A new method based on Type-2 fuzzy neural network for accurate wind power forecasting under uncertain data

Nowadays, due to some environmental restrictions and decrease of fossil fuel sources, renewable energy sources and specifically wind power plants have a major part of energy generation in the industrial countries. To this end, the accurate forecasting of wind power is considered as an important and influential factor for the management and planning of power systems. In this paper, a novel intelligent method is proposed to provide an accurate forecast of the medium term and long-term wind power by using the uncertain data from an online supervisory control and data acquisition (SCADA) system and the numerical weather prediction (NWP). This new method is based on the particle swarm optimization (PSO) algorithm and applied to train the Type-2 fuzzy neural network (T2FNN) which is called T2FNN-PSO. The presented method combines both of fuzzy system's expert knowledge and the neural network's learning capability for accurate forecasting of the wind power. In addition, the T2FNN-PSO can appropriately handle the uncertainties associated with the measured parameters from SCADA system, the numerical weather prediction and measuring tools. The proposed method is applied on a case study of a real wind farm. The obtained simulation results validate effectiveness and applicability of the proposed method for a practical solution to an accurate wind power forecasting in a power system control center. (C) 2017 Elsevier Ltd. All rights reserved.