Hierarchical control for the start-up procedure of solar thermal fields with direct storage. (February 2020)

• Record Type: Journal Article
• Authors: Gil, Juan D.; Roca, Lidia; Zaragoza, Guillermo; Normey-Rico, Julio E.; Berenguel, Manuel
• Is Part Of: Control engineering practice. Volume 95(2020)
• Publication Date: 2020-02
• Subjects: Process control -- Solar energy -- Hierarchical control -- Model predictive control; Automatic control -- Periodicals; 629.89
• DOI: 10.1016/j.conengprac.2019.104254
• Languages: English
• ISSNs: 0967-0661

Abstract

Abstract: Thermal energy storage tanks are habitually combined with solar thermal fields to improve the dispatchability of these facilities. From the dynamical point of view, the start-up phase is relevant since if the storage device is unloaded in terms of energy or widely stratified, the transitory regime can take long time until reaching the operating point. In this paper, an optimal real-time procedure based on a hierarchical controller for improving the start-up phase is proposed. The hierarchical controller is composed of two layers based on a Model Predictive Control (MPC) technique and Proportional Integer Derivative (PID) controllers. Real experimental tests were performed in a pilot facility located at Plataforma Solar de Almería (Almería, Spain). In addition, a comparison in simulation with the typical manual procedure and with two techniques proposed previously in the literature for the same plant is provided. The results demonstrate the benefits obtained by using the proposed method; since it reduces the start-up phase in 34 [min] in comparison with the manual operation, and in 26 and 6 [min] with respect to the two previous techniques. Highlights: This paper proposes an optimal start-up policy for solar thermal fields. The procedure is based on a hierarchical controller that uses a real time optimizer. Experimental tests were conducted in a real facility with direct storage. A simulated comparative analysis with other start-up techniques is also provided. The start-up stage time can be reduced up to 10.5% with the proposed methodology.