The technology of wind speed forecasting along high-speed railway, which obtains the predictions based on the data of wind stations, can reserve time for high-speed railway dispatching and ensure the safety of high-speed railway system under the strong wind. In this paper, a novel wind speed prediction model is proposed based on the deep learning, and neural architecture learning. In particular, a new mix-way structure layer is designed and used in the proposed model. Besides, the technology of adaptive structural learning of artificial neural network and long short term memory network are also used in the proposed model. The experimental tests and comparison models are used to investigate the performance of the proposed model. The results show that the proposed model produces the highest prediction accuracy among all the comparison models.

CICTP