Multi-objective optimization method, device, electronic device and storage medium

Abstract

The present invention provides a multi-objective optimization method, device, electronic device and storage medium, wherein a server obtains initial parameters and selected evaluation indicators; based on the initial parameters, a GCAir platform is driven to simulate a system to be simulated to obtain a simulation result; based on the evaluation indicators and the simulation result, a comprehensive evaluation function is obtained; if it is determined based on the comprehensive evaluation function that the simulation requirement is not met, or it is determined that the comprehensive evaluation function is not a target solution, the multi-objective optimization algorithm of the system is called to optimize the initial parameters to obtain the parameter optimization value; based on the parameter optimization value, the GCAir platform is continuously driven to simulate the system to be simulated until the comprehensive evaluation function corresponding to the obtained simulation result meets the simulation requirement, and the parameter optimization value is used as the target parameter of the system to be simulated. The present invention realizes multi-objective parameter optimization by calling the multi-objective optimization algorithm and continuously adjusting the system parameters through the GCAir platform, and can improve the efficiency of multiple target optimization.

Description

Technical Field

[0001] The present invention relates to the technical field of simulation data processing, and in particular to a multi-objective optimization method, device, electronic equipment and storage medium. Background Art

[0002] In complex system simulation engineering design, system construction needs to consider multiple goals such as performance, cost and reliability.

[0003] At present, each objective is often optimized in turn through a single-objective optimization method. Since these objectives are usually contradictory, optimizing one objective may cause the values ​​of other objectives to decrease or increase, thereby affecting the optimization efficiency of multiple objectives. Summary of the invention

[0004] In view of this, an embodiment of the present invention provides a multi-objective optimization method, device, electronic device and storage medium to solve the problem of affecting the optimization efficiency of multiple objectives in the prior art.

[0005] To achieve the above objectives, the embodiments of the present invention provide the following technical solutions:

[0006] The first aspect of the embodiment of the present invention shows a multi-objective optimization method, which is applied to a server, and the server is connected to the GCAir platform and the interaction module respectively. The method includes:

[0007] Obtaining initial parameters and selected evaluation indicators based on the interactive module input;

[0008] Based on the initial parameters, the GCAir platform is driven to simulate the system to be simulated to obtain a simulation result, wherein the system to be simulated is a pre-selected system engineering;

[0009] Performing processing based on the evaluation index and the simulation result to obtain a comprehensive evaluation function;

[0010] If it is determined based on the comprehensive evaluation function that the simulation requirements are not met, calling its own multi-objective optimization algorithm to optimize the initial parameters to obtain parameter optimization values;

[0011] The GCAir platform is continuously driven to simulate the system to be simulated based on the parameter optimization value until the comprehensive evaluation function corresponding to the obtained simulation result meets the simulation requirement, and the parameter optimization value is used as the target parameter of the system to be simulated.

[0012] Optionally, the processing based on the evaluation index and the simulation result to obtain a comprehensive evaluation function includes:

[0013] Obtaining a simulation value corresponding to each of the evaluation indicators in the simulation result;

[0014] For each of the evaluation indicators, a simulation value corresponding to the evaluation indicator and a corresponding indicator weight are calculated to obtain a first value, wherein the indicator weight corresponding to each of the evaluation indicators is obtained by processing the indicator comparison value input in advance by the interaction module;

[0015] Processing is performed based on the first numerical value corresponding to each evaluation index to obtain a comprehensive evaluation function.

[0016] Optionally, also include:

[0017] If it is determined based on the comprehensive evaluation function that the simulation requirements are met, the initial parameters used for the simulation are used as target parameters;

[0018] The target parameters are displayed through the interactive module for users to view.

[0019] Optionally, determining that the simulation requirement is not met based on the comprehensive evaluation function includes:

[0020] Determine whether the comprehensive evaluation function is greater than or equal to a preset parameter;

[0021] If it is greater than or equal to, it is determined that the comprehensive evaluation function meets the simulation requirements.

[0022] Optionally, also include:

[0023] Constructing a starting population based on the initial values ​​of the initial parameters and the optimization range corresponding to each initial value of the parameters;

[0024] Calculate the corresponding fitness value based on each individual in the starting population;

[0025] For the initial value of each initial parameter, sort them in descending order according to the fitness value, and determine the individual corresponding to the fitness value ranked first;

[0026] Take the individual corresponding to the first fitness value in each ranking as the target solution;

[0027] Determining whether the comprehensive evaluation function is the same as the target solution;

[0028] If the comprehensive evaluation function is not the target solution, the step of calling its own multi-objective optimization algorithm to optimize the initial parameters to obtain the optimized values ​​of the parameters is executed.

[0029] Optionally, calling its own multi-objective optimization algorithm to optimize the initial parameters to obtain parameter optimization values ​​includes:

[0030] Constructing a starting population based on the initial values ​​of the initial parameters and the optimization range corresponding to each initial value of the parameters;

[0031] Calling its own multi-objective optimization algorithm to perform multi-objective optimization from the starting population to obtain offspring;

[0032] The initial parameters are updated based on the offspring to obtain optimized parameter values.

[0033] Optionally, if it is determined based on the comprehensive evaluation function that the simulation requirement is not met, the method further includes:

[0034] Get the current number of optimization attempts;

[0035] Determine whether the number of optimization searches reaches a preset number of iterations;

[0036] If so, the initial parameters used for simulation are used as target parameters;

[0037] If not, the process executes the step of calling its own multi-objective optimization algorithm to optimize the initial parameters and obtain the optimized values ​​of the parameters.

