one-time point-by-point optimization cycle is finished. if i is equal to i +1, go to step thirteen if i > m, otherwise go to step eight.
And step thirteen, judging whether the control rule optimization process is finished. And if the optimization process reaches the end condition, turning to the step fourteen, otherwise, turning to the step eight, wherein i is equal to 1.
And step fourteen, the control law corresponding to the current elite individual set is the optimization result of the dynamic control law of the engine.
Effects of the invention
The invention has the technical effects that: by applying the method for optimizing the dynamic control rule of the engine, the problem that the traditional method for designing the dynamic control rule is not suitable for designing the dynamic control rule of the advanced aeroengine with multiple adjustable components can be solved, and the problems that the traditional method is greatly influenced by human factors, needs a reciprocating iteration process and cannot obtain the optimal control rule can be solved.
Drawings
FIG. 1 is a schematic flow chart of an alternative method for optimizing engine dynamic control laws according to an embodiment of the present disclosure;
FIG. 2 is a schematic diagram illustrating a flow of optimizing each sub-control law point in an optional method for optimizing a dynamic engine control law according to an embodiment of the present invention;
wherein the following parameters are included in the figures: i: the sequence number of the current evolutionary optimized sub-population; m: and the total number of sub-control rules obtained by the decomposition of the control rules, namely the total number of corresponding evolutionary optimization sub-populations.
Detailed Description
Referring to fig. 1-2, in order to make the technical solution of the present invention better understood, the technical solution of the embodiment of the present invention will be clearly and completely described below with reference to the drawings in the embodiment of the present invention, and it is obvious that the described embodiment is only a part of the embodiment of the present invention, but not all of the embodiment. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Firstly, the optimization object of the method is a dynamic control rule between any two steady-state working points of the engine. Therefore, the adjustment parameters of each component of the start steady state point A and the final steady state point B of the dynamic process of the engine to be optimized and the performance parameters of the engine need to be obtained firstly. Taking the dynamic control rule optimization from the ground slow vehicle to the ground takeoff of the turbofan engine as an example, determining the start and final steady state points of the dynamic process of the engine is to obtain the adjustment parameters of each component of the ground slow vehicle steady state point and the ground takeoff steady state point of the turbofan engine according to the design result of the steady state performance of the turbofan engine. The method comprises the following steps: high and low pressure rotor speed, nozzle throat section area, compression component inlet guide vane angle, turbine component inlet guide vane section area, combustion chamber oil supply, variable area guide vane adjustment, and compression component surge margin.
And step two, determining a dynamic control rule optimization target and constraint of the engine from a steady-state point A to a steady-state point B. Taking the optimization of the dynamic control rule from the ground slow vehicle to the ground takeoff of the turbofan engine as an example, the design target is that the time required by the whole dynamic state is shortest, namely the engine is transited from the steady state point A to the steady state point B as fast as possible. And the constraint conditions are that in the whole dynamic state, the surge margin of each compression part of the engine is not higher than a threshold value, the high-low pressure rotating speed acceleration rate is not higher than the threshold value, the combustion chamber oil-gas ratio change rate is not higher than the threshold value, and the threshold values of the three are different and unequal.
And step three, establishing an initial control rule. The general initial control law can be simply set as a linear function of the control parameter and the argument.
And step four, the control law established in the decomposition step three is m sub-control law points, wherein m represents the decomposed sub-control law point number. It should be noted that the m sub-control law points obtained by decomposition do not include dynamic start and end steady-state points, which are the premise and constraint of control law optimization.
And step five, establishing m corresponding sub-population evolutionarily optimized with the size of n according to the m sub-control rule points obtained by decomposition in the step four.
And step six, establishing an elite individual set. Randomly selecting an individual from each evolution optimization sub-population as an elite individual of the sub-population, wherein the elite individual set is a set of elite individuals of the m evolution optimization sub-populations. It should be noted that any individual in each evolutionary optimization sub-population corresponds to a sub-control law point in a dynamic control law, and a set formed by these individuals corresponds to a dynamic control law, that is, an elite individual set also corresponds to a dynamic control law.
