A comparative analysis of power demand forecasting with artificial intelligence and traditional approach. (1st January 2013)

• Record Type: Journal Article
• Authors: Zahin, Sadia; Latif, Hasan Habibul; Paul, Sanjoy Kumar; Azeem, Abdullahil
• Is Part Of: International journal of business information systems. Volume 13:Number 3(2013)
• Page Start: 359
• Page End: 380
• Publication Date: 2013-01-01
• Subjects: forecasting -- demand of power generation -- artificial intelligence -- adaptive neuro-fuzzy inference system -- ANFIS -- seasonal forecasting -- artificial neural network -- ANN -- Bangladesh; Information storage and retrieval systems -- Business -- Periodicals; Business -- Data processing -- Periodicals; Management information systems -- Periodicals; System design -- Periodicals; System analysis -- Periodicals; 651.8
• Languages: English
• ISSNs: 1746-0972

Abstract

Power demand forecasting is a significant factor in the planning and economic and secure operation of modern power system. This research work has compared different forecasting techniques and opted to find out better technique in context of power generation, which varies rapidly from time to time. The dataset has been generated from yearly demand of electricity of Bangladesh for last five years. Year, irrigation season, temperature and rainfall amount have been considered as input parameters where as single output is demand of load in adaptive neuro-fuzzy inference system (ANFIS). Another artificial intelligence technique, artificial neural network (ANN) has been used to validate the output results. The best suited traditional technique for forecasting power generation is seasonal forecasting. Seasonal forecasting is also used to compare with ANFIS and ANN to find out better technique. The result of experiment indicates that ANFIS is superior method to tackle forecasting of power generation from different error measures.