DESIGN OF DISTANCE-BASED FUZZY SLIDING MODE CONTROLLER WITH A VARYING CONTROL LAW

• Main Authors: Min-Cheng Yang, 楊旻政
• Other Authors: Prof. Hung-Chin Lu
• Format: Others
• Language: zh-TW
• Published: 2003
• Online Access: http://ndltd.ncl.edu.tw/handle/78524209871772853350

碩士 === 大同大學 === 電機工程研究所 === 91 === In this thesis, a distance-based fuzzy sliding mode controller with a varying control law is proposed. In sliding mode control, the most important drawback is control chattering. To eliminate chattering, boundary layer or fuzzy logic control usually is applied into sliding mode control. For fuzzy sliding mode control, a fuzzy rule table is generally constructed in a two-dimensional input space but it requires a huge number of control rules. A new variable called the signed distance, which is a sole fuzzy variable in fuzzy logic control, is known as is introduced. Using the as control input, the total number of rules is greatly reduced and the tuning of control rules becomes easy. In sliding-mode control, the use of a boundary layer is a common technique to reduce chattering of the control signal, but robust performance, accurate control results are compromised. In order to increase robustness and accuracy as the system approaches steady state, the saturation functions are dynamically adjusted. Thus, distance-based fuzzy sliding mode controller with a varying control law is proposed. By using this method, fuzzy rules can be constructed in a one-dimensional space to reduce control rules and we can change the slope of the control law and the boundary layer width depending on distance, which is between states and the sliding surface. Furthermore, the stability is guaranteed by using the Lyapunov theorem. As a result, the system response, robustness, and accuracy are improved. According to aforementioned concepts, both theoretical analysis and simulation studies are carried out via computer simulations using a cart-pole system and a ball-beam system to demonstrate the effectiveness of the method.