Research On Wind Speed Forecast And Early Warning Along High-speed Railway

In recent years,the speed of China's high-speed rail development has been staggering.It has been hailed as the four major inventions of China in the new era.Its future development is even more exciting.Compared with traditional railways,high-speed railways are fast and more sensitive to Wind speed change.However,the existing prediction methods have low resolution and low prediction accuracy,which is difficult to meet the demand for high-speed railway wind speed prediction.Therefore,how to predict the high winds that endanger the safety of high-speed rail in time and accurately,so that the train can adjust to a safe speed in time,and ensure the safe and efficient operation of high-speed rail is of great significance.Based on this,this paper proposes a combined model prediction method and designs a set of high-speed rail wind speed forecast and early warning system based on this.The main contents of the research are as follows:Based on the idea of a combined forecasting model,a CEEMDAN-SE-GRU-BP-based ultra-short-term wind speed forecasting model along the high-speed rail is proposed.Firstly,the original wind speed sequence is processed by the complete ensemble empirical mode decomposition(CEEMDAN)with adaptive noise,and after several eigenmode functions are obtained,the sample entropy is used to measure the complexity of each eigenmode function.The eigenmode function that approximates the entropy of the sample is recomposed into three components: random component,detail component,and trend component from high to low,using gated recurrent unit(GRU)and back-propagation neural network to model and predict,and finally Add up to get the final predicted value.The data of the Jiangsu section of the Beijing-Shanghai line along the high-speed rail line in 2018 is used for wind speed prediction.MAE,RMSE,and MAPE are used as evaluation indicators.The experimental results prove that this method has a better prediction effect compared with traditional prediction methods.According to the high-speed rail operation technical specifications,the recall and precision rates in the field of information retrieval are introduced to verify the effect of this method in the high-speed rail gale warning.According to the railway standard,the real data and forecasts are based on the season,topography,and time distribution.Statistics and analysis of the data have been carried out.Compare the prediction effect and early warning effect of the forecast data in different wind speed intervals.The experimental results verify its accuracy and applicability.Finally,a high-speed rail wind speed prediction system suitable for the target high-speed railway is designed,the above algorithm is integrated into the algorithm layer module of the prediction system,and the system displays the information in a visual manner to provide users with reference.