And step seven, initializing a control rule overall sub-population point-by-point optimization process. Taking i as 1, wherein i represents the evolutionarily optimized sub-population number to be operated in the current and subsequent steps.
And step eight, performing one-step optimization iteration on the ith evolutionary optimization sub-population. And (3) operating the current n individuals in the ith evolutionary optimization sub-population, namely parent individuals, by using mutation and crossover operators of a differential evolution algorithm to generate n sub-population.
And step nine, calculating the fitness of each individual in the ith evolutionary optimization sub-population.
Step ten, screening the ith evolutionary optimization sub-population. And D, according to the fitness calculated in the ninth step, carrying out fitness sequencing from small to large on n individuals in the ith evolutionary optimization sub-population, only reserving the first n individuals in the current evolutionary optimization sub-population, and discarding the rest individuals.
Step eleven, updating the elite individual set. And updating the elite individuals of the ith evolutionary optimization sub-population into the individuals with the minimum fitness in the sub-population.
And step twelve, judging whether the one-time point-by-point optimization cycle is finished. if i is equal to i +1, go to step thirteen if i > m, otherwise go to step eight.
And step thirteen, judging whether the control rule optimization process is finished. And if the optimization process reaches the end condition, turning to the step fourteen, otherwise, turning to the step eight, wherein i is equal to 1.
And step fourteen, the control law corresponding to the current elite individual set is the optimization result of the dynamic control law of the engine.
And (3) dispersing the dynamic control rule of the engine into a plurality of sub-control rule points according to the time step or other independent variable parameters. And establishing a corresponding sub-population for optimization aiming at each sub-control rule point except the starting point and the end point, selecting the optimal individual as an elite individual of the sub-population after the optimization of each step is finished, freezing the optimization process of the sub-population to perform the next sub-population optimization, and circularly optimizing point by point until the optimization process meets the termination condition. The sub-population optimization target is set as each sub-population elite individual, namely each sub-control rule point, the engine target performance parameter obtained by the dynamic control rule formed by combining the sub-population elite individual points is optimal, and the limitation condition of the sub-population optimization can be obtained by decomposing the design limitation condition of the dynamic control rule of the engine.
Further, discretizing the steady-state or dynamic control law in an independent variable space to form discrete control law points, and establishing corresponding sub-populations for each control law point to optimize respectively.
Further, after each optimization of the sub-populations is finished, the optimal individual of the current sub-population is selected as the elite individual of the sub-population, the objective function of the sub-population optimization is set as the elite individual of each sub-population, namely each optimal sub-control rule point, the engine performance parameter obtained by the combined control rule is optimal, and the constraint condition of the sub-population optimization is obtained by decomposing the constraint condition of the control rule.
And further, updating the elite individual set immediately after acquiring the elite individual of each sub-population optimization process every step, pausing the current sub-population optimization process, and performing next sub-population optimization. And when all the sub-populations are optimized in one step, acquiring the elite individuals after the optimization is finished, updating the set of the elite individuals, and continuing the optimization from the first sub-population until the end condition is met.
A specific embodiment of the inventive scheme comprises the following steps.
Firstly, the optimization object of the method is a dynamic control rule between any two steady-state working points of the engine. Therefore, the adjustment parameters of each component of the start steady state point A and the final steady state point B of the dynamic process of the engine to be optimized and the performance parameters of the engine need to be obtained firstly. Taking the dynamic control rule optimization from the ground slow vehicle to the ground takeoff of the turbofan engine as an example, determining the start and final steady state points of the dynamic process of the engine is to obtain the adjustment parameters of each component of the ground slow vehicle steady state point and the ground takeoff steady state point of the turbofan engine according to the design result of the steady state performance of the turbofan engine. The method comprises the following steps: high and low pressure rotor speed, nozzle throat section area, compression component inlet guide vane angle, turbine component inlet guide vane section area, combustion chamber oil supply, variable area guide vane adjustment, and compression component surge margin.
