Aircraft optimization method based on mixed radial basis function neural network

Abstract

The invention discloses an aircraft optimization method based on a mixed radial basis function neural network, and belongs to the field of aircraft design optimization. The method comprises the following steps: constructing initial sample points by adopting a Latin hypersquare test design method, randomly grouping the initial sample points into a training sample point set and an evaluation sample point set, training the neural networks of different radial basis function types by using the training sample points, solving the weight coefficients of the neural networks of different radial basis function types by using the evaluation sample points and the generalized inverse matrix, and weighting, so as to fully utilize the advantages of different types of radial basis functions to improve the approximation precision and effectively reduce the construction risk of the mononuclear radial basis function neural network caused by lack of prior information, and finally, optimizing the constructed mixed radial basis function neural network in combination with an intelligent optimization algorithm so as to realize rapid optimization of aircraft system performance. The method has important significance in the aspects of relieving the calculation time consumption of aircraft optimization, improving the optimization efficiency and the like.

Description

technical field [0001] The invention relates to an aircraft optimization method based on a mixed radial basis function neural network, belonging to the field of aircraft design optimization. Background technique [0002] With the development of computer software and hardware technology, high-precision simulation analysis models have been widely used in the field of aircraft overall design, such as finite element analysis (FEA), computational fluid dynamics (CFD), etc. Although the high-precision analysis model improves the analysis accuracy and design confidence, it also significantly increases the calculation cost of the simulation. For example, the CFD model of an aircraft usually takes several hours to complete an aerodynamic simulation analysis. Because traditional optimization methods (such as genetic algorithm, particle swarm optimization algorithm, etc.) often need to directly call thousands of analysis models to realize the exploration of the design space, the optimi...

AI Technical Summary

Problems solved by technology

Considering that the aircraft analysis model is a "black box" function with an unknown mathematical structure, it is difficult to determine the optimal radial basis function type in advance

Images