A stochastic model based energy management system for off-grid solar houses

This paper discusses the challenges to net zero energy houses to the next level, which is to take them off the grid. This raises the stakes significantly as there is no reliance on the infinite storage that the grid provides, but only on the local energy storage. Overall reliability and resilience are the basic demands that need to be satisfied by the design and system sizing. We investigate a special aspect of power reliability, arguing that maintaining thermal comfort and guaranteeing power availability will be the two most important performance indicators in an off-grid house design. Due to the dynamic characteristic of a renewable energy generation system it is often difficult to maintain a stable thermal comfort level without investing in an oversized renewable power system. This paper presents a stochastic model based energy management system that distributes energy based on acceptable ranges of thermal comfort levels and manages electricity storage in batteries, driven by the projected energy demand and the projected energy generation. The objective of the stochastic model based control is to guide system operation and reach a balance between acceptable thermal comfort and general power reliability in an off-grid residence. Simulation results indicate that the proposed stochastic model based controller decreases the risk of power unavailability by up to 24% without sacrificing indoor thermal comfort dramatically. A follow-up sensitivity analysis demonstrates versatilities of the controller in practical application. Certain limitations in its application are discussed at the end of this paper. Published by Elsevier Ltd.