And step two, determining a dynamic control rule optimization target and constraint of the engine from a steady-state point A to a steady-state point B. Taking the optimization of the dynamic control rule from the ground slow vehicle to the ground takeoff of the turbofan engine as an example, the design target is that the time required by the whole dynamic state is shortest, namely the engine is transited from the steady state point A to the steady state point B as fast as possible. And the constraint conditions are that in the whole dynamic state, the surge margin of each compression part of the engine is not higher than a threshold value, the high-low pressure rotating speed acceleration rate is not higher than the threshold value, the combustion chamber oil-gas ratio change rate is not higher than the threshold value, and the threshold values of the three are different and unequal.
And step three, establishing an initial control rule. The general initial control law can be simply set as a linear function of the control parameter and the argument. For the selection of the initial value of the dynamic control law with the fuel flow as the control parameter and the independent variable as the time, the fuel flow at the initial and final steady state points is assumed to be Wf_startAnd Wf_endThen the initial control law may be set as follows:
wherein WfFor burning control parameters in dynamic control lawsAnd (3) oil flow, wherein T is independent variable time in the dynamic control rule, and T is the maximum value of the independent variable T in the dynamic control rule, namely the total time of the dynamic control rule.
It should be noted that, the control parameters and the independent variables of the dynamic control law have various choices, for example, the control parameters may select the fuel flow, the high-low pressure rotor speed, the adjustment amount of each adjustable component, and the like, and the independent variables may select the time, the high-low pressure rotor speed, and the like. The control parameters and the independent variables are combined randomly on the premise of no conflict, and a dynamic control rule can be formed. The dynamic control law may also include the change laws of a plurality of control parameters, and the change laws of all the control parameters need to be initialized according to the method described in the foregoing step. The initial control law is only used as an initial value of the subsequent optimization, the specific value of the initial control law does not influence the final result, but the initial control law is ensured to meet the constraint of the control law determined in the step two.
And step four, the control law established in the decomposition step three is m sub-control law points, wherein m represents the decomposed sub-control law point number. The value range of the value range is suggested to be [10,100] by combining the performance of the current computer and the time-consuming requirement of the optimization of the dynamic control rule of a general engine. It should be noted that the m sub-control law points obtained by decomposition do not include dynamic start and end steady-state points, which are the premise and constraint of control law optimization.
And step five, establishing m corresponding sub-population evolutionarily optimized with the size of n according to the m sub-control rule points obtained by decomposition in the step four. Wherein n represents the size of the evolutionary optimized sub-population, i.e. each evolutionary optimized sub-population comprises n individuals, which value can be customized by a user, and is generally set to [50,200 ]. And each individual comprises all control parameters of the sub-control rule points corresponding to the evolution optimization sub-population.
And step six, establishing an elite individual set. Randomly selecting an individual from each evolution optimization sub-population as an elite individual of the sub-population, wherein the elite individual set is a set of elite individuals of the m evolution optimization sub-populations. It should be noted that any individual in each evolutionary optimization sub-population corresponds to a sub-control law point in a dynamic control law, and a set formed by these individuals corresponds to a dynamic control law, that is, an elite individual set also corresponds to a dynamic control law.
And step seven, initializing a control rule overall sub-population point-by-point optimization process. Taking i as 1, wherein i represents the evolutionarily optimized sub-population number to be operated in the current and subsequent steps.
And step eight, performing one-step optimization iteration on the ith evolutionary optimization sub-population. And (3) operating the current n individuals in the ith evolutionary optimization sub-population, namely parent individuals, by using mutation and crossover operators of a differential evolution algorithm to generate n sub-population.
And step nine, calculating the fitness of each individual in the ith evolutionary optimization sub-population.
And step one, initializing the ith evolutionary optimization sub population fitness calculation process. And j is taken as 1, wherein j represents the serial number of the individual currently calculating the fitness in the ith evolutionary optimization sub-population. It should be noted that, at this time, the ith evolutionary optimization sub-population includes 2 × n individuals, that is, the ith evolutionary optimization sub-population includes n parent individuals and n child individuals mentioned in step eight.
