论文部分内容阅读
螺旋输送器的动态特性设计可归结于特征值反问题的求解。针对结构参数到结构响应之间的非线性映射关系,通过一种基于神经网络代理模型的优化策略,采用正交试验设计在设计空间中选择初始样本点,构造神经网络代理模型,神经网络结合遗传算法求解,利用神经网络的非线性拟合能力和遗传算法的非线性寻优能力,引入训练后的BP神经网络预测结果作为个体适应度值,获得全局最优值及对应输入值。解决了遗传算法能获全局最优解与有限元大量结构重分析之间的矛盾,是结构反问题的一种有效求解策略。
The dynamic design of the screw conveyor can be attributed to the eigenvalue inverse problem. Aiming at the nonlinear mapping relationship between structural parameters and structural response, an orthogonal experimental design was used to select the initial sample points in the design space through an optimization strategy based on the neural network proxy model. The neural network proxy model was constructed, and neural network combined with genetic By using the nonlinear fitting ability of neural network and the nonlinear optimization ability of genetic algorithm, the prediction result of BP neural network after training is introduced as individual fitness value to obtain the global optimum value and the corresponding input value. It solves the contradiction between the global optimal solution of genetic algorithm and the finite element analysis of a large number of structures. It is an effective solution to the structural inverse problem.