Authors: Assistant Professor Sourabh Jain, Prachi Gupta, Manish Yadav, Pooja Srivastava
Abstract: In this day and age when we need to manage resources prudently, accurately predicting how much energy one would require is an extremely important task. In this abstract, we showcase how the application of advanced technologies – machine learning, data mining, and artificial intelligence techniques can be blended with energy management systems to enhance the efficiency of forecasting energy consumption rates. The sample we use in this research contains a wide array of data, including casual and seasonal weather data, time, building occupancy figures, as well as the figures attained for energy consumption during the various time slices. Some of the various approaches to solve the problem we are working on that we analyze include: linear regression, decision tree regression, random forest regression, and artificial neural networks. It is vital to accurately predict future power consumption considering factors like resource optimization and sustainable energy management. This work describes an approach that uses advanced methodologies in machine learning techniques for precise forecasting based on historical data alongside a variety of descriptive features to predict energy consumption within the foreseeable future. In this research, we use an extensive dataset containing weather data, timestamps, occupancy statistics, and previous energy consumption data. We apply many algorithms that include linear regression, decision trees, random forests, and neural networks to energy consumption prediction and analyze which model best performs the prediction task.
