Comparison of ANN, PSO-ANN and GA-ANN models in forecasting peak daily electricity prices, Case study: Iran Electricity Market

• Main Authors: Hamed Mazandarani Zadeh, Maryam Parhizkari
• Format: Article
• Language: fas
• Published: Iranian Society of Heating, Refrigeration and Air Conditioning Engineers (IRSHRAE) 2019-09-01
• Series: انرژی‌های تجدیدپذیر و نو
• Subjects: electricity prices prediction; artificial neural networks; electricity market
• Online Access: http://www.jrenew.ir/article_87328_8d73c412a543a7a0e9241265f20d1724.pdf

Hydro-power is one of the most important ways of providing energy in peak hours. Restructuring in the electricity industry has created rivalry among the country's electricity suppliers. In order to increase the profitability of investment and better utilization of resources, estimating the future price of electricity is of particular importance to producers. Artificial Neural Networks (ANNs), as one of the most important methods of artificial intelligence, have many uses in predicting and predicting phenomena. Recently, in order to improve the performance of the model of artificial intelligence models, their combination with optimization models has become widespread. The purpose of this study was to compare the performance of ANN, PSO-ANN and GA-ANN models in predicting the dispersed and sinusoidal data of peak daily electricity prices in Iran. The results show that the use of PSO-ANN and GA-ANN models in this case study has no superiority to the ANN model and has not improved the performance and forecast of the electricity market data.