On optimal scheduling of residential energy management systems

Abstract

Demand response is an attractive topic due to the opportunities offered by the coming of Smart Grid, and increasing concern of greenhouse gas emissions. To increase the responsiveness of demand, residential energy management systems (REMS) have been proposed to automate and schedule electric energy consumption within a dwelling. This work addresses several challenges and implications associated to REMS in the electricity grid. First, this work presents a framework for modelling residential electric load and tariff structures for creating optimal consumption schedules for electrical devices in a dwelling. Second, this work provides a framework for evaluating a multi-criteria performance of REMS, and compares the performance of REMS under various electricity tariff structures. Third, this work presents two scheduling algorithms that reduces the impact of scheduling errors and provides analysis on the improvement on computational tractability. Numerical simulations based on real-life data are provided. Each of the simulations includes optimizing the energy consumption patterns of several load types, subject to forecasting errors and variable electricity prices. The results allow us to describe the trade-offs between solution quality and various parameters.