Nonlinear Coupled Tank System Control Using Optimal Evolutionary Neural Sliding Mode Technique

This study proposes the novel DANSMC-MDE algorithm for the coupled tank system (CTS) liquid level control, which consists of the Modified Difference Evolution (MDE) method for optimizing parameters of the Direct Adaptive Neuro Sliding Mode Controller (DANSMC). The CTS plant represents a nonlinear object with delay and uncertainties, including varying parameters, sensors and output valve noises, etc. The suggested controller contains a direct adaptive controller directly approximated by a Radial Basis Function (RBF) neural network combined with a sliding mode controller used to compensate for the errors of the RBF network and to ensure system stability. The Lyapunov stability criteria is used to construct both of sliding-mode control system and adaptive rule. Simulations are conducted to demonstrate the effectiveness of the proposed DANSMC-MDE control method. Furthermore, to demonstrate the superiority of the suggested control method, it is contrasted with the optimal SMC and the direct adaptive neuro sliding mode control (DANSMC) approaches. DOI: 10.61416/ceai.v27i3.9508