AI Enabled Energy Consumption Predictor for Smart Buildings

Artificial intelligence has been taking the prior position in the technology domain. Unlike other technologies, smart energy management is one of the most likely topics for some specialists in parts of better comprehension of the unpredictable example of energy creation, dissemination and utility line. For the quicker and efficient prior energy demand prediction an AI enabled energy consumption predictor has been developed. A largely varying energy consumption demand within a city is the case where statistical analytics could improve the efficiency of the energy demand prediction. In this paper, data analytics methods have emphasized to aggregate false outliers within widely varying training data. The main focus of appropriate training data selection though statistical analysis could be able to increase the efficiency of the prediction score from 0.90 to 0.96. Various regression models have been implemented over a publicly available data set. Finally, a smart energy consumption predictor could be able to increase overall prediction accuracy up to 6.66%.