Long Term Load Demand Forecasting in Bali Province Using Deep Learning Neural Network

PT PLN (Persero) is a company of Indonesian electricity state, one of the duty is to deliver the electricity whole of Indonesia. Every ten years, PLN conducts a planning to develop electric infrastructure that referring to a document named RUPTL (Rencana Usaha Penyediaan Tenaga Listrik), one of the parameters that estimated is load consumption in MWh until next ten years. Currently, the existing method forecasting is using Multiple Linear Regression (MLR), but it has deviation against the realization. This paper proposes the Deep Learning Neural Network method which offers better advantage in time series forecasting get smaller deviation in forecasting. This paper using a case by forecasting the load demand in Bali province based on historical data from 2004 - 2018. The evaluation is conducted by comparing the proposed method Deep Learning Neural Network against the existing method Multiple Linear Regression by obtaining the Root Mean Square Error (RMSE) and Mean Absolute Percentage Error (MAPE) value.