Novel approach for reducing chattering effects in sliding mode control system

Abstract

The sliding mode controller (SMC) is a type of variable structure control system (VSCS), which is an authoritative tool for dealing with uncertainty, variations in parameter systems, nonlinear systems and external disturbances. Although significant advantages are associated with SMC such as robustness, the conventional sliding mode controller (CSMC) does not cover most of the requirements of the system, especially near the equilibrium point because of the high chattering which occurs as a result of high-speed switching (high frequencies of control signal near sliding line). This thesis is concerned with developing a novel controller and algorithms to reduce the effect of the chattering phenomenon, in order to achieve an efficient system performance. It includes three novel sliding mode concepts; sliding mode with state feedback controller (SMSFC), pre-programmed exponential sliding mode controller (PPESMC), and combination of nonlinear functions with sliding mode controller (CNFSMC). These are based on the SMC concept.The SMSFC is designed to reduce the effect of the chattering phenomenon that is present with the use of CSMC when noise and uncertainties occur. This is accomplished by refining the gain amplitude of CSMC, obtaining the convergence states properties of the system. The state feedback controller reformats and combines seamlessly with the CSMC to produce an integrated controller called a sliding mode with state feedback controller (SMSCF), whereas PPESMC relies on the value of an error signal and generates an exponential gain which is proportional to the error signal. Finally, a combination of nonlinear functions with sliding mode controller (CNFSMC) can be constructed from a combination of SMSFC and PPESMC. This method depends on two interrelated and nonlinear state-exponential properties. These new controllers have proved to be a robust and effective integrated control strategy for uncertain, varied-parameter, linear, and nonlinear systems, in addition to reducing the effect of the chattering phenomenon. Performance evaluations, comparisons, and analysis for the three methods (SMSFC, PPESMC and CNFSMC) for the SMC system are presented in this thesis, and their performance compared with the super-twisting (STW), boundary layer sliding mode (BLSMC) and low pass filter (LPFSMC) with SMC methods respectively when applied to a DC motor and robotics. The main conclusion drawn in this thesis was that the SMSFC as developed and implemented exhibited robust and high performance and trajectory tracking control given modeling uncertainties and noise. The evaluation and analysis were performed for different performance indexes and under different operational conditions. The results showed that under various external loads, external noise, and variations in system parameters SMSFC, PPESMC, CNFSMC, STW, BLSMC and LPFSMC with respect to reduction of the effect of the chattering phenomenon by 95%, 94%, 97%, 68%, 78% and 89% respectively.