Smart Energy Meter Analyzer Using IOT and Machine Learning

This paper presents a Smart Energy Meter Analyzer that integrates IoT (Internet of Things) and Machine Learning (ML) to monitor and classify power consumption in real-time. The system employs three current transformers (CTs) connected to an ESP32 microcontroller to measure current across three load phases, converting readings to power (W) and energy units (kWh). Data is transmitted via MQTT protocol to a cloud-based dashboard, which visualizes live consumption metrics (wattage, frequency, and units) and employs a Random Forest classifier to predict consumption classes: Low, Average, or High. The ML model, trained on a balanced dataset of 4 input features (watt1, watt2, watt3, unit), achieves accurate classification, enabling users to identify inefficiencies and optimize energy usage. Experimental results demonstrate the system’s effectiveness in real-world scenarios, with live hardware data updates and predictive alerts for high consumption. This work bridges IoT-based sensing with ML-driven analytics, offering a scalable solution for smart energy management.