Modelling and control of multi-energy systems through Multi-Prosumer Node and Economic Model Predictive Control

• Published in: International journal of electrical power & energy systems 2020-06, Vol.118 (105778), p.105778, Article 105778
• Main Authors: Blaud, Pierre Clément, Haurant, Pierrick, Claveau, Fabien, Lacarrière, Bruno, Chevrel, Philippe, Mouraud, Anthony
• Format: Article
• Language: English
• Subjects: Automatic Control Engineering; Computer Science; Economic Model Predictive Control; Energy hub; Engineering Sciences; Mechanics; Multi-energy system; Multi-Prosumer Node; Thermics
• ISSN: 0142-0615; 1879-3517
• DOI: 10.1016/j.ijepes.2019.105778

•MPN state-space model and power balance considers different energy carriers.•Energy systems, like converters, loads, storages, renewable sources are considered.•MPN deals with system dynamics, this bring minutes scale optimisation capabilities.•Optimisation is formulated without integer number.•MPN deals on-grid or off-grid systems, and energy flows can be bidirectional. The present study deals with Multi-Energy Systems (MES) modelling and advanced control with Economic Model Predictive Control (EMPC). MES provide energy flexibility, efficiency, and adaptability thanks to several energy carriers. MES are identified as a lever for integrating renewable energy. A MES novel formulation technique called Multi-Prosumer Node (MPN) is developed in this paper. MPN makes possible the modeling of MES, considering MES dynamics, several energy carriers, converters, on-grid, and off-grid. In addition, this MES modeling approach is compatible with predictive control strategies like the EMPC. In fact, EMPC is able to take into account loads, weather, renewable power and energy grid cost predictions to minimise economic costs. A real case study is implemented to examine MPN capabilities, which it is composed of renewable generators, loads, storages from two-energy carriers. Two real scenarios have been developed in order to represent realistic winter and summer cases. Simulation results, thanks to modelling with MPN and EMPC advanced control, demonstrate that the node is optimally controlled, devices dynamics are considered on a minute scale, and energy conversion from one carrier to another one is taken into account while economic cost minimisation is performed. The gained results indicate that the presented MPN modelling and optimisation approach reduces economic cost by 8.21% in winter case and 84.24% in summer case compared to the benchmarks which are composed of rule-based control.