Sub-optimal Linear Parameter Varying Model Predictive Control for Solar Collectors

Abstract

This short paper investigates the temperature control of a flat-plate water-heating solar collector. This nonlinear system is modelled via a quasi-linear parameter varying setting. To address this control problem, a model predictive control algorithm is formulated, considering a frozen guess for the evolution of the scheduling parameters, set-sequence constraints and a Lyapunov-decreasing terminal cost. The advantage of this method is that it uses standard quadratic programming problems and does not have to resort to nonlinear optimization. Through simulation, it is demonstrated that it can yield successful performances.