Robust Adaptive Neural Network Control for Nonlinear CSTRs with Prescribed Control Performance and Event-triggered Inputs

This paper presents a robust adaptive neural network event-triggered controller (NN ETC) for nonlinear continuous stirred tank reactors (CSTRs) with prescribed performance. The proposed controller employs a globally invertible prescribed performance function to ensure the tracking error remains within predefined bounds. A radial basis function (RBF) NN approximates the unknown nonlinear system dynamics, reducing reliance on precise modeling. A switching threshold event-triggering mechanism (SWT-ETM) is adopted that dynamically switches between fixed and relative thresholds to reduce communication and actuation frequency while avoiding Zeno behavior. Stability analysis proved that all system signals are globally bounded and the tracking error exponentially converges to an adjustable neighborhood. Simulation results demonstrate the effectiveness of the proposed method.