当前位置: X-MOL 学术Asian J. Control › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Finite-time command filtered control combined with pi-sigma fuzzy neural network for hydraulic control system
Asian Journal of Control ( IF 2.7 ) Pub Date : 2020-09-28 , DOI: 10.1002/asjc.2400
Wei Shen 1, 2 , Chao Shen 1 , Xiaoyu Su 3 , Jiehao Wang 1
Affiliation  

This paper considers the problem of the finite-time command filtered control for hydraulic motor system with external load disturbance using a new hydraulic transformer. Based on the backstepping design method and pi-sigma fuzzy neural network, a controller is designed. Firstly, the mathematical model of the hydraulic motor system is established, and the command filtered method is employed to solve the problem of the “explosion of complexity” caused by repeated differentiation of the virtual control signals in the backstepping design process, and the error compensation mechanisms are designed to reduce the errors produced by command filters. Then, the finite-time control approach is adopted to guarantee the tracking errors converge in finite time. In addition, the pi-sigma fuzzy neural network is used to approximate the nonlinear functions. The designed controller not only solves the strong nonlinear coupling but also enables the system to work efficiently with stronger robustness and faster error convergence. Finally, the effectiveness of the proposed controller is shown by comparisons with the traditional PID and pi-sigma fuzzy neural networks methods.

中文翻译:

液压控制系统结合pi-sigma模糊神经网络的有限时间指令滤波控制

本文考虑了采用新型液压变压器对有外部负载扰动的液压马达系统进行有限时间指令滤波控制的问题。基于反步法设计方法和pi-sigma模糊神经网络,设计了控制器。首先建立液压马达系统的数学模型,采用指令滤波的方法解决反步设计过程中虚拟控制信号反复微分引起的“复杂度爆炸”问题,并进行误差补偿。机制旨在减少命令过滤器产生的错误。然后,采用有限时间控制方法保证跟踪误差在有限时间内收敛。此外,pi-sigma 模糊神经网络用于逼近非线性函数。所设计的控制器不仅解决了强非线性耦合问题,而且使系统能够以更强的鲁棒性和更快的误差收敛速度高效工作。最后,通过与传统的 PID 和 pi-sigma 模糊神经网络方法进行比较,表明了所提出的控制器的有效性。
更新日期:2020-09-28
down
wechat
bug