An Application of the Novel Hybrid Model in Monthly Prediction of the Electricity Cost in the Iraqi Market System

Abstract

Electricity-related industries have changed organizationally. Its monopolistic structure is competitive-like. This market's transactions are based on buyers and sellers' one- or more-day-old proposals to buy and sell electricity. The purpose of this study was to examine the methods proposed to enhance the performance of the support vector machine model and their application to the process of predicting future electricity prices. A single attempt was made to estimate the model's performance using a wavelet combination. In the electricity markets of Spain and Iraq, which are real systems with readily accessible data, the proposed method was discussed and investigated. The most significant aspect of this study is the application of and the support vector machine training model to determine the optimal inputs while accounting for uncertainty. The results showed that the price prediction ability of the proposed model is good in both Spanish and Iraqi system. MAPE, MAE, and RMSE values for the test phase are 2.22, 3.12, and 4.01, while those for Iraqi market system and the training phase are 2.76, 3.71, and 4.32.