A non-linear ARMAX for short term load forecasting

Abstract

This paper presents a methodology for forecasting the hourly load demand of the power provider enterprise using a non-linear autoregressive moving average model with exogenous variable (NARMAX). The NARMAX method can use fewer order terms and more combination terms to capture the dynamic characteristics of highly non-linear system with orthogonal parameter estimation. And the NARMAX method is expected to be improved the model identification and fewer orders in the short term load forecasting. In this paper, the proposed NARMAX method is used to test on a practice power provider enterprise (Taiwan power supply networks) for short term load forecasting to illustrate its performance.