Design of Super Twisting Algorithm for Chattering Suppression in Machine Tools

Sliding mode control is known for its robustness against plant uncertainties and disturbance rejection. An extension of this technique is the addition of super twisting algorithm that suppresses the chattering effect commonly observed with sliding mode control application. This paper elaborates on the design and analysis of the super twisting algorithm with focus on optimizing the gain parameters. Two main components of the sliding mode super twisting controller were designed and validated; that is, the switching function and the control laws. The traditional sliding surface was used as the switching function while control laws with continuous control action were applied. Two gain parameters of the control laws were designed and analyzed in terms of tracking error reduction and chattering effect. A Kalman-Bucy continuous filter was applied for velocity signal estimation. Fast Fourier Transform was applied on the control inputs to investigate the chattering amplitude and root mean square of the tracking error was used as performance indicator for tracking error reduction. Analyses were performed on the effect of gains variation to control performance. Experimental results showed that gain L was sensitive to chattering amplitude while gain W provided wider range of tuning. The control performance was compared to the traditional pseudo-sliding mode controller in terms of tracking error reduction and chattering effect. Results showed that the super twisting algorithm produced superior results in tracking error reduction and was effective in suppressing chattering effect compared to the pseudo-sliding mode controller.