Power transformer demand forecast with Box Jenkins ARIMA model

Year 2020, , 95 - 100, 05.10.2020

Özlem Kuvat , Ege Adalı

https://doi.org/10.31593/ijeat.771010

Abstract

Demand forecasting is based on the principle of trying to forecast the demand for the outputs of enterprises in the field of manufacturing or service for the next periods. It requires the estimation of various future scenarios, if necessary, taking measures and taking steps, and during the application phase, the technique that is most suitable for the characteristics of the examined data set is selected and used. As a result of a healthy analysis carried out in this way, detailed plans and strict measures can be taken for the unknown, negative scenarios of the future.
This study analyzes the characteristics of a series of power transformers of a company operating in the electromechanical industry in the past years, and as a result of this analysis, the Box Jenkins Autoregressive Integrated Moving Average method (ARIMA), which best fits the results, is expected to occur for power transformers in the future. It was made to estimate the amount of demand.
Within the scope of this study, firstly, the most suitable model was tried to be determined by taking into consideration the past 132 months data of PTS. It was decided that the best choice among the alternative models was the ARMA (4,4) x (0,1) 12 model. The model was found to be stable and it was decided that the root mean square error (RMSE), mean absolute percentage error (MAPE) and Theil inequality coefficient values determined in the performance measurements were appropriate.

Keywords

ARIMA,, Box Jenkins,, Transformer demand forecast

References

• Büyükşahin, Ü. and Ertekin, Ş. 2020. A feature-based hybrid ARIMA-ANN model for univariate time series forecasting. Journal of the Faculty of Engineering & Architecture of Gazi University, 35(1), 468-478.
• Karakaş E. 2019. Forecasting Automotive Export Revenue of Turkey using ARIMA Model. Journal of Yasar University, 14/55, 318-328.
• Kaya L., Doğan, Z. and Binici, T. 2015. A comparative investigation of alternative estimation methods in non-stationary time series: analysis of cotton price. Atatürk University Journal of Graduate School of Social Sciences 19(2), 401-421.
• Contreras, J., Espinola, R., Nogales, F. J. and Conejo, A. J. 2003. ARIMA models to predict next-day electricity prices. IEEE Transactions On Power Systems, 18(3), 1014-1020.
• Ordóñez, C., Lasheras, F. S., Roca-Pardiñas, J. and de Cos Juez, F. J. 2019. A hybrid ARIMA–SVM model for the study of the remaining useful life of aircraft engines. Journal of Computational and Applied Mathematics, 346, 184-191.
• Lai, Y. and Dzombak, D. A. 2020. Use of the autoregressive ıntegrated moving average (ARIMA) model to forecast near-term regional temperature and precipitation. Weather and Forecasting, 35(3), 959-976.
• Wang, H., Huang, J., Zhou, H., Zhao, L. and Yuan, Y. 2019. An integrated variational mode decomposition and ARIMA model to forecast air temperature. Sustainability, 11(15), 4018.
• Guha, B. and Bandyopadhyay, G. (2016). Gold price forecasting using ARIMA model. Journal of Advanced Management Science, 4(2), 117-121.
• Fattah, J., Ezzine, L., Aman, Z., El Moussami, H. and Lachhab, A. (2018). Forecasting of demand using ARIMA model. International Journal of Engineering Business Management, 10, 1847979018808673.
• Dickey, D. A. and Fuller, W. A. 1979. Distribution of the estimators for autoregressive time series with a unit root. Journal of The American Statistical Association, 74(366a), 427-431.
• Phillips, P. C. and Perron, P. 1988. Testing for a unit root in time series regression. Biometrika, 75(2), 335-346.
• Kwiatkowski, D., Phillips, P. C., Schmidt, P. and Shin, Y. 1992. Testing the null hypothesis of stationarity against the alternative of a unit root. Journal of Econometrics, 54(1-3), 159-178.
• Nyoni, T. 2019. Modeling and forecasting inflation in Tanzania using ARIMA models. MPRA Paper No. 92458, posted 03 Mar 2019 19:07 UTC, 12p.
• Şahan M. and Okur Y. 2016. Estimation of the solar radiation using some meteorological data for the mediterranean region with the artificial neural network. SDU Journal of Science (E-Journal), 11(1), 61-71
• Bozkurt, H. Y. Zaman Serileri Analizi, 2. Baskı, Ekin Kitapevi, Bursa, Türkiye, 2013.