[0038] The second aspect of the embodiment of the present invention shows a multi-objective optimization device, including a server, a GCAir platform and an interaction module, wherein the server is connected to the GCAir platform and the interaction module respectively;

[0039] The server obtains initial parameters and selected evaluation indicators based on the input of the interactive module; drives the GCAir platform to simulate the system to be simulated based on the initial parameters to obtain simulation results, wherein the system to be simulated is a pre-selected system engineering; processes based on the evaluation indicators and the simulation results to obtain a comprehensive evaluation function; if it is determined based on the comprehensive evaluation function that the simulation requirements are not met, calls its own multi-objective optimization algorithm to optimize the initial parameters to obtain parameter optimization values; based on the parameter optimization values, continues to drive the GCAir platform to simulate the system to be simulated until the comprehensive evaluation function corresponding to the obtained simulation results meets the simulation requirements, and uses the parameter optimization values ​​as the target parameters of the system to be simulated.

[0040] A third aspect of an embodiment of the present invention shows an electronic device, which is used to run a program, wherein the program, when running, executes the multi-objective optimization method shown in the first aspect of the embodiment of the present invention.

[0041] A fourth aspect of an embodiment of the present invention shows a storage medium, which includes a stored program, wherein when the program is running, the device where the storage medium is located is controlled to execute the multi-objective optimization method shown in the first aspect of the embodiment of the present invention.

[0042] Based on the multi-objective optimization method, device, electronic device and storage medium provided by the above-mentioned embodiment of the present invention, the server is respectively connected with the GCAir platform and the interaction module, and the method includes: obtaining initial parameters and selected evaluation indicators based on the input of the interaction module; driving the GCAir platform to simulate the system to be simulated based on the initial parameters to obtain simulation results, wherein the system to be simulated is a pre-selected system engineering; processing based on the evaluation indicators and the simulation results to obtain a comprehensive evaluation function; if it is determined based on the comprehensive evaluation function that the simulation requirements are not met, calling its own multi-objective optimization algorithm to optimize the initial parameters to obtain parameter optimization values; based on the parameter optimization values, continue to drive the GCAir platform to simulate the system to be simulated until the comprehensive evaluation function corresponding to the obtained simulation result meets the simulation requirements, and use the parameter optimization values ​​as the target parameters of the system to be simulated. After the initial parameters and evaluation indicators are set, the server sends them to the system parameter GCAir platform so that the GCAir platform performs simulation based on the system parameters. The server calls the multi-objective optimization algorithm to continuously adjust the system parameters, i.e., the parameter optimization value, through the GCAir platform, and drives the GCAir engineering simulation until the comprehensive evaluation function corresponding to the obtained simulation result meets the simulation requirements, and the parameter optimization value is used as the target parameter of the system to be simulated. In this way, multi-objective parameter optimization is realized, and the efficiency of multiple target optimization can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0044] Figure 1 A schematic diagram of the architecture of a multi-objective optimization device according to an embodiment of the present invention;

[0045] Figure 2 This is a diagram showing the interaction between the GCAir platform and the server according to an embodiment of the present invention;

[0046] Figure 3 A schematic diagram of a specific architecture of a multi-objective optimization device according to an embodiment of the present invention;

[0047] Figure 4 A schematic diagram of interface settings shown in an embodiment of the present invention;

[0048] Figure 5 A schematic diagram of a multi-objective optimization method according to an embodiment of the present invention;

[0049] Figure 6 A schematic diagram of a multi-objective optimization example shown in an embodiment of the present invention;

[0050] Figure 7 The present invention is a flowchart of another multi-objective optimization method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0051] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0052] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0053] It should be noted that the descriptions of "first", "second", etc. in the present invention are only used for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In addition, the technical solutions between the various embodiments can be combined with each other, but they must be based on the ability of ordinary technicians in the field to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such a combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0054] In this application, the terms "comprises", "comprising" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprising a ..." does not exclude the presence of other identical elements in the process, method, article or device comprising the element.

[0055] See also Figure 1 , is a schematic diagram of the architecture of a multi-objective optimization device according to an embodiment of the present invention, wherein the device comprises:

[0056] The GCAir platform 10 is connected to the server 20 and the interaction module 30 respectively.

[0057] The interaction module 30 may be a human-machine interaction interface. Specifically, the human-machine interaction interface may be integrated on the GCAir platform 10 in the form of a plug-in. That is, the GCAir platform 10 and the interaction module 30 are communicatively connected.

[0058] The GCAir platform 10 is provided with an interactive plug-in, so that the GCAir platform 10 can feed back the simulation results to the server 20 provided with a multi-objective optimization algorithm, i.e., an intelligent optimization algorithm. The server 20 provided with a multi-objective optimization algorithm, i.e., an intelligent optimization algorithm, can call the GCAir platform 10 to perform system simulation; it can also adjust parameters based on the simulation results and repeatedly call the GCAir platform 10 to perform system simulation, such as Figure 2 shown.

[0059] The integrated simulation test and verification GCAir platform 10 is a system integration simulation test software used to build more realistic and complex system engineering in different fields.

[0060] For example: the real complex system engineering includes semi-rocker landing gear, double-chamber single throttle buffer, spherical adaptive throttle valve, UH60 model whole machine model, and 3D tire model.

[0061] Specifically, a parameter optimization implementation script is set in the server 20 and deployed on the server. The implementation script is connected to the GCAir platform 10 through the application programming interface API, and the server 20 and the interaction module 30 are also connected to each other.

[0062] Among them, the implementation script of parameter optimization can be implemented by computer programming language Python, and can also be implemented by other computer programming languages, which is not limited by this application.

