A multi-objective optimization method for aircraft based on adaptive surrogate model

Abstract

The invention discloses an aircraft multi-objective optimization method based on a self-adaptive agent model, relates to a multi-objective optimization method for treating complex aircraft design, and belongs to the field of aircraft design optimization. According to the aircraft multi-objective optimization method, an integrated preference function is constructed by use of a physical planning method to realize the conversion of the multi-objective optimization problem into a single-objective optimization problem reflecting design preference; next, the self-adaptive agent model is constructed from the integrated preference function and constraint conditions to take the place of a high-accuracy analysis model, and therefore, the problem of great time taken in calculation of optimization design is solved; finally, the constraint problem is converted into a non-constraint problem by use of an augmentation Lagrange multiplier method, and the non-constraint problem is solved by use of a genetic algorithm. The aircraft multi-objective optimization method has the advantages that the solving process of the aircraft multi-objective optimization method taking much time in calculation is simple and efficient, and therefore, a Pareto noninferior solution meeting the requirements of a user can be obtained quickly to shorten the design period of the aircraft, and the design cost is reduced. Besides, the aircraft multi-objective optimization method is high in universality and convenient for program development.

Description

technical field [0001] The invention relates to a multi-objective optimization method for processing complex product design, in particular to a multi-objective optimization method for processing complex aircraft design, belonging to the field of aircraft design optimization. Background technique [0002] In aircraft engineering design, it is often necessary to comprehensively consider various performance indicators of the aircraft system, and to deal with multi-objective optimization problems. In addition, it will consume a lot of time to call high-precision subject analysis models for performance calculations. Therefore, in the multi-objective optimization design of aircraft The existing weights are difficult to choose and time-consuming, and a set of practical optimization design methods is needed for design guidance. [0003] In dealing with multi-objective optimization problems, the commonly used methods are Pareto (Pareto) evolutionary algorithm, weighted coefficient me...

AI Technical Summary

Problems solved by technology

[0044] The present invention constructs a comprehensive preference function through the physical programming method, realizes the transformation of the multi-objective optimization problem into a single-objective optimization problem reflecting the design preference, and then constructs an adaptive proxy model for the comprehensive preference function and constraint conditions, replacing the high-precision analysis model, and solves the problem of In order to solve the problem of time-consuming calculation of optimal design, the augmented Lagrange multiplier method was used to transform the constrained problem into an unconstrained problem, and the genetic algorithm (GA) was used to solve the problem

Images