Optimizing Energy Consumption and an IoT-Based System for Real-Time Monitoring and Management of Electricity Bills

This paper presents an IoT-based framework for real-time monitoring and intelligent management of household electricity consumption. The system integrates voltage, temperature, and current sensors with an ESP32 microcontroller, which applies a moving-average filter to smooth high-frequency noise while preserving meaningful usage patterns. The filtered data is segmented into fixed-length windows, from which both time-domain statistics and frequency-domain features-using discrete Fourier transforms-are extracted. These features are used to train a Support Vector Machine (SVM) to classify energy usage into “low”, “moderate‟, or “high” categories. Experimental evaluation shows high performance with 92.3 % accuracy, 91.8 % precision, 92. 0% recall, and a 91.9% F<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</inf>-score. Deployed in a three-month pilot, the system achieved an average update latency of 1.4 seconds and 99.7% uptime for the cloud service. A web and mobile dashboard provides users with real-time insights. The results demonstrate that the framework is practical for real-time monitoring, enabling energy-conscious decision-making and supporting sustainable electricity usage.