Robust pole assignment for synthesizing fractional-order control systems via neurodynamic optimization

In this paper, a neurodynamic optimization approach is proposed for robust pole assignment of fractional-order control systems. Compared with integral-order systems, the pole assignment of fractional-order systems is more challenging due to variability of stability region. The robust pole assignment is formulated as a constrained optimization problem, and a robustness measure is derived as a pseudoconvex objective function to be minimized. A recurrent neural network is employed for computing optimal solutions in real time. Simulation results are given to substantiate the efficacy and superiority of the proposed neurodynamic optimization approach.