IoT-based demand-side energy management: Enhancing peak hour efficiency through automated control of appliances

Abstract

Industrial solar microgrids experience pronounced peak demand due to aggregated production processes, auxiliary systems, and shift-based operation. Conventional demand response approaches rely on voluntary participation and aggregated energy metrics, which limits their effectiveness in industrial environments. This study presents an IoT-based demand-side energy management framework for industrial microgrids that explicitly distinguishes active power, reactive power, and apparent power during peak operation. The system integrates real-time power factor monitoring, automated reactive power compensation, and prioritized load control using low-latency IoT communication. A high-fidelity simulation model was developed in Proteus 8.15 and validated using a physical IoT prototype calibrated against industrial measurement data. Performance evaluation combined experimental measurements with historical load data from an operating industrial facility. Automatic power factor correction reduced apparent power demand by 41.58% due to reactive power mitigation. Real power decreased by 1.6 and 2.6% for both hardware and simulation respectively. The result shows that power factor correction reduces current and releases inverter capacity. It does not produce large reductions in real energy consumption. Automated shedding of non-critical loads reduced real power demand by 23.46% during peak periods. The non-critical loads include auxiliary lighting and support equipment rated at approximately 0.063 kW. These loads represent 4 percent of the monitored site load. Classification followed production continuity, safety requirements, and operational redundancy. Load factor increased by 23.4%, and both peak demand and peak-to-average ratio decreased by 20% relative to baseline operation. The results demonstrate that automated industrial demand-side energy management improves electrical performance and peak demand characteristics without reliance on voluntary user response. The proposed framework provides a practical foundation for power-quality-aware demand management in industrial solar microgrids.