And a second substep of calculating the current individual fitness. And combining the jth individual in the ith evolution optimization sub-population with the elite individuals of the rest evolution optimization sub-populations to form a set of control rules. And calculating the dynamic target performance of the engine according to the control rule, evaluating the target performance and obtaining the fitness of the individual. Generally, if the control law optimization target is that the dynamic time is shortest, the fitness can be taken as the reciprocal of the dynamic time. I.e., generally equating the optimization problem objective to a minimum fitness.
And thirdly, judging whether all the individual fitness degrees are calculated. If j is j +1, go to step ten if j >2 × n, otherwise go to substep two.
Step ten, screening the ith evolutionary optimization sub-population. And D, according to the fitness calculated in the ninth step, carrying out fitness sequencing from small to large on n individuals in the ith evolutionary optimization sub-population, only reserving the first n individuals in the current evolutionary optimization sub-population, and discarding the rest individuals.
Step eleven, updating the elite individual set. And updating the elite individuals of the ith evolutionary optimization sub-population into the individuals with the minimum fitness in the sub-population.
And step twelve, judging whether the one-time point-by-point optimization cycle is finished. if i is equal to i +1, go to step thirteen if i > m, otherwise go to step eight.
And step thirteen, judging whether the control rule optimization process is finished. And if the optimization process reaches the end condition, turning to the step fourteen, otherwise, turning to the step eight, wherein i is equal to 1.
And step fourteen, the control law corresponding to the current elite individual set is the optimization result of the dynamic control law of the engine.
While the above is merely a preferred embodiment of the present invention for optimizing the acceleration control law of a mixed-exhaust turbofan engine, it should be noted that those skilled in the art can make several improvements and modifications without departing from the principle of the present invention, and these improvements and modifications should be considered as the protection scope of the present invention.
Claims (1)
1. The method for optimizing the dynamic control law of the engine is characterized by comprising the following steps of:
the method comprises the following steps: assuming that the starting steady state point of the engine in the dynamic process is A and the final steady state point is B, the optimization objects are as follows: a change rule of controllable parameters in a transition state from A to B and a change rule of adjustable component adjustment quantity in a transition state from A to B, wherein a plurality of adjustment parameters and engine performance parameters comprise high-low pressure rotor rotating speed, nozzle throat section area, compression component inlet guide vane angle, turbine component inlet guider section area, combustion chamber oil supply quantity, variable area guider adjustment quantity and compression component surge margin;
step two: determining a design target from A to B and a constraint condition required to be ensured by a dynamic process part of an engine, or determining a design target from A to B and a constraint condition required to be ensured by a complete machine parameter; the design objective is that the time required by the dynamic process from A to B is shortest, the surge margin of a compression part, the high-low pressure rotating speed acceleration rate and the combustion chamber oil-gas ratio change rate are different and unequal, and the constraint parameters are that the surge margin is not higher than a threshold value, the high-low pressure rotating speed acceleration rate and the combustion chamber oil-gas ratio change rate are not higher than a threshold value;
step three: establishing an initial control law, wherein the initial control law can be set according to the following formula:
X＝(XB-XA)×t/T
wherein T is the independent variable time in the dynamic control rule, T is the total time of the dynamic control rule, X is the vector formed by the parameters in the step one, and X is the vector formed by the parameters in the step oneBVector of control parameters, X, for the final steady-state pointAA control parameter vector which is a starting steady-state point;
step four: dispersing the control law established in the third step into m sub-control law points according to the independent variable,
wherein the m sub-control law points do not include a point A and a point B;
