柔性机械臂的智能极值响应面优化设计方法

Intelligent Extremum Response Surface Optimization Design Method of Flexible Manipulator

在满足柔性机械臂系统可靠性前提下,为了节省制造用料,提出了智能极值响应与分阶层粒子群算法相结合的优化设计方法。BP神经网络与极值响应面相结合的智能极值响应面,可以有效计算机械系统可靠性;给出了优化量灵敏度计算方法,选择连杆横截面长与宽作为优化量;建立了连杆可靠性优化设计模型;采用了分阶层粒子群算法解算此优化模型,结果显示,二次项极值面法优化后横截面积减小了0.21%,而智能极值面法优化后减少了2.5%;与传统粒子群算法相比,分阶层粒子群算法优化的连杆横截面积更小。

To save the manufacturing material on the premise of satisfying flexible manipulator reliability,manipulator design optimization method combined intelligent extremum response and stratified particle swarm algorithm is proposed.Intelligent extremum response combining BP neutral network and extremum response can calculate mechanical system reliable validly.Optimization sensitivity calculation method is given,and long and wide of link intersecting surface are chosen as optimizations.Reliability design optimization model of link is built.Stratified particle swarm algorithm is applied to solve the model.The result shows that intersecting surface area optimized by quadratic term decrease by 0.21%,and 2.5%by intelligent extremum response surface.Compared with traditional particle swarm algorithm,link intersecting surface area is smaller optimized by stratified particle swarm algorithm.