Chattering Free Neuro-Sliding Mode Control of DC Drives

In this paper, a synergistic combination of neural network (NN) with sliding mode control (SMC) methodology is proposed. The main purpose is to eliminate the chattering phenomenon in sliding mode control. The reduction of the chattering is achieved by using a distance function that measures the distance between the trajectory of state errors and the sliding surface as the corrective control term instead of discontinuous sign function. The NN is utilized to estimate the equivalent control of SMC. The network weights are adjusted using a modified online error back propagation algorithm. The proposed scheme is applied to control the speed of a DC drive. Simulation verifies that the proposed control scheme has the advantage of less chattering in SMC