Modeling and forecasting the mean hourly wind speed time series using GMDH-based abductive networks

Wind power forecasting : state-of-the-art 2009.

“…The best result was the one obtained with the RBF neural network, but accuracy increases when all of the models are combined (i.e., into an ensemble of models). [114] demonstrated the application of abductive networks based on the group method of data handling (GMDH) [115] to forecast the mean hourly wind speed. The authors demonstrated that the main advantage of the abductive networks over the NN is the fast convergence during training and automatic selection of both input variables and model structure.…”

“…[109], [110], [118], [119] Smooth Transition Autoregressive [136]- [138] Discrete Hilbert Transform [120], [121] Markov-switching Autoregressive [136]- [138] Abductive Networks (GMDH) [114] Adaptive Fuzzy Logic Models [122], [123] Adaptive Linear Models [122], [123] ARIMA time series models [94]- [100], [106], [128]- [130] Neural Networks [104]- [108], [112], [131] Adaptive Neural Fuzzy Inference System [106], [116], [127] …”

Short-Term and Very Short-Term Wind Power Forecasting Using a Hybrid ICA-NN Method

“…This model has the capability of prediction for horizons between 3-6 hours according to this consideration that they merely use the past production data. In a hierarchical classification of very short-term wind power prediction approaches, two major groups are consisting: 1) those that use manufacturer's power curve or empirical curves of wind turbines to map predicted wind speed to the generation output power, and [25] Locally Recurrent Neural Networks [3] Discrete Hilbert Transform [26] Autoregressive with Exogenous input (ARX) [34] Abductive Networks (GMDH) [27] Autoregressive with Exogenous Input and Multi-timescale Parameter (ARXM) [35] Adaptive Neural Fuzzy Inference System [28] Random Forests [36] Grey Predictor [29] Neural Networks [37] 2)techniques in which wind generation power can be calculated directly without interfering the turbines' characteristics and role of NWP model is intensively highlighted for the short-term wind power forecasting.…”

A quick guide to wind power forecating : state-of-the-art 2009.

“…Kalman Filter [8], [9] Fuzzy Time Series [17], [19] Grey Predictor [10] Self-exciting Threshold Autoregressive [20]- [22] Takagi-Sugeno [11]- [14] Smooth Transition Autoregressive [20]- [22] Discrete Hilbert Transform [15], [16] Markov-switching Autoregressive [20]- [22] Abductive Networks (GMDH) [18] Adaptive Fuzzy Logic Models [23], [24] Adaptive Linear Models [23], [24] ARIMA time series models [25]- [35] Neural Networks [19], [36]- [41] Adaptive Neural Fuzzy Inference System [31], [42], [43] …”