Optimization back propagation algorithm with feature engineering to predict building hourly consumption energy

Optimizer function in Back propagation algorithm are responsible for reducing the losses and to provide prediction the most accurate results. An Back propagation algorithm with feature engineering was used to predict hourly energy consumption on buildings. It is believed to be an effort to reduce energy consumption. Hourly energy consumption on buildings data comes from ASHRAE data warehouse, consists of three tables, these are data train, weather, and building meta. This paper aim to find best optimizer function on Back propagation algorithm with feature engineering by compared the result to predict building hourly consumption energy. In addition this paper also analyze utilization feature engineering to accuracy of prediction. The results show that Feature engineering has given 1-22 percent influence in improving accuracy of prediction than without using features engineering. Root mean square (RMSE) Adam W optimizer function on Back propagation with feature engineering is 22.61 percent, 6.12 percent when combined with Adam optimizer, and 1.67 with Stochastic Gradient Descent (SGD) optimizer. Though, Compering’s third optimizer showed that SGD optimizer has a high quality of results than both of them.