Energy Internet In Building Energy Management System (Ei-Bems)

ASM Sc. J., 21(1), 2026

Published on June 26, 2026

https://doi.org/10.32802/asmscj.2026.1002

Author: Sim Sy Yi, Loh Wei Kiat, Siti Nur Afifah Mohd Suhaimi, Alvin John Lim Meng Siang, Husam S. Samkar, Muhamad Azfar Azri

Abstract

Malaysia’s National Energy Transition Roadmap (NETR) identifies Energy Efficiency as one of the six key levers for achieving a sustainable energy future. Commercial buildings account for approximately 26–27% of national electricity consumption, mainly due to lighting and air- conditioning systems. Conventional Building Energy Management Systems (BEMS) commonly employ Proportional–Integral–Derivative (PID) controllers because of their simplicity. However, PID controllers are less effective in handling the non linear and dynamic characteristics of real building environments. To address this limitation, this paper proposes an Artificial Neural Network (ANN)-based controller integrated into an Energy-Internet Building Energy Management System (EI-BEMS). ANN and PID controllers were developed and evaluated in MATLAB Simulink for lighting and HVAC subsystems. The results show that the ANN controller achieved superior predictive accuracy, with regression coefficients R = 0.94 for HVAC and R = 0.99 for lighting, while also providing faster response times, lower overshoot, and improved control stability compared to PID control. The novelty of this study lies in demonstrating ANN as an intelligent alternative to PID for EI-BEMS applications, enhancing both energy efficiency and indoor comfort. These finding s support Malaysia’s NETR energy efficiency initiative and align with United Nations Sustainable Development Goal (SDG) 7.3, which aims to double the global rate of energy efficiency improvement by 2030.

Keywords: AI integration, building energy management system, cost-effectiveness, energy efficiency, energy transition, sustainable future