A Hybrid Optimized Model of Adaptive Neuro-Fuzzy Inference System, Recurrent Kalman Filter and Neuro-Wavelet for Wind Power Forecasting Driven by DFIG. (15th January 2022)

• Record Type: Journal Article
• Authors: Aly, Hamed H.H.
• Is Part Of: Energy. Volume 239:Part E(2022)
• Publication Date: 2022-01-15
• Subjects: Wind power forecasting -- WNN -- RKF -- ANFIS -- DFIG; Power resources -- Periodicals; Power (Mechanics) -- Periodicals; Energy consumption -- Periodicals; 333.7905
• DOI: 10.1016/j.energy.2021.122367
• Languages: English
• ISSNs: 0360-5442

Abstract

Abstract: Renewable energy resources are playing a compromising role in the new generation of sustainable energy and smart grid. Wind power is playing a crucial role these days to minimize the fossil fuel emissions. Their integration is depending on a highly accurate forecasting model due to its intermittency, nonlinearity, and fluctuation. This work presents a hybrid optimized model of Adaptive Neuro-Fuzzy Inference System (ANFIS), Recurrent Kalman Filter (RKF) and Neuro-Wavelet (WNN) for Wind Power Forecasting Driven by doubly fed induction generator (DFIG). The predictions of individual models and hybrid of ANFIS, RKF and WNN models for wind speed and power generated are compared with other published work results in the literature. Six different hybrid models are proposed (ANFIS+WNN+RKF, ANFIS+RKF+WNN, WNN+ANFIS+RKF, WNN+RKF+ANFIS, RKF+WNN+ANFIS, RKF+ANFIS+WNN). The results of this work indicate that all proposed hybrid models are performing well but the hybrid of ANFIS+RKF+WNN in sequence has the optimal performance compared to other models. Highlights: Six different hybrid models are proposed for wind power forecasting. The hybrid approaches are combinations of ANFIS, WNN and RKF. The most accurate hybrid model is the hybrid of ANFIS+RKF+WNN in sequence with nRMSE of 0.03542. The models are tested and validated by comparing them with two different models used in the literature. Model parameters are optimized based on thousands of runs and IGWO for the best performance.