[0063] The process of implementing multi-objective optimization based on the above-mentioned architecture includes:

[0064] The server 20 obtains initial parameters and selected evaluation indicators from the interaction module 30, wherein the number of initial values ​​of the parameters in the initial parameters is multiple;

[0065] The server 20 drives the GCAir platform 10 to simulate the system to be simulated based on the initial parameters to obtain a simulation result, wherein the system to be simulated is a pre-selected system engineering;

[0066] The server 20 processes the evaluation index and the simulation result to obtain a comprehensive evaluation function; if it is determined based on the comprehensive evaluation function that the simulation requirements are not met, its own multi-objective optimization algorithm is called to optimize the initial parameters to obtain parameter optimization values, and the number of the parameter optimization values ​​is the same as the number of the initial values ​​of the parameters in the initial parameters; based on the parameter optimization values, the GCAir platform 10 is continuously driven to simulate the system to be simulated until the comprehensive evaluation function corresponding to the obtained simulation result meets the simulation requirements, and the parameter optimization values ​​are used as the target parameters of the system to be simulated.

[0067] Optionally, the GCAir platform 10 includes a control module, a controlled object module and a system engineering, and the server 20 includes a multi-objective optimization algorithm and an evaluation function module; the interactive interface 30 is connected to the GCAir platform 10 through a dynamic link library DLL; the interactive interface 30 includes a module for setting the initial value and optimization range of the parameter to be optimized, i.e., the initial parameter, a user-defined evaluation index module, and a display of the simulation results of key variables and the optimal parameters, i.e., the target parameters, such as Figure 3 shown.

[0068] The user can set the initial value of the parameter to be optimized through the interactive module 30, and the user can customize the module of the evaluation index; when the GCAir platform 10 feeds back the simulation results, the interactive module 30 obtains the simulation results related to the evaluation index, that is, the simulation results of the key variables, from the simulation results and sends them to the server 20.

[0069] The multi-objective optimization interface of the GCAir platform 10 includes an interactive interface of the interactive module 30, which can be used for configuration selection, i.e., selecting the system to be simulated, optimizing parameter settings including the initial value and optimization range settings of the parameters to be optimized, and the settings of simulation variable monitoring, i.e., the system optimization results returned by the GCAir platform 10, the number of evaluation indicators, the setting of evaluation indicators, and the display of the optimal parameters, i.e., the target parameters, such as Figure 4 shown.

[0070] The initial parameters include the initial value of the parameter to be optimized and the optimization range.

[0071] The multi-objective optimization device provided in the present invention is a tool for solving optimization problems of multiple objective functions. It can help users solve multi-objective optimization problems in the system to be simulated, that is, complex system design, through a specific intelligent optimization algorithm. The multi-objective optimization algorithm can consider multiple objectives at the same time, and through intelligent search and optimization processes, it helps engineering designers make reasonable decisions while considering multiple objectives, and realize the optimal comprehensive performance design of complex systems.

[0072] The multi-objective optimization device can consider multiple objectives at the same time and find a set of optimal solutions, thereby providing a more comprehensive and integrated optimization solution.

[0073] In the embodiment of the present invention, when the system parameters, i.e., the initial values ​​or optimized values ​​of the parameters in the initial parameters and the evaluation index are set, the server sends them to the system parameter GCAir platform so that the GCAir platform performs simulation based on the system parameters, and the server calls the multi-objective optimization algorithm to continuously adjust the system parameters, i.e., the optimized values ​​of the parameters, through the computer program programming interface API of the GCAir platform and drives the GCAir engineering simulation until the comprehensive evaluation function corresponding to the obtained simulation result meets the simulation requirements, and the optimized values ​​of the parameters are used as the target parameters of the system to be simulated. Thus, multi-objective parameter optimization is achieved.

[0074] This article provides a method for multi-objective optimization of complex systems based on GCAir and the server, using computer programming to build intelligent optimization algorithms, and using GCAir to build complex system engineering. It is not only universal, but also can simplify the preliminary preparation work of technicians and reduce the cost of software learning. GCAir software has the functions of system simulation design and application programming API interface, which gives it a natural advantage in building complex environment system models and integrating intelligent optimization algorithms.

[0075] The multi-objective optimization device establishes data communication between the system engineering constructed by GCAir and the server, provides a display interface for the system parameter optimization process, and provides a configuration interface for the optimization objectives. This device is convenient and universal, which can simplify the preliminary preparation work of system designers and reduce system design costs.

[0076] Optionally, based on the multi-objective optimization device shown in the above embodiment of the present invention, the server 20 that processes based on the evaluation index and the simulation result to obtain a comprehensive evaluation function is specifically used for:

[0077] Acquire the simulation value corresponding to each of the evaluation indicators in the simulation result; for each of the evaluation indicators, calculate the simulation value corresponding to the evaluation indicator and the corresponding indicator weight to obtain a first numerical value, wherein the indicator weight corresponding to each of the evaluation indicators is obtained by processing the indicator comparison value input in advance through the interactive module; obtain a comprehensive evaluation function based on the first numerical value corresponding to each evaluation indicator.

[0078] In a specific implementation, the simulation value corresponding to each of the evaluation indicators is extracted from the simulation results; for each of the evaluation indicators, the simulation value corresponding to the evaluation indicator and the indicator weight corresponding to the evaluation indicator are weightedly calculated to obtain a first numerical value, and then the first numerical value corresponding to each evaluation indicator is obtained.

[0079] It should be noted that the indicator weight corresponding to each evaluation indicator is obtained by processing the indicator comparison value input by the interaction module in advance.

[0080] Specifically, for each evaluation indicator, the process of obtaining the indicator weight by processing the indicator comparison value input by the interaction module includes:

[0081] Obtain the index comparison value of each evaluation index input in advance through the interactive module; construct a judgment matrix using the index comparison value of each evaluation index; analyze the correlation between each index of different layers and the upper layer index, and output the index weight of each evaluation index.

