Cost-optimal model predictive scheduling of freezers. (November 2018)

• Record Type: Journal Article
• Authors: Bálint, Roland; Fodor, Attila; Hangos, Katalin M.; Magyar, Attila
• Is Part Of: Control engineering practice. Volume 80(2018)
• Page Start: 61
• Page End: 69
• Publication Date: 2018-11
• Subjects: Smart grids -- Demand side management -- Model predictive control -- Heuristics -- Scheduling algorithms; Automatic control -- Periodicals; 629.89
• DOI: 10.1016/j.conengprac.2018.08.009
• Languages: English
• ISSNs: 0967-0661

Abstract

Abstract: A cost-optimal model predictive scheduling algorithm is presented that operates in a day-ahead market. The underlying optimizer is a heuristic branch and bound algorithm that finds the constrained optimal scheduling of a freezer with respect to hourly changing energy price. The method is also able to iteratively re-estimate the heat capacity of the freezer. Simulation experiments were performed on a freezer model identified from measurement data. Results show that the proposed algorithm successfully decreased the cost of operation, however the computational complexity increases when the price is growing. The proposed method can be generalized for home appliances of different kind. Highlights: Demand side management in a day-ahead market is important. Freezers and other home appliances are potential applications. Model predictive control with a heuristic branch and bound optimizer is applied. Online estimating the key model parameters yield a better result. Falling price—easy to be optimal; Rising price—not trivial.