step five, aiming at the m sub-control rule points obtained by decomposition in the step four, establishing m corresponding evolution optimization sub-populations with the size of n;
step six, establishing an elite individual set; randomly selecting an individual from each evolution optimization sub-population as an elite individual of the sub-population, wherein an elite individual set is a set of elite individuals of m evolution optimization sub-populations; it should be mentioned that any individual in each evolution optimization sub-population corresponds to a sub-control law point in a dynamic control law, and a set formed by these individuals corresponds to a dynamic control law, that is, an elite individual set also corresponds to a dynamic control law;
step seven, initializing a control rule overall sub-population point-by-point optimization process; taking i as 1, wherein i represents the sequence number of the evolutionary optimization sub-population to be operated in the current and subsequent steps;
step eight, performing one-step optimization iteration on the ith evolutionary optimization sub-population; performing operation on the current n individuals in the ith evolutionary optimization sub-population, namely parent individuals, by using a mutation operator and a crossover operator of a differential evolution algorithm to generate n sub-generation individuals;
calculating the fitness of each individual in the ith evolutionary optimization sub-population;
step ten, screening the ith evolutionary optimization sub-population; according to the fitness calculated in the ninth step, carrying out fitness sequencing from small to large on n individuals in the ith evolutionary optimization sub-population, only reserving the first n individuals in the current evolutionary optimization sub-population, and discarding the rest individuals;
step eleven, updating the elite individual set; updating the elite individual of the ith evolutionary optimization sub-population into the individual with the minimum fitness in the sub-population;
step twelve, judging whether one point-by-point optimization cycle is finished; if i is equal to i +1, if i is greater than m, go to step thirteen, otherwise go to step eight;
step thirteen, judging whether the control rule optimization process is finished or not; if the optimization process reaches the end condition, turning to the step fourteen, otherwise, turning to the step eight if the optimization process is not finished, and turning to the step i;
and step fourteen, the control law corresponding to the current elite individual set is the optimization result of the dynamic control law of the engine.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910055751.6A CN109634107B (en) | 2019-01-22 | 2019-01-22 | An optimization method of engine dynamic control law |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910055751.6A CN109634107B (en) | 2019-01-22 | 2019-01-22 | An optimization method of engine dynamic control law |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109634107A CN109634107A (en) | 2019-04-16 |
| CN109634107B true CN109634107B (en) | 2021-07-16 |
Family
ID=66062260
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910055751.6A Active CN109634107B (en) | 2019-01-22 | 2019-01-22 | An optimization method of engine dynamic control law |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109634107B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021097696A1 (en) * | 2019-11-20 | 2021-05-27 | 大连理工大学 | Adaptive dynamic programming method for aero-engine in optimal acceleration tracking control |
| CN111783239B (en) * | 2020-06-05 | 2022-07-08 | 西北工业大学 | Fuzzy reliability analysis method for turbine tenon connection structure |
| CN112085145B (en) * | 2020-09-04 | 2023-02-21 | 北京航空航天大学 | Method for designing control law of mode conversion process of self-adaptive circulating aero-engine |
| CN115587446B (en) * | 2022-10-24 | 2024-05-28 | 南京航空航天大学 | A method for optimizing multi-task steady-state control law of variable cycle engine |
| CN119335862B (en) * | 2024-10-17 | 2025-09-05 | 西北工业大学 | A hierarchical optimization method for dynamic control laws of aircraft engines |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104375421A (en) * | 2014-11-24 | 2015-02-25 | 西北工业大学 | Aircraft engine part modeling method based on control system development platform |
| CN104392039A (en) * | 2014-11-19 | 2015-03-04 | 中国航空工业集团公司沈阳飞机设计研究所 | Transition state engine simulation modeling method |
| CN105302124A (en) * | 2015-12-03 | 2016-02-03 | 西北工业大学 | Turboshaft engine control performance index extraction method based on test data |
| CN106970523A (en) * | 2017-03-01 | 2017-07-21 | 南京航空航天大学 | A kind of energy management strategies of aircraft self-adapting power and heat management system |
| CN107357176A (en) * | 2017-08-05 | 2017-11-17 | 中国科学院工程热物理研究所 | A kind of aeroengine test run Data Modeling Method |
