IOT DEVICE FOR ENERGY CONSUMPTION PREDICTION USING DEEP LEARNING TECHNIQUES

Abstract

The growing demand for electricity necessitates efficient energy monitoring and prediction systems. Conventional energy meters provide only cumulative readings and lack predictive capabilities, limiting their effectiveness in modern energy management. This paper presents an Internet of Things (IoT)-based energy consumption prediction system integrated with machine learning and deep learning techniques.

The proposed system utilizes sensors to measure electrical parameters such as voltage and current from household appliances. The collected data is transmitted to a cloud platform for processing and analysis. Machine learning models, including Random Forest, and deep learning models such as Long Short-Term Memory (LSTM) networks are employed to predict energy consumption at multiple time intervals, including 1 minute, 5 minutes, 15 minutes, 1 hour, and 1 day.

The results are visualized through a cloud-based dashboard, enabling users to monitor real-time and predicted energy usage. Experimental results demonstrate that the system provides reliable and consistent predictions with minimal deviation from actual values. The proposed approach offers a scalable and cost-effective solution for intelligent energy management in smart homes and small-scale applications.