Support Vector Regression Analysis for Electricity Load

Abstract

Need or consumption of electricity in various regions from time to time are always different, so that
the supply of electric power and generators are used also differ from one area to another. Predictions
from this need for electricity can help electrical energy service providers, so that the demand for
electricity and the availability of electricity is balanced. A Support Vector Machine (SVM) technique
called Support Vector Regression is applied in regression cases. Overfitting can be addressed using
the SVR approach, resulting in strong performance, results with good generalization and accuracy.
The Radial Basis Function (RBF) and the Linear Function are the kernel functions employed in this
study. The purpose of this research is model prediction and forecasting on time series data. The
results obtained in this study are the model formed from the training data has a good performance
in forecasting the test data.

Keywords: Support Vector Regression, Radial Basis Function Kernel, Linear Function Kernel,
Time Series