Hourly Load Forecasting of Electricity in Bali, Indonesia using Adaptive Neuro Fuzzy Inference System

Today electricity has been basic need for economical growth. One of measurement to identify electricity capacity in an area or a country is electricity consumption index per capita. The index in Bali, Indonesia is still lower than other developing countries in Asia. Therefore load forecasting of electricity in Bali is required to yield good electricity capacity planning. Thus this paper investigates accuracy of ANFIS implementation on forecasting electricity consumption hourly. The accuracy is characterized by MAPE, modus of MAPE, variance of MAPE, and its correlation. All the results confirm that ANFIS model is sufficient as short term electricity load predictor.