基于神经网络和遗传算法的车身骨架结构优化设计

Optimal Design of Bus Frame Structure on the Basis of ANN and GE

在对客车车身骨架结构进行有限元静、动力分析的基础上,利用神经网络对有限元分析得出的样本数据建立骨架结构设计参数(输入)与位移、应力及频率(输出)的全局性映射关系,获得遗传算法求解结构优化问题所需的目标函数值。最后,用遗传算法进行了车身骨架结构的优化设计。结果表明,基于神经网络和遗传算法的优化技术应用在汽车结构的优化设计中是有效、合理的。

With the development of optimal technology, optimal design of structure has been applied in the vehicle field. On the base of finite element method analysis on bus frame structure, a non-linear mapping function from multiple input data （design variables） to multiple output data （displacement, stresses and frequency calculated by FEM） was constructed within BP neural networks. It is necessary to obtain the objective function values in optimum design of structures by using genetic algorithms. Optimal design of structures was put forward by using genetic algorithms. The results showed that on the basis of ANN and GE, optimal technology applied in optimal design of bus structure was effective and reasonable.