AI Based Smart Energy Saving and Consumption Management for Industries

Authors:

Abstract
This paper introduces an intelligent energy management system (EMS) designed to enhance energy efficiency in office, academic, and industrial environments through real-time human presence detection and IoT-based monitoring. Unlike conventional motion sensors, the proposed system employs image processing and machine learning algorithms implemented on a Raspberry Pi platform to accurately and contactless identify occupants. Based on detected occupancy, the system dynamically controls connected electrical loads via Arduino-based relays, minimizing unnecessary power usage and improving operational efficiency. In parallel, a real-time three-phase monitoring module was developed using PZEM-004T sensors and ESP8266 microcontrollers to measure voltage, current, and power factor across industrial networks. The data is processed locally and uploaded to a cloud dashboard for visualization, fault detection, and maintenance planning. Experimental validation demonstrates that both systems operate reliably under varying load conditions and significantly reduce energy wastage. The proposed dual-module approach—combining intelligent occupancy-based control with IoT-enabled monitoring—provides a cost-effective, scalable, and practical solution for modern energy management. It not only improves efficiency and safety but also establishes a foundation for future integration with renewable energy systems, predictive analytics, and privacy-aware smart building architectures aimed at promoting sustainability in industrial and institutional settings.