A multi-level Demand-Side Management algorithm for offgrid multi-source systems. (15th January 2020)

• Record Type: Journal Article
• Authors: Roy, Anthony; Auger, François; Dupriez-Robin, Florian; Bourguet, Salvy; Tran, Quoc Tuan
• Is Part Of: Energy. Volume 191(2020)
• Publication Date: 2020-01-15
• Subjects: Demand side management -- Offgrid -- Multi-source system -- Marine renewable energy; Power resources -- Periodicals; Power (Mechanics) -- Periodicals; Energy consumption -- Periodicals; 333.7905
• DOI: 10.1016/j.energy.2019.116536
• Languages: English
• ISSNs: 0360-5442

Abstract

Abstract: The use of renewable sources for electricity supply of islands is faced with technical and economic constraints. To ensure that demand is fully met in the event of low generated power, batteries and gensets are often used and these can be expensive due to fuel import and battery costs. To provide more degrees of freedom for these offgrid networks, a multi-level algorithm based on several Demand Side Management strategies is proposed in this paper. The simulated case study concerns an island supplied by a multi-source system including solar, wind, tidal, wave energies and a battery storage solution. To limit the inconvenience for users, a hierarchical application of the proposed strategies is studied, according to a day-ahead forecast. Strategies based on anticipation are firstly applied for electric room heaters and water heaters in order to use the excess of generated power. For the most critical situations, strategies based on load shifting and load shedding are studied. In these cases, the best solution is found using a genetic algorithm. The application of the proposed Demand Side Management algorithm was found to help reduce the unmet load demand rate and adapt load demand according to the power generated. Highlights: A hierarchical application of different Demand-Side Management strategies is proposed. The anticipation of water heaters and electric room heaters demand is favoured. The most critical situations are scheduled with an optimization problem. Fully discharged battery occurrence is avoided by applying the proposed algorithm. Improvements are brought to the energy management of a 100% renewable energy system.