Encore abstract "Online learning of windmill time series using Long Short-term Cognitive Networks"

Abstract

The amount of data generated by windmill farms make online learning the most viable forecasting strategy. However, updating a forecasting model with a new batch of data is often very expensive when using recurrent neural network models. Long Short-term Cognitive Networks (LSTCNs) are a novel gated neural networks consisting of chained Short-term Cognitive Network blocks, each processing a temporal data chunk. The learning algorithm of these blocks is based on a very fast, deterministic learning rule that makes LSTCNs suitable for online learning tasks. The simulations using a case study involving four windmills showed that our approach reported the lowest forecasting errors and training time with respect to traditional models.