Industry Energy Consumption Prediction Using Data Mining Techniques

Abstract

Predicting energy consumption is an essential part of the electricity company supply. This paper presents and explores energy consumption prediction models using data mining approach for the steel industry. DAEWOO steel industry energy consumption data is used in this study. Data used include lagging and leading current reactive power, lagging and leading current power factor, carbon dioxide (tCO2) emission, and load types. The prediction models are trained with its best hyperparameters selected using repeated cross-validation and are evaluated using a test set: (a) General Linear Regression, (b) Classification and Regression Trees (c) Support Vector Machine with Radial Basis Kernel (d) K Nearest Neighbor, (e) Random Forest. Four evaluation indices such as Root Mean Squared Error, Mean Absolute Error, Mean Absolute Percentage Error and Coefficient of Variation are used to measure the prediction efficiency of regression models. The results show that the Random Forest model can best predict energy consumption and outperforms other conventional algorithms in comparison.