A radial basis function(RBF)neural network evaluation algorithm ISSAGWO based on improved grey wolf optimiza-tion algorithm is proposed to address the issue of poor reliability and low efficiency of current reliability evaluation methods for thermal protection structures of hypersonic aircraft.Among them,the Grey Wolf algorithm was selected as the optimization method for tradi-tional RBF neural networks,and the population and convergence factor of the Grey Wolf algorithm were optimized to further improve the accuracy of reliability evaluation results.The results show that compared with other optimization algorithms,ISSAGWO has faster convergence speed,better search ability,and can quickly obtain the optimal value of the test function;Compared with other reliability evaluation methods,the ISSAGWO based reliability evaluation method has a significant advantage in convergence speed,and accurate evaluation results can be obtained with only a small amount of sample data.The above results indicate that the proposed RBF neural network reliability evaluation method based on the improved Grey Wolf optimization algorithm has good performance and can be ap-plied to the reliability evaluation of actual aircraft thermal protection structures with high reliability.
关键词
高超声速飞行器/热防护/可靠性评估/RBF神经网络/灰狼算法
Key words
hypersonic aircraft/thermal protection/reliability assessment/RBF neural network/Grey Wolf Algorithm
维普
万方数据