[0082] Specifically, the analytic hierarchy process (AHP) is used to analyze the judgment matrix in accordance with the thinking mode of decomposition, comparative judgment and synthesis, so as to analyze the influence of each indicator comparison value in each level on the result, and quantify it to obtain the indicator weight of each evaluation indicator.

[0083] Optionally, based on the multi-objective optimization device shown in the above-mentioned embodiment of the present invention, the server 20 is also used for: if it is determined based on the comprehensive evaluation function that the simulation requirements are met, using the initial parameters used for simulation as target parameters; and displaying the target parameters through the interactive module for user viewing.

[0084] Optionally, based on the multi-objective optimization device shown in the above embodiment of the present invention, the server 20 that does not meet the simulation requirements is determined based on the comprehensive evaluation function, specifically for:

[0085] Determine whether the comprehensive evaluation function is greater than or equal to a preset parameter; if so, determine that the comprehensive evaluation function meets the simulation requirements.

[0086] Optionally, based on the multi-objective optimization device shown in the above embodiment of the present invention, the server 20 is further used for:

[0087] The method comprises the following steps: constructing a starting population based on the initial values ​​of the initial parameters and the optimization range corresponding to each initial value of the parameter; calculating the corresponding fitness value based on each individual in the starting population; sorting the initial values ​​of each initial parameter in descending order according to the fitness value, and determining the individual corresponding to the fitness value ranked first; taking the individual corresponding to each fitness value ranked first as the target solution; judging whether the comprehensive evaluation function is the same as the target solution; and executing the step of calling its own multi-objective optimization algorithm to optimize the initial parameters and obtain the parameter optimization value if the comprehensive evaluation function is not the target solution.

[0088] Specifically, for each initial value of the initial parameters, a value obtained by adding any number in the corresponding optimization range to the initial value of the parameter, or by subtracting any number in the corresponding optimization range from the initial value of the parameter is taken as an individual; the individuals corresponding to the initial value of the parameter are formed into a starting population, thereby obtaining multiple starting populations, and initializing them;

[0089] Each of these entities typically represents a potential solution to the problem, the specific form of which depends on the GCAir system engineering, that is, the characteristics of the performance of the system to be simulated.

[0090] Optionally, before executing step S21, the population parameters of the multi-objective optimization algorithm are set, such as the population size POP, the number of iterations M, the function f for calculating the fitness, the number of clusters m, etc.

[0091] Among them, the fitness value is usually determined by the evaluation index of the problem, which is used to measure the quality of individual solutions; determine the index that affects each individual, that is, the change of the individual will affect the change of the index; then, based on each solution, use the GCAir platform to simulate the simulation system to obtain the simulation results corresponding to each solution; extract the index corresponding to each individual, and perform the above-mentioned weighted calculation to obtain the fitness of the individual. For the initial value of each initial parameter, compare the size of the fitness value, sort them in order from large to small, and determine the individual corresponding to the fitness value ranked first; and take the individual corresponding to the fitness value ranked first as the target solution, that is, the best solution. Determine whether the initial comprehensive evaluation function is the same as the target solution. If so, the input initial parameter can be directly used as the target parameter. If not, call its own multi-objective optimization algorithm to optimize the initial parameter to obtain the parameter optimization value.

[0092] Optionally, based on the multi-objective optimization device shown in the above embodiment of the present invention, the server 20 calls its own multi-objective optimization algorithm to optimize the initial parameters to obtain the parameter optimization value, which is specifically used for:

[0093] The initial population is constructed based on the initial values ​​of the initial parameters and the optimization range of each initial value of the parameter; the multi-objective optimization algorithm itself is called to perform multi-objective optimization from the initial population to obtain offspring; the initial values ​​of the parameters in the initial parameters are updated based on the offspring to obtain optimized values ​​of the parameters.

[0094] Specifically, the initial value of the initial parameter is used as the position of the individual, and an exploration direction and a distance less than the corresponding optimization range are selected from the position of each individual in the initial population; the position of the individual is updated according to the exploration direction and distance, that is, a new individual is obtained and used as an offspring; the offspring replaces the corresponding parent generation to form a new population, that is, the parameter optimization value is obtained;

[0095] It should be noted that the exploration direction and distance are pre-set according to the actual situation, and the distance cannot be greater than the optimization range of the initial parameter value.

[0096] Among them, according to the specific replacement strategy, such as retaining the best individual and updating the entire population, that is, the GCAir parameters are constantly changing, and the process of updating the individual position is the process of continuously adjusting the GCAir system parameters.

[0097] In an embodiment of the present invention, by continuously generating new solutions, i.e., current parameter optimization values, and replacing old solutions, i.e., initial parameters or parameter optimization values ​​obtained last time, a multi-objective optimization algorithm is used to gradually optimize individuals in the starting population so that the parameter optimization values ​​gradually approach or reach the optimal solution to the problem.

[0098] Optionally, based on the multi-objective optimization device shown in the above embodiment of the present invention, the server 20 is further used for:

[0099] If it is determined based on the comprehensive evaluation function that the simulation requirements are not met, the current number of optimization searches is obtained; whether the number of optimization searches reaches the preset number of iterations is determined; if so, the initial parameters used for simulation are used as target parameters; if not, the step of calling its own multi-objective optimization algorithm to optimize the initial parameters and obtain the optimized parameter values ​​is executed.

[0100] Optionally, each time the multi-objective optimization algorithm is called to optimize the initial parameters, the number of optimizations is incremented by 1 and recorded.

[0101] Specifically, the current solution, i.e., the number of iterations of the initial parameters or parameter optimization values, i.e., the number of optimizations, is determined. It is judged whether the number of optimizations reaches the preset number of iterations. If so, the initial parameters can be directly used as target parameters. If not, the multi-objective optimization algorithm is called to optimize the initial parameters to obtain the parameter optimization values.

