An execution, monitoring and replanning approach for optimal energy management in microgrids

This work develops a new approach for optimal energy management of electrical distribution ‘smart-grids’. Optimality aims at improving sustainability through the minimization of carbon emissions and at reducing production costs and maximizing quality. Input data are the forecasted loads and productions from renewable generation units, output data are a set of control actions for the actuators. The considered electrical distribution system includes storage units that must be considered over a 24 h time interval, to consider an entire charge and discharge cycle. The objectives for the optimal management of distributed (renewables and not) generation are technical, economical and environmental. It is thus required to solve a multi-objective optimization problem over a 24 h time interval considering the uncertainty associated to weather conditions and loads profiles. The novelty of the proposed approach resides in considering the optimal scheduling of generation units an automatic planning process in a dynamic, non-deterministic and not fully observable environment, as it is, getting closer to actual conditions. The system proposed here is a planning and execution scheduler which allows the central controller to monitor the execution of a scheduling plan, interrupt the monitoring to input new information and repair the plan under execution every time interval.