Periodic constraint-tightening MPC for switched PV battery operation

• Published in: IET control theory & applications Vol. 12; no. 15; pp. 2010 - 2021
• Main Authors: Hashikura, Kotaro, Namba, Kazuki, Kojima, Akira
• Format: Journal Article
• Language: English
• Published: The Institution of Engineering and Technology 16.10.2018
• Subjects: charge–discharge schedules; constraint‐tightening gain; constraint‐tightening technique; design freedom; MPC design; operational limits; periodic constraint‐tightening MPC; periodic operation; photovoltaic batteries; photovoltaic power systems; prediction error; prediction uncertainty; predictive control; primary references; primary schedule; PV battery operation problem; PV battery systems subject; reference switching; Research Article; secondary reference; secondary references; solar irradiance; supply–demand balance; switched PV battery operation; switching model predictive control; weather changes; supply–demand balance; constraint-tightening technique; PV battery operation problem; MPC design; secondary reference; switched PV battery operation; primary references; periodic operation; photovoltaic batteries; charge–discharge schedules; photovoltaic power systems; PV battery systems subject; weather changes; secondary references; switching model predictive control; solar irradiance; reference switching; operational limits; prediction uncertainty; primary schedule; constraint-tightening gain; periodic constraint-tightening MPC; prediction error; predictive control; design freedom
• ISSN: 1751-8644; 1751-8652
• DOI: 10.1049/iet-cta.2017.1174

Charge–discharge schedules of photovoltaic (PV) batteries are planned based on the prediction of solar irradiance, and various methods have been studied to cope with prediction uncertainty. Among them, the switching of charge–discharge schedules could be a measure for keeping supply–demand balance against abrupt weather changes. Those methods motivate a PV battery operation problem to retain the trackability of the secondary reference even when the primary schedule is ongoing. This study addresses the periodic operation of PV battery systems subject to the reference switching, and designs a switching model predictive control (MPC) to track the primary and secondary references with guaranteed feasibility. The proposed MPC design also incorporates the constraint-tightening technique, by which the operational limits are kept under the prediction error associated with each of the schedules. In order to utilise the design freedom in compensating the prediction error, the authors construct the constraint-tightening gain based on economic dispatching control, and discuss the features in terms of design simplicity and fuel cost efficiency.