[0102] See also Figure 5, is a flow chart of a multi-objective optimization method according to an embodiment of the present invention, which is applied to a server, and the method includes:

[0103] Step S501: obtaining initial parameters and selected evaluation indicators based on input of an interactive module, wherein the number of the initial parameters is multiple;

[0104] Optionally, the user may set the initial parameters of each system to be simulated, that is, the initial values ​​and optimization ranges of the parameters to be optimized, through the human-computer interaction interface of the interaction module.

[0105] The optimization range includes an upper limit and a lower limit of each parameter initial value that can be adjusted.

[0106] At the same time, the evaluation index of each system to be simulated is customized, that is, the index related to the performance and status of the device itself, that is, the key index, is determined in each system to be simulated.

[0107] It should be noted that, in addition to the user-defined method, indicators related to the performance and status of the device itself, namely key indicators, can also be automatically selected through the implementation manual and implementation functions of the system to be simulated.

[0108] Next, the interaction module sends the initial parameters and the evaluation index to the server so that the server can receive them.

[0109] Step S502: driving the GCAir platform to simulate the system to be simulated based on the initial parameters to obtain simulation results.

[0110] Among them, the system to be simulated is a pre-selected system engineering;

[0111] Optionally, the configuration is selected in advance through the GCAir platform, and the system engineering that needs to be simulated is selected so that the GCAir platform can build a complex system engineering.

[0112] System engineering may include semi-levered landing gear, dual-chamber single throttle buffer, spherical adaptive throttle, UH60 model complete machine model, and 3D tire model waiting for simulation system.

[0113] In the specific implementation process of step S502, the GCAir platform is controlled to execute the corresponding simulation process on the simulation system according to the initial values ​​of the parameters in the initial parameters to obtain the simulation results.

[0114] It should be noted that the simulation results may be in the form of a curve or in the form of data, including data or curves corresponding to multiple indicators.

[0115] Step S503: performing processing based on the evaluation index and the simulation result to obtain a comprehensive evaluation function;

[0116] It should be noted that the specific process of implementing step S503 includes the following steps:

[0117] Step S11: obtaining a simulation value corresponding to each evaluation index in the simulation result;

[0118] In the specific implementation of step S11, the simulation value corresponding to each evaluation index in the simulation result is extracted.

[0119] Step S12: For each of the evaluation indicators, the simulation value corresponding to the evaluation indicator and the corresponding indicator weight are calculated to obtain a first value.

[0120] It should be noted that the indicator weight corresponding to each evaluation indicator is obtained by processing the indicator comparison value input by the interaction module in advance.

[0121] Specifically, for each evaluation indicator, the process of obtaining the indicator weight by processing the indicator comparison value input by the interaction module includes:

[0122] Obtaining an indicator comparison value of each evaluation indicator input in advance through the interactive module;

[0123] Constructing a judgment matrix using the indicator comparison value of each of the evaluation indicators;

[0124] The correlation between each indicator at different levels and the indicators at the upper level is analyzed, and the indicator weight of each evaluation indicator is output.

[0125] In the specific implementation, the analytic hierarchy process (AHP) is used to analyze the judgment matrix according to the thinking mode of decomposition, comparative judgment and synthesis, so as to analyze the influence of each indicator comparison value in each level on the result, and quantify it to obtain the indicator weight of each evaluation indicator.

[0126] In the specific implementation of step S12, for each evaluation indicator, a simulation value corresponding to the evaluation indicator and an indicator weight corresponding to the evaluation indicator are weightedly calculated to obtain a first value, thereby obtaining a first value corresponding to each evaluation indicator.

[0127] Step S13: Processing is performed based on the first numerical value corresponding to each evaluation index to obtain a comprehensive evaluation function.

[0128] In the specific implementation of step S13, the first value corresponding to each evaluation index is summed up to obtain a comprehensive evaluation function.

[0129] Step S504: Determine whether the comprehensive evaluation function meets the simulation requirements. If so, execute step S505; if not, execute step S506:

[0130] In the process of implementing step S504, it is determined whether the comprehensive evaluation function is greater than or equal to the preset parameters, that is, the simulation requirements. If it is greater than or equal to, it means that the comprehensive evaluation function meets the simulation requirements, and step S505 is executed; if it is less than, it means that the comprehensive evaluation function does not meet the simulation requirements, and step S506 is executed.

[0131] It should be noted that, in addition to the above method of determining whether it is necessary to optimize the initial values ​​of the parameters in the initial parameters, it can also be determined in the following way.

[0132] In one implementation, it is determined whether the comprehensive evaluation function is a target solution. If so, the initial parameters can be directly used as target parameters. If not, step S506 is executed.

[0133] The specific implementation process includes the following steps.

[0134] Step S21: constructing a starting population based on the initial parameters, the initial values ​​of the parameters and the optimization range corresponding to each initial value of the parameters.

[0135] In the specific implementation of step S21, for the initial value of each initial parameter, a value obtained by adding any number in the corresponding optimization range to the initial value of the parameter, or by subtracting any number in the corresponding optimization range from the initial value of the parameter is taken as an individual; the individuals corresponding to each initial value of the parameter are formed into a starting population, thereby obtaining multiple starting populations, and initializing them;

[0136] Each of these entities typically represents a potential solution to the problem, the specific form of which depends on the GCAir system engineering, that is, the characteristics of the performance of the system to be simulated.

[0137] Optionally, before executing step S21, the population parameters of the multi-objective optimization algorithm are set, such as the population size POP, the number of iterations M, the function f for calculating the fitness, the number of clusters m, etc.

[0138] Step S22: Calculate the corresponding fitness value based on each individual in the initial population.

