Active demand side management for households in smart grids using optimization and artificial intelligence

This work aims to develop a methodology to perform the active demand side management for households in smart grids, which contain distributed solar photovoltaic generation and energy storage. Such methodology outcomes a decision-making system that manages the battery aiming to reduce the consumer electricity cost. It also contributes to postpone the investments in expansion of the electricity grid if the higher loading period coincides with the higher electricity tariff of the day. The decision-making system is a validated neural network, trained with optimized data, which can be used in any household meeting certain conditions - specific location and electricity tariff, and consumption profile like to the standard verified by the local electricity utility. To validate this methodology, it was created three consumption and three solar generation profiles, which were combined to each other. The results show that the ANN-based decision-making system operates the battery efficiently to achieve the minimum electricity bill.