Computational Comparison Between MPC and SR-MPC For Fast Dynamic System in Presence of Hard Constraints

MPC is an optimal control technique and has become a wide-spread solution in numerous industrial applications due to its constraints handling ability. However, MPC requires high computational time to compute the control signal because it uses the state space model of the system in optimization process and thus it suits to regulate the slow dynamic processes i-e petrochemical, fluid catalic cracking, temperature control etc. Therefore, in recent years researchers have proposed number of methods i-e Laguerre Functions based, Scale Reduction technique etc to reduce computational burden of conventional MPC and improve its feasibility for fast dynamic systems. In this research work computational comparison between MPC and Scale Reduction based MPC for a fast dynamic system i-e speed control of a DC motor in presence of hard constraints is presented. The simulations are done by first developing the state space model of DC motor and then simulating it in MATLAB. The results have been investigated in two modes. In first mode speed of DC motor is controlled by both techniques in absence of hard constraints and in second mode hard constraints are applied. The results have shown that Scale Reduction technique improves computational efficiency of MPC and make it feasible for the regulation of fast dynamic systems.