[0139] Among them, the fitness value is usually determined by the evaluation index of the problem and is used to measure the quality of individual solutions;

[0140] The specific implementation process of step S22 includes determining the indicators that affect each individual, that is, changes in the individual will affect changes in the indicators; then, based on each solution, the GCAir platform is used to simulate the simulation system to obtain the simulation results corresponding to each solution; the indicators corresponding to each individual are extracted, and weighted calculations are performed through the above steps S11 to S12 to obtain the fitness of the individual.

[0141] A set of input parameters of the GCAir system engineering shown in this application is equivalent to a solution, and the fitness value of each individual is obtained by the evaluation function of the algorithm, and the data source of the evaluation function of the algorithm is the GCAir simulation result data.

[0142] Step S23: sorting the initial values ​​of each initial parameter in descending order of the fitness values, and determining the individual corresponding to the fitness value ranked first;

[0143] Step S24: Take the individual corresponding to each first-ranked fitness value as the target solution.

[0144] In the process of implementing step S23 and step S24, the fitness of each starting population is compared with respect to the initial value of each initial parameter to compare the corresponding fitness values, and the fitness values ​​are sorted in order from large to small to determine the individual corresponding to the fitness value ranked first; and the individual corresponding to the fitness value ranked first is taken as the target solution, that is, the best solution.

[0145] Step S25: Determine whether the comprehensive evaluation function is the same as the target solution. If so, the initial parameters can be directly used as the target parameters. If not, execute step S506.

[0146] In the specific implementation of step S25, it is determined whether the comprehensive evaluation function is identical to the target solution. If so, the initial parameters can be directly used as the target parameters. If not, step S506 is executed.

[0147] Step S505: using the initial parameters used for simulation as target parameters, and displaying the target parameters through the interactive module for users to view.

[0148] Step S506: calling its own multi-objective optimization algorithm to optimize the initial parameters to obtain parameter optimization values.

[0149] Wherein, the number of the parameter optimization values ​​is the same as the number of the initial values ​​of the parameters in the initial parameters;

[0150] Optionally, before executing step S506, the number of optimization searches t is recorded.

[0151] It should be noted that there are multiple implementation methods for the specific implementation of step S506, including the following steps:

[0152] Step S31: constructing a starting population based on the initial values ​​of the initial parameters and the optimization range corresponding to each initial value of the parameters.

[0153] It should be noted that the specific implementation process of step S31 is the same as the specific implementation process of the above-mentioned specific implementation step S21, and they can refer to each other.

[0154] Optionally, after executing step S31, the following steps are further included:

[0155] Step S41: Calculate the corresponding fitness value based on each individual in the initial population.

[0156] It should be noted that the specific implementation process of step S41 is the same as the specific implementation process of the above-mentioned specific implementation step S22, and they can refer to each other.

[0157] Step S42: Selecting an individual as a parent according to the fitness value.

[0158] In the specific implementation of step S42, the fitness values ​​are compared, and the individual with higher fitness among the individuals corresponding to each initial parameter is taken as the parent.

[0159] Optionally, in a specific implementation, the parent generation may replace the initial parameters and the process returns to step S502.

[0160] Step S32: calling its own multi-objective optimization algorithm to perform multi-objective optimization from the starting population according to the initial parameters to obtain offspring.

[0161] In the specific implementation of step S32, the initial value of the initial parameter is used as the position of the individual, and an exploration direction and a distance less than the corresponding optimization range are selected from the position of each individual in the starting population; the position of the individual is updated according to the exploration direction and distance, that is, a new individual is obtained and used as an offspring;

[0162] It should be noted that the exploration direction and distance are set in advance according to the actual situation.

[0163] Optionally, since the specific implementation of step S32 is similar to the movement process of birds when searching for food, the multi-objective optimization algorithm can be a sparrow search algorithm, which is not limited in this application.

[0164] Step S33: Update initial parameters based on the offspring to obtain optimized parameter values.

[0165] In the specific implementation of step S33, the corresponding parent generation is replaced by the offspring to form a new population, that is, the parameter optimization value is obtained;

[0166] Among them, according to the specific replacement strategy, such as retaining the best individual and updating the entire population, that is, the GCAir parameters are constantly changing, and the process of updating the individual position is the process of continuously adjusting the GCAir system parameters, that is, the initial values ​​of the parameters.

[0167] In an embodiment of the present invention, by continuously generating new solutions, i.e., current parameter optimization values, and replacing old solutions, i.e., initial parameters or previous parameter optimization values, a multi-objective optimization algorithm is used to gradually optimize individuals in the starting population so that the parameter optimization values ​​gradually approach or reach the optimal solution to the problem.

[0168] Step S507: Based on the parameter optimization value, the GCAir platform is continued to be driven to simulate the system to be simulated, that is, the execution of step S503 is returned until the comprehensive evaluation function corresponding to the obtained simulation result meets the simulation requirement, and the parameter optimization value is used as the target parameter of the system to be simulated.

[0169] In the process of implementing step S508, after obtaining the simulation result, continue to return to step S503 to step S505 based on the parameter optimization value to re-judge whether the comprehensive evaluation function corresponding to the simulation result meets the simulation requirements. If so, use the parameter optimization value as the target parameter of the system to be simulated. If not, call its own multi-objective optimization algorithm to optimize the parameter optimization value to obtain a new parameter optimization value, and so on, until it is determined that the comprehensive evaluation function corresponding to the simulation result meets the simulation requirements.

[0170] In order to better understand the multi-objective optimization method shown above, an example is given below. Figure 6 shown.

