A multi-model and multi-feature fusion method for wind speed prediction along high-speed railway

Abstract

The invention discloses a multi-model and multi-feature fusion wind speed prediction method along a high-speed railway. The method comprises the following steps: 1. Installing 5 auxiliary wind measuring stations around the position of the wind measuring station; Multi-model Kalman filter method for processing; step 3: use the filtered data for wavelet processing, and construct a prediction sub-model for the low-frequency data after wavelet processing; step 4: advance the multi-step prediction model of the space-target wind station, The self-target anemometer multi-step forecast model and the weather-target anemometer multi-step forecast model get the multi-step forecast value of the target anemometer and the weather forecast wind speed forecast value of the target anemometer input into the Bayesian combination model , to obtain the final predicted value of the target wind measuring station; the invention can not only avoid the data interruption caused by the hardware failure of a single wind measuring station, but also provide longer emergency processing time for the safe driving of the high-speed railway under the severe wind environment.

Description

technical field [0001] The invention belongs to the field of railway wind speed prediction, in particular to a multi-model and multi-feature fusion method for wind speed prediction along a high-speed railway. Background technique [0002] With the sustained and stable development of my country's economy, my country's railway construction has entered a period of rapid development. With the increase of railway operating lines and the improvement of train speed, more and more attention has been paid to the safety, stability and comfort of train operation. Strong wind is one of the main natural disasters leading to train accidents. Strong winds often occur in some areas along the railways in my country, which brings great challenges to the safe and stable operation of trains. In order to prevent train accidents, it is necessary to establish a railway gale monitoring and early warning system so that the railway department can dispatch and command in advance. The wind speed predi...

AI Technical Summary

Problems solved by technology

At the same time, due to the reliability problems of hardware equipment, the prediction effect of a single model is unstable, and the wind speed prediction data along the railway line is not allowed to interrupt the output, so the wind speed prediction along the railway line must have stable performance and be able to output high precision continuously under various abnormal conditions forecast data

Images