结合面法向和切向接触刚度的MPSO-BP神经网络算法的建模  

Machine joints normal and tangential contact stiffness modeling based on MPSO-BP neural network algorithm

作　　者：杨红平[1] 傅卫平[1] 王雯[1] 师彪[2] 杨世强[1] 

机构地区：[1]西安理工大学机械与精密仪器工程学院,西安710048 [2]西安理工大学水利水电学院,西安710048

出　　处：《仪器仪表学报》2012年第8期1856-1861,共6页Chinese Journal of Scientific Instrument

基　　金：国家973计划(2009CB724406);国家重大科技专项课题(2009ZX04014-32)资助项目

摘　　要：提出一种基于改进的粒子群和BP神经网络相结合的算法模型,用改进粒子群优化算法优化BP神经网络参数。以机械结合面法向接触刚度和切向接触刚度作为算例。考虑结合面配对副材料、接触载荷、表面加工方法、表面粗糙度和结合面间的介质作为主要因素,对8组结合面法向和切向接触刚度进行预测建模,并对仿真与实验结果进行比较与误差分析。结果表明,该方法实现了多种影响因素组合下的机械结合面法向和切向接触刚度较高精度的建模和预测。An algorithm model combining the modified particle swarm optimizer （MPSO） and the BP neural network is proposed. The normal and tangential contact stiffness of machined joints is influenced by many factors such as joint material, surface machining method, interface pressure, contact roughness and interface medium. The normal and tangential contact stiffness of 8 groups of machined joints is forecasted under certain experimental conditions. The simulation and experimental results are compared and error analyzed. The results show that the method is high precision to achieve the normal and tangential contact stiffness modeling and forecasting of machined joints under various affecting factors.

关 键 词：改进PSO—BP神经网络算法 法向和切向接触刚度 预测 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置] TH123[自动化与计算机技术—控制科学与工程]