摘要
扫雷犁电液伺服系统是一类复杂非线性系统,基于传统建模方法构造的线性模型难以反映系统的本质非线性特征,因此提出采用基于正交最小二乘法的径向基函数神经网络对该系统进行精确建模;已有的手动分级控制或传统的PID控制方式不能满足精确控制的战术要求,提出采用基于广义训练和专门训练相结合的神经网络直接逆控制方法,实现系统吃土深度的有效控制。实验仿真结果以及与其他建模和控制方法的比较,验证了提出方法的有效性。
The electrohydraulic servo system of a certain mine sweeping plough is complex nonlinear system.It is difficult for the linear model obtained by first principle method to represent the essential nonlinear characteristics of the system,so the radial basis function neural network based on the orthogonal least square method is employed to model the system.It is also difficult to satisfy the tactical requirement of precise control by the existing manual control or the conventional PID control,so the neural network based direct inverse control combined with the generalized training and specialized training methods is used to realize effective control of the system.The experimental results and comparisons with other approaches clearly show the validities of the proposed techniques.
出处
《火力与指挥控制》
CSCD
北大核心
2010年第4期141-146,共6页
Fire Control & Command Control
基金
南京理工大学科技发展基金(XKF09003)
北京市教育委员会学科建设与研究生教育建设基金资助项目
关键词
电液伺服系统
神经网络
建模
逆控制
electrohydraulic servo system
neural networks
modelling
inverse control