| CN108573116A (en) * | 2018-05-11 | 2018-09-25 | 南京航空航天大学 | An Algorithm for Thrust Estimation in Transition State of Aeroengine Based on Long Short-term Memory Network |
| CN108762089A (en) * | 2018-06-15 | 2018-11-06 | 大连理工大学 | A kind of aero-engine on-line optimization and multivariable Control design method based on model prediction |
| CN108828947A (en) * | 2018-07-13 | 2018-11-16 | 南京航空航天大学 | A kind of uncertain dynamic fuzzy model modeling method of the aero-engine containing time lag |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9715221B2 (en) * | 2014-05-01 | 2017-07-25 | Aspen Technology, Inc. | Online control calculation for models containing near colinearity and uncertainty |
-
2019
- 2019-01-22 CN CN201910055751.6A patent/CN109634107B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104392039A (en) * | 2014-11-19 | 2015-03-04 | 中国航空工业集团公司沈阳飞机设计研究所 | Transition state engine simulation modeling method |
| CN104375421A (en) * | 2014-11-24 | 2015-02-25 | 西北工业大学 | Aircraft engine part modeling method based on control system development platform |
| CN105302124A (en) * | 2015-12-03 | 2016-02-03 | 西北工业大学 | Turboshaft engine control performance index extraction method based on test data |
| CN106970523A (en) * | 2017-03-01 | 2017-07-21 | 南京航空航天大学 | A kind of energy management strategies of aircraft self-adapting power and heat management system |
| CN107357176A (en) * | 2017-08-05 | 2017-11-17 | 中国科学院工程热物理研究所 | A kind of aeroengine test run Data Modeling Method |
| CN108573116A (en) * | 2018-05-11 | 2018-09-25 | 南京航空航天大学 | An Algorithm for Thrust Estimation in Transition State of Aeroengine Based on Long Short-term Memory Network |
| CN108762089A (en) * | 2018-06-15 | 2018-11-06 | 大连理工大学 | A kind of aero-engine on-line optimization and multivariable Control design method based on model prediction |
| CN108828947A (en) * | 2018-07-13 | 2018-11-16 | 南京航空航天大学 | A kind of uncertain dynamic fuzzy model modeling method of the aero-engine containing time lag |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109634107A (en) | 2019-04-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109634107B (en) | An optimization method of engine dynamic control law | |
| CN108416086B (en) | An adaptive correction method for aero-engine model based on deep learning | |
| CN112528575B (en) | Multi-dimensional collaborative design method for multistage axial flow compressor of ship gas turbine | |
| WO2019144337A1 (en) | Deep-learning algorithm-based self-adaptive correction method for full-envelope model of aero-engine | |
| CN106951634B (en) | A Design Method of Robust Tracking Controller for Aeroengine | |
| CN112131670B (en) | Aero-engine model iterative algorithm based on hybrid adaptive differential evolution | |
| CN108223140A (en) | A kind of three shaft gas turbine power turbine control of inlet guide vanes rule optimization method | |
| CN108108528B (en) | A one-dimensional matching design method of split-shaft aeroderivative gas turbine power turbine | |
| CN113656907B (en) | Three-dimensional steady-state simulation matching iteration method for aircraft engine | |
| CN109031951B (en) | On-line establishment of aero-engine state variable model based on exact partial derivatives | |
| JP2006500697A (en) | Optimization expert system | |
| Egorov et al. | IOSO optimization toolkit-novel software to create better design | |
| CN110647052B (en) | An adaptive ID card model construction method for variable cycle engine mode switching | |
| Samuelsson et al. | Consistent conceptual design and performance modeling of aero engines | |
| CN110489877B (en) | Interpolation method suitable for real-time model of aero-engine | |
| CN108733906A (en) | Aero-engine component-level model construction method based on accurate partial derivative | |
| Marchukov et al. | Improving of the working process of axial compressors of gas turbine engines by using an optimization method | |
| CN117932960A (en) | Agent model-based aero-engine overall performance multi-objective optimization method | |
| CN115903484A (en) | Multivariate Robust Controller Optimization Method for Aeroengine Based on Cooperative Game | |
| CN115963727A (en) | A Model-Based Method for Designing Acceleration and Deceleration Control Laws of Turbofan Engines | |
| Lengyel-Kampmann et al. | Generalized optimization of counter-rotating and single-rotating fans | |
| CN110321586A (en) | A kind of aero-engine deviates the obtaining value method of design point working condition iterative solution | |
| Joachim | An enhanced compressor map extension method suited for spool speeds down to 1% | |
| CN115470575B (en) | Control parameter optimization method for variable cycle aeroengine acceleration process | |
| CN111624886B (en) | A SARSA-based variable cycle aero-engine thrust control method |
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 |