摘要
惯性参数大范围变化和低速状态下的非线性摩擦是制约机械伺服系统跟踪性能的主要因素,基于LuGre动态摩擦模型和干扰观测器的补偿控制可以实现非线性摩擦力矩的动态补偿,但状态观测器的设计是基于被控对象的数学模型,当负载惯性参数大范围变化时,上述控制系统性能无法保障,针对上述问题提出一种基于模糊神经网络补偿的状态观测器复合控制,分析了基于模糊神经网络补偿复合控制的理论与实现方法,并以直流电机飞行仿真转台作为被控对象进行了仿真试验,试验结果表明了控制方法的有效性。
Inertial parameters varying in large range and nonlinear frictional torque are the main factors that lead to mechanical servo system tracking performance declination.The LuGre dynamic f riction model and the disturbance observer realize the dynamic compensation of nonlinear friction torque.When servo system inertial parameters varying in large range,the tracking performance can not be guaranteed because of the design of disturbance observer based on the mathematic model of control system.A complex control scheme is proposed to deal with the above problems.Fuzzy-neural network is used to compensate the Inertial parameters varying of mechanical servo system.The theory and realizing way of the propsed method are analyzed.Experimental results for a DC motor servo-system show the effectiveness of the proposed method.
出处
《控制工程》
CSCD
北大核心
2010年第2期146-148,153,共4页
Control Engineering of China
基金
国家自然基金资助项目(E06A30020)
关键词
机械伺服系统
干扰观测器
模糊神经网络
摩擦力矩
补偿
跟踪控制
mechanical servo systems
disturbance observer
fuzzy-neural network
frictional torque
compensation
tracking controll
