Publication Details

Abstract

Energy consumption in smart homes has increased significantly due to the rapid adoption of Internet of Things (IoT) devices and smart appliances. Accurate prediction of power consumption helps improve energy efficiency, reduce electricity costs, and support sustainable energy management. This paper presents a machine learning approach for predicting household power consumption using environmental and appliance usage data. A synthetic dataset representing temperature, humidity, appliance usage, and power consumption was generated and analyzed. A regression model based on scikit-learn Linear Regression was implemented to estimate power usage. Data visualization techniques were applied to analyze the relationship between environmental factors and energy consumption. Experimental results demonstrate that the proposed model effectively predicts power consumption using multiple features. The developed system can support intelligent energy management systems in smart homes.