[0171] For example, if the selected system to be simulated is a landing gear system, firstly, the optimization parameters are set in the interactive module, that is, the initial values ​​of the initial parameters are set to Ds2, Ns2, Dpl1, and Dpl2; and the optimization range of each initial value of the parameter is set respectively, that is, the upper and lower limits of the parameters of Ds2, Ns2, Dpl1, and Dpl2.

[0172] Next, set the evaluation indicators, such as the maximum vertical load of the tire, buffer efficiency, overload factor, etc.

[0173] Input the comparison values ​​between indicators, construct the AHP judgment matrix, and then calculate the indicator weights corresponding to each indicator so as to construct the evaluation function later.

[0174] The Sparrow Search Algorithm (SSA) can be selected as the multi-objective optimization algorithm, and the algorithm initialization settings (population size POP, number of iterations M, function f used to calculate fitness, number of clusters m, etc.) can be performed.

[0175] The server drives the GCAir platform to run a landing gear system drop analysis simulation project based on the initial values ​​of the parameters in the initial parameters to obtain simulation results.

[0176] Obtaining indicators such as the maximum vertical load of the tire, the buffer efficiency, and the overload factor from the simulation results or simulation curves; and calculating the current evaluation function based on the indicators such as the maximum vertical load of the tire, the buffer efficiency, and the overload factor and the indicator weights corresponding to each indicator;

[0177] The judgment function is compared with the measured / experimental data or with the target solution to determine whether the comprehensive evaluation function meets the simulation requirements. If so, the initial values ​​of the parameters in the initial parameters are used as the target parameters, that is, the optimal solution, and the optimal solution is output through the interactive module;

[0178] If not, determine whether the comprehensive evaluation function is the best, that is, the target solution. If not, call the multi-objective optimization algorithm to continuously adjust the system parameters through the computer program programming interface API of the GCAir platform, that is, the initial values ​​of the parameters, and construct new parameters Ds2, Ns2, Dpi1, and Dpi2, that is, the optimized values ​​of the parameters; and return to the steps of executing the landing gear system drop analysis simulation project.

[0179] This application can consider multiple objectives at the same time and find the most appropriate solution through a multi-objective optimization algorithm, which can help engineering designers make reasonable decisions while considering multiple objectives and improve the overall performance and competitiveness of the system.

[0180] In the embodiment of the present invention, after the system parameters, i.e., the initial values ​​of the parameters and the evaluation indexes of the initial parameters are set, the server sends them to the system parameter GCAir platform so that the GCAir platform performs simulation based on the system parameters, and the server calls the multi-objective optimization algorithm to continuously adjust the system parameters, i.e., the parameter optimization values, through the computer program programming interface API of the GCAir platform and drives the GCAir engineering simulation until the comprehensive evaluation function corresponding to the obtained simulation result meets the simulation requirements, and the parameter optimization values ​​are used as the target parameters of the system to be simulated. Thus, multi-objective parameter optimization is achieved.

[0181] Based on the multi-objective optimization method shown in the above embodiment of the present invention, the embodiment of the present invention also discloses a flow chart of another multi-objective optimization method, such as Figure 7 As shown, the method includes:

[0182] Step S701: obtaining initial parameters and selected evaluation indicators, wherein the number of the initial parameters is multiple;

[0183] Optionally, the initial parameters may be input by the user or obtained through processing by a multi-objective optimization algorithm. For details, see steps S41 to S42, in which the initial values ​​of the parameters in the initial parameters of the optimized parent generation are updated to obtain new initial parameters.

[0184] Step S702: driving the GCAir platform to simulate the system to be simulated based on the initial parameters to obtain simulation results.

[0185] Step S703: performing processing based on the evaluation index and the simulation result to obtain a comprehensive evaluation function;

[0186] Step S704: Determine whether the comprehensive evaluation function meets the simulation requirements. If so, execute step S705; if not, execute step S706:

[0187] Step S705: using the initial parameters for simulation as target parameters, and displaying the target parameters through the interactive module for users to view.

[0188] Step S706: Get the current number of optimization attempts.

[0189] In the specific implementation of step S706, the current solution, that is, the number of iterations of the initial parameters or parameter optimization values, that is, the number of optimization searches, is determined.

[0190] Optionally, each time the multi-objective optimization algorithm is called to optimize the initial values ​​of the initial parameters, the number of optimizations is increased by 1 and recorded.

[0191] Step S707: Determine whether the number of optimization attempts reaches a preset number of iterations. If so, the initial value of the initial parameter can be directly used as the target parameter. If not, execute step S708.

[0192] Step S708: calling its own multi-objective optimization algorithm to optimize the initial parameters and obtain parameter optimization values.

[0193] Step S709: Continue to drive the GCAir platform to simulate the system to be simulated based on the parameter optimization value, that is, return to step S703; until the comprehensive evaluation function corresponding to the obtained simulation result meets the simulation requirements, and use the parameter optimization value as the target parameter of the system to be simulated.

[0194] It should be noted that the specific implementation process of step S708 and step S709 is the same as the specific implementation process of step S506 and step S508 mentioned above, and they can refer to each other.

[0195] In the embodiment of the present invention, when the system parameters, i.e., the initial parameters and the evaluation indexes are set, the server sends them to the system parameter GCAir platform so that the GCAir platform performs simulation based on the system parameters. If it is determined that the comprehensive evaluation function does not meet the simulation requirements and the number of optimizations does not reach the preset number of iterations, the server calls the multi-objective optimization algorithm to continuously adjust the system parameters, i.e., the parameter optimization values, through the computer program programming interface API of the GCAir platform and drives the GCAir engineering simulation until the comprehensive evaluation function corresponding to the obtained simulation result meets the simulation requirements, and the parameter optimization values ​​are used as the target parameters of the system to be simulated. Thus, multi-objective parameter optimization is achieved.

