Limited Complexity Predictive Controller for Load Management in Isolated Photo-voltaic Systems

Stand-alone photovoltaic systems are suitable electrification solutions for isolated buildings. Those systems rely exclusively on solar radiation, a source that cannot be controlled, therefore an improper operation of the system may lead to service interruptions and reduction of lifespan. Energy management systems allow proper scheduling of sources, storage, and load consumption, usually through complex optimal predictive control strategies. This paper proposes a novel load management system, based on a simplified model predictive controller aimed at minimizing service interruptions while guaranteeing a specific depth of discharge of the battery. The proposed solution avoids the use of complex optimization solvers and/or forecasting algorithms by introducing proper approximations to the model that lead to a simple decision algorithm based on open-loop dynamics simulations and priority-based schedules. Simulations and Hardware in the Loop tests show that the proposed controller can be implemented on low-cost microprocessors, providing adequate performance and satisfying the constraints on maximum load curtailment and depth of discharge.