A new approach for HEMS to optimize cost, emissions and comfort through a smart integration of V2G, load scheduling, and PVT generation

This paper introduces a new Home Energy Management System (HEMS) strategy designed for Smart Homes that performs load scheduling and contains Photovoltaic/Thermal (PVT) generation and Electric Vehicle (EV) using two approaches: one designed for mono-directional EV charger and the other for bi-directional one, which can perform vehicle-to-grid (V2G). The problem is solved using the Non-dominated Sorting Genetic Algorithm III metaheuristic and Long Short-Term Memory to predict the PVT generation. The occupant behavior applied utilizes 3 feature parameters to define the usage profile of controllable loads and determine the periods in which the user is most adapted to using the equipment to quantify their comfort, in addition to modeling the dependency between the operation of some loads. The electrical and thermal loads are categorized into non-controllable, deferrable, and thermo-controllable. An annual case study shows an average cost reduction of 14.57% achieved by leveraging the flexibility of the bidirectional charger for similar values of emissions and user comfort. This reduction occurs by exploiting time-of-use tariffs (which lead to an average savings of 22.23%) and reducing the maximum demand (resulting in an average reduction of 21.73%). These savings are sufficient to offset the increase in battery losses and degradation costs to perform V2G. Finally, the comparison of various HEMS architectures highlights the advantages of EV adoption through V2G implementation, positioning EVs as a more competitive solution for promoting clean and affordable energy in residential buildings.