Abstract

In order to improve energy sustainability and economic stability, it is essential to pursue efficient energy consumption forecast in order to achieve the dual goals of energy saving and power generation cost reduction. There is growing interest in leveraging, according to recent surveys. Machine learning methods for accurate energy consumption forecasts in smart homes. Because these algorithms rely on particular use cases and datasets, selecting them manually still presents a significant difficulty for data analysts and decision makers, even when these algorithms are evaluated using accuracy criteria. This work presents a revolutionary decision algorithm model that combines data mining and machine learning with the creative idea of picture fuzzy operators to address this problem. The technique of the model entails testing and training. Machine learning techniques for forecasting energy use, particularly with regard to meteorological data. The score values of Lasso Regression are used to decipher the complex patterns and characteristics present in meteorological data, particularly in the microclimates of smart homes. The paper also suggests a complex decision matrix that uses fuzzy operators to make algorithm aggregation and ranking via a score function easier. In the end, this study advances our knowledge of how much electricity is used by specific appliances as well as how much energy is used overall in smart homes, measured in kilowatts (KW).