A hybrid approach for optimal energy management system of internet of things enabled residential buildings in smart grid

Abstract

Nowadays, a large number of smart devices in residential buildings is integrated with the evolution of internet of things (IoT), where it is necessary to effectively manage energy to meet the increase in demand. Therefore, in this paper, an optimal energy management strategy using the hybrid Gradient Boosting Decision Tree -Artificial Transgender Longicorn Algorithm (GBDT-ALTA) is proposed for IoT-enabled residential buildings. The main objective of the proposed approach is to minimize the electricity bill of customer, thereby Peak-to-Average Ratio (PAR) is reduced. Due to the scheduling process of residential electric devices, the proposed approach utilizes the waiting time threshold. Here, three types of appliances, such as shiftable electrical devices, thermostatically controlled electrical devices, and generally operated electrical devices are considered. Related to real-time price signals (RTPS) in utilities, scheduling the shiftable electrical devices is processed. The cost minimization with less acceptable time of waiting is achieved by GBDT-ATLA method, which considers the trade-off between the electricity cost and waiting time. Moreover, the proposed approach maintains the stability of the grid, because the stability of grid depends on the PAR. The proposed method is carried out in MATLAB/Simulink site and the simulated results prove that the better performance of GBDT-ATLA approach compared to the existing approaches, like Slime Mould Optimization (SMO), Chaos Game Optimization Algorithm (CGO), and Side-Blotched Lizard Optimization Algorithm (SBLO). The efficiency of the proposed approach under 100, 200, 500, and 1000 trails are 99.7300%, 99.6513%, 99.8363%, and 99.7916% respectively.