Abstract—Whole countries of the world needed electrical energy to use in our daily life that have to control about purchasing and distribution for people and organization. In Thailand, Provincial Electricity Authority is the organization about provided and managed purchasing and distribution of electrical energy to people. If the balance of purchasing and distribution of electrical energy were out of controlled, other risk factors would be consequences. In this research, the control of purchasing and distribution of Provincial Electricity Authority was studied to forecasting electrical energy for finding the best of demand and supply by using ARIMA model integrated with Extreme Learning Machine model to find the best solution of forecasting. Experiment results show that Root Mean Square Error of the proposed model compared with real data of purchasing and distribution in November 2017 were 1.9799e-05 and 3.8798e-03 respectively.
Index Terms—Forecasting, ARIMA model, extreme learning machine model.
The authors are with the Bureau of Core Business Software Phase 2, Provincial Electricity Authority, Bangkok, Thailand (e-mail: email@example.com, firstname.lastname@example.org, email@example.com, firstname.lastname@example.org).
Cite: Sarunyoo Boriratrit, Suphakorn Parakul, Nongnooch Khunsaeng, and Patsada Taecharoenchai, "Integration Extreme Learning Machine with ARIMA Model for Forecasting Electricity Purchasing and Distribution Data in Thailand," International Journal of Machine Learning and Computing vol. 8, no. 6, pp. 559-564, 2018.