A method for the optimization of the drawbead geometrical parameters was built , which was based on the response surface method and genetic algorithm. 引入響應面方法和遺傳演算法建立基於近似模型的拉延筋幾何參數反求方法。
Classical response surface method approaches the structural stochastic response using a group of general polynomials, but this method can not guarantee convergence. 摘要傳統的響應面方法以一般多項式逼近結構的隨機響應，但這種方式並不能保證收斂性。
