Adaptative state-space model predictive control of a parabolic-trough field

Model Predictive control algorithms for trough solar plants make use of solar radiation measurements and the overall efficiency knowledge which are very useful in order to reject disturbances. However, direct solar radiation, mirror reflectivity and metal absorptance which affect overall efficiency, can only be measured locally. In this paper, an adaptative model predictive control using an unscented Kalman filter (UKF) to estimate both the effective solar radiation and the metal–fluid temperature profiles is proposed. The control algorithm is validated by tests with real data taken from the ACUREX field of the PSA in Almería (Spain).

► A state-space model predictive control of a parabolic trough field is proposed.
► It makes use of effective solar radiation and the temperature profiles of metal and fluid.
► Global efficiency and temperatures profiles are very difficult to measure.
► We develop an unscented Kalman filter to estimate them.
► The algorithm has been tested with real data taken from ACUREX field.