A method of wind speed prediction along high-speed railway for instantaneous calculation and decision-making

Abstract

The invention provides an instantaneous calculation decision-oriented wind speed prediction method along a high-speed railway. According to the recent wind speed condition, a target wind measurement station and a time shift wind measurement station are constructed at a target wind measurement point, after denoising processing is conducted on data of the wind measurement station, an extreme learning machine is used for training the denoised wind speed data, and wind speed prediction models of all the wind measurement stations under various steps are constructed; screening prediction task iteration vectors constructed by using various step sizes in advance, obtaining optimal prediction combinations of various step sizes of each model, realizing multi-step iteration prediction, improving prediction precision, and reducing interference of random errors; the method achieves the advanced prediction of the wind speed along the railway, can obtain the wind speed environment condition of an accident-prone area in advance, timely and effectively guides the operation of a train, and guarantees the operation safety of the train.

Description

technical field [0001] The invention belongs to the field of railway wind speed prediction, in particular to a wind speed prediction method along a high-speed railway oriented to instantaneous calculation and decision-making. Background technique [0002] Strong wind is one of the important reasons affecting the safety of railway transportation. It makes the running train suffer from strong cross wind. When the train runs to the winding line and the terrain is undulating, derailment and overturning accidents are very easy to occur under the action of strong wind. . On December 25, 2005, when the Japanese Shinkansen express train "Inaho No. 14" was traveling to Shonai Town, Yamagata Prefecture, northern Japan, it encountered strong winds and the train derailed, killing 4 people and injuring 33 others. At 2:05 on February 28, 2007, when the No. 5807 train from Urumqi to Aksu, a city in southern Xinjiang, was running at a distance of 42 kilometers + 300 meters between Pearl Sp...

AI Technical Summary

Problems solved by technology

The current research on wind speed prediction is time-consuming and the prediction accuracy is not ideal

Images