Design of a Proposed Smart Plug Framework Based on IoT and Machine Learning for Energy Monitoring and Abnormal Consumption Detection

Design of a Proposed Smart Plug Framework Based on IoT and Machine Learning for Energy Monitoring and Abnormal Consumption Detection

azadeh askarifar1

1) Graduate in Hardware System Technology Engineering

IOI Code : ACCD-ACZKBZ
Publication : 10th International Conference on Applied Researches in Science & Engineering (carseconf.com)
Abstract :
    The proliferation of electrical appliances and the variability of energy consumption patterns have heightened the need for intelligent methods to monitor consumption at the individual device level. This research presents a framework for an IoT- and machine learning-based smart socket capable of measuring electrical parameters, transmitting data, and detecting anomalous consumption patterns. In the proposed architecture, the smart socket measures voltage, current, power, and energy, transmitting the data to the processing unit via an IoT platform. Following data preprocessing and feature extraction, an anomaly detection algorithm is employed to identify irregular consumption behaviors. MATLAB serves as the simulation environment for the framework, allowing for the generation of simulated data to evaluate the proposed method. The study focuses on presenting this architecture and the associated methodology.

Keywords : smart plug internet of things (iot) energy consumption anomaly detection machine learning energy monitoring