摘要
针对三维桥式吊车的防摇摆控制,基于径向基函数神经网络(Radial Basis Function,RBF)研究了一种自适应防摇摆控制算法。在神经网络控制中,如果持续激励(Persistent Excitation,PE)条件得不到满足,便不能保证权值收敛,执行相同的控制任务时,仍需对神经网络进行重复训练。所设计的自适应神经网络控制器不仅能够快速精确地定位负载,并能有效抑制吊车系统的摆动;同时在对于周期或回归轨迹的跟踪中,系统中信号的一致有界得到了证明,在稳定控制中实现了部分权值的收敛以及未知闭环动态的局部准确逼近,即确定学习。最后,通过仿真验证了所设计控制器的正确性和有效性,为桥式吊车系统的防摇摆控制提供了新的控制算法。
Aiming at the anti-sway control of three-dimensional overhead crane systems,an adaptive neural control approach is proposed by using radial basis function（RBF）neural networks. The convergence of neural weights cannot be guaranteed because of dissatisfying the persistent excitation（PE）condition,as a result,the training of the controller has to be repeated even for the same control task. The designed controller not only achieves accurate position tracking,short transportation time and sway suppression,but also achieves uniformly ultimately boundness of all signals in the control system,the convergence of partial neural weights and locally-accurate approximate of unknown system dynamics for periodic or recurrent tracking control,i.e.,deterministic learning. Finally,numerical studies indicate the effectiveness and correctness of the approach. A new control strategy is designed for anti-sway control for overhead crane systems.
作者
赖啸
刘勇
代艳霞
罗文果
LAI Xiao;LIU Yong;DAI Yan-xia;LUO Wen-guo(The Department of Modern Manufacturing, Yibin Vocational and Technical College, Sichuan Yibin 644003, China;Yibin Sanjiang Machinery Co., Ltd., Sichuan Yibin 644007, China)
出处
《机械设计与制造》
北大核心
2018年第6期114-117,共4页
Machinery Design & Manufacture
