A Cohesive Machine Learning-Based Model for Energy Consumption Prediction in Smart Homes

Abstract

Energy is the most important and costly resource and its plays a vital role in our daily lives. With the passage of time technologies are advancing, hence, the demand for energy uses is also increasing. In this work, a model has been proposed for energy consumption prediction in the smart home. The proposed model consists of four modules, namely data acquisition, data pre-processing, prediction, performance evaluation, and application. The pre-processing module has four sub-modules namely, output rectification, data cleaning, data transformation, and data reduction. The processed data is then fed to the prediction module, and different machine learning algorithms have been applied to the pre-processed data to predict energy consumption in smart homes. Next, the performance of these algorithms has been evaluated in the performance evaluation stage, in this stage different performance metrics have been considered, such as mean absolute error (MAE), mean absolute percentage error (MAPE), and root mean squared error (RMSE) in order to measure the performance of machine learning algorithms on the given data. The results indicate that the random forest algorithm performance is better as compared to other counterpart algorithms on the given data. The trained random forest algorithm is then used in the web-based interface in order to make able a layman to use the system for energy consumption prediction.