An Approach of Correlation Inter-variable Modeling with Limited Data for Inter-Bus Transformer Weather Sensitive Loading Prediction

This paper deals with a method to determine short-term loading estimation of inter-bus transformers based only a limited data of variables. A grey predicting approach is implemented in this paper to solve dynamic short term load forecasting affected by the climate changes in Indonesian EHV grid, as provided by Jawa Bali Control Center. A dynamic forecasting model is needed due to the uncertain nature of the load predicting process specifically when load changes is correlated with the external effect. Traditional GM (1,1) as basic model is presented to compare with the correlation inter-variable model in grey method for inter-bus loading transformers loading conditions refer to the temperature variation pattern during only the year 2009. GM(1,2) model denotes the correlation and relationship of hourly load demand affected to the 2 (two) tropical seasons, in this case by using local ambient temperature as dynamic variable. The daily load curve for Cawang and Cibatu distribution areas, as representing big local substations in Indonesia spatially, was used to validate the models. The model adequacy and weekly forecasting result had indicated a good and justified grade in error diagnostic checking (MAPE) in each location.