Interval Type-2 Intuitionistic Fuzzy Logic System for Forecasting the Electricity Load

A study on load forecasting prediction is important for efficient management of users' demands for any utility such as electricity and gas. Many computational intelligence approaches have been adopted for short term electricity load forecasts. For instance, artificial neural network, traditional fuzzy neural networks, support vector machines and other intelligent methods have been adopted in the literature. It is well known that load forecast is an important input for management decision systems in every power supply organization. Fuzzy logic systems (type-1 and type-2) have proven to be one of the effective tools for many load forecasting problems. In this study, interval type-2 intuitionistic fuzzy logic system equipped with membership functions, non-membership functions and hesitation indices is proposed for very short and short-term load forecasting for the first time. Gradient descent method is used to optimize the parameters of the developed model. The proposed model is compared with the traditional interval type-2 fuzzy logic and traditional neural network systems. Experimental analyses reveal that the proposed model outperform the two traditional systems in many load instances.