利用遗传算法优化的神经网络求解并联机构位置正解

Using Genetic Algorithm to Optimize Neural Network to Solve Parallel Mechanism's Forward Solution of Position

位置正解是研究并联机构运动学的基本问题,采用遗传算法优化神经网络权值的方法对模型进行了求解。利用位置反解求得训练样本,然后训练网络,最后对其进行仿真。结果表明,此算法运算结果快、求解简单、精度高,具有发展潜力。

Positional forward solution is the basic problems of the kinematics parallel mechanism. This paper uses the genetic algorithm to optimize neural network weights method to solve the model. Use the inverse position solution to acquire samples. Training the network and simulating it. The results show that this algorithm operation result quickly, the solution is simple, high precision and it ＇has great potential.