[0196] An embodiment of the present application provides an electronic device, which includes a processor and a memory, the memory is used for program codes and data for multi-objective optimization, and the processor is used to call program instructions in the memory to execute the steps shown in the multi-objective optimization method in the above embodiment.

[0197] An embodiment of the present application provides a storage medium, which includes a stored program, wherein when the program is running, the device where the storage medium is located is controlled to execute the multi-objective optimization method shown in the above embodiment.

[0198] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can refer to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system or system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can refer to the partial description of the method embodiment. The system and system embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without creative work.

[0199] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in the above description according to function. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0200] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A multi-objective optimization method, characterized in that: Applied to a server, the server is connected to the GCAir platform and the interaction module respectively, and the method includes: Obtaining initial parameters and selected evaluation indicators based on the interactive module input; Based on the initial parameters, the GCAir platform is driven to simulate the system to be simulated to obtain a simulation result, wherein the system to be simulated is a pre-selected system engineering; Performing processing based on the evaluation index and the simulation result to obtain a comprehensive evaluation function; If it is determined based on the comprehensive evaluation function that the simulation requirements are not met, calling its own multi-objective optimization algorithm to optimize the initial parameters to obtain parameter optimization values; Based on the parameter optimization value, the GCAir platform is continuously driven to simulate the system to be simulated until the comprehensive evaluation function corresponding to the obtained simulation result meets the simulation requirement, and the parameter optimization value is used as the target parameter of the system to be simulated; The processing based on the evaluation index and the simulation result to obtain a comprehensive evaluation function includes: Obtaining a simulation value corresponding to each of the evaluation indicators in the simulation result; For each of the evaluation indicators, a simulation value corresponding to the evaluation indicator and a corresponding indicator weight are calculated to obtain a first value, wherein the indicator weight corresponding to each of the evaluation indicators is obtained by processing the indicator comparison value input in advance by the interaction module; Processing is performed based on the first numerical value corresponding to each evaluation index to obtain a comprehensive evaluation function.

2. The method according to claim 1, characterized in that Also includes: If it is determined based on the comprehensive evaluation function that the simulation requirements are met, the initial parameters used for the simulation are used as target parameters; The target parameters are displayed through the interactive module for users to view.

3. The method according to claim 1, characterized in that Determining based on the comprehensive evaluation function that the simulation requirement is not met includes: Determine whether the comprehensive evaluation function is greater than or equal to a preset parameter; If it is greater than or equal to, it is determined that the comprehensive evaluation function meets the simulation requirements.

4. The method according to claim 1, characterized in that: Also includes: Constructing a starting population based on the initial values ​​of the initial parameters and the optimization range corresponding to each initial value of the parameters; Calculate the corresponding fitness value based on each individual in the starting population; For the initial value of each initial parameter, sort them in descending order according to the fitness value, and determine the individual corresponding to the fitness value ranked first; Take the individual corresponding to the first fitness value in each ranking as the target solution; Determining whether the comprehensive evaluation function is the same as the target solution; If the comprehensive evaluation function is not the target solution, the step of calling its own multi-objective optimization algorithm to optimize the initial parameters to obtain the optimized values ​​of the parameters is executed.

5. The method according to claim 1, characterized in that Calling its own multi-objective optimization algorithm to optimize the initial parameters to obtain the optimized parameter values, including: Constructing a starting population based on the initial values ​​of the initial parameters and the optimization range corresponding to each initial value of the parameters; Calling its own multi-objective optimization algorithm to perform multi-objective optimization from the starting population to obtain offspring; The initial parameters are updated based on the offspring to obtain optimized parameter values.

6. The method according to claim 1, characterized in that If it is determined based on the comprehensive evaluation function that the simulation requirement is not met, the method further includes: Get the current number of optimization attempts; Determine whether the number of optimization searches reaches a preset number of iterations; If so, the initial parameters used for simulation are used as target parameters; If not, the process executes the step of invoking its own multi-objective optimization algorithm to optimize the initial parameters and obtain the optimized parameter values.

7. A multi-objective optimization device, characterized in that: It includes a server, a GCAir platform and an interaction module, wherein the server is connected to the GCAir platform and the interaction module respectively; The server obtains initial parameters and selected evaluation indicators based on the input of the interactive module; drives the GCAir platform to simulate the system to be simulated based on the initial parameters to obtain simulation results, wherein the system to be simulated is a pre-selected system engineering; processes based on the evaluation indicators and the simulation results to obtain a comprehensive evaluation function; if it is determined based on the comprehensive evaluation function that the simulation requirements are not met, calls its own multi-objective optimization algorithm to optimize the initial parameters to obtain parameter optimization values; based on the parameter optimization values, continues to drive the GCAir platform to simulate the system to be simulated until the comprehensive evaluation function corresponding to the obtained simulation results meets the simulation requirements, and uses the parameter optimization values ​​as the target parameters of the system to be simulated; The processing based on the evaluation index and the simulation result to obtain a comprehensive evaluation function includes: Acquire the simulation value corresponding to each of the evaluation indicators in the simulation result; for each of the evaluation indicators, calculate the simulation value corresponding to the evaluation indicator and the corresponding indicator weight to obtain a first numerical value, wherein the indicator weight corresponding to each of the evaluation indicators is obtained by processing the indicator comparison value input in advance through the interactive module; obtain a comprehensive evaluation function based on the first numerical value corresponding to each evaluation indicator.

8. An electronic device, characterized in that: The electronic device is used to run a program, wherein the program, when running, executes the multi-objective optimization method as described in any one of claims 1-6.

9. A storage medium, characterized in that: The storage medium includes a storage program, wherein when the program is running, the device where the storage medium is located is controlled to execute the multi-objective optimization method as described in any one of claims 1-6.

Images