Research And Implementation Of Track Irregularity Prediction System For Rail Detection

In the case of carrying trains,the rails will gradually produce geometric irregularities,posing a threat to the safety of the train.Therefore,the railway engineering department will first test the tracks and then formulate a track maintenance and repair plan based on the test results.With the continuous expansion of the scale of China's railway network,the existing maintenance mode has made the work department fatigued to cope with various line diseases,resulting in waste of resources and untimely maintenance.If we can give an accurate prediction of the development trend of track irregularity,it will have great practical significance for track maintenance and train safety and comfort.Therefore,this topic has studied how to improve the accuracy of track irregularity prediction.By studying the existing track irregularity prediction methods,analyzing the inadequacies of the existing work,combined with the characteristics of the track detection data,a new track irregularity prediction model is proposed.Then based on the model,combined with actual needs,The track irregularity prediction system was designed and implemented.On the one hand,based on the shortcomings of the existing work and the law of the development of the track quality,this paper proposes a method for predicting the quality of the track..In the first step,the paper uses the weight matrix to improve the accuracy of the gray model prediction,and uses the gray model to predict the overall development trend of TQI.The second step is to use the cyclic neural network to develop the TQI.The randomness of the trend is predicted;the third step is to combine the prediction results of the two parts as the final prediction value.The verification is carried out on the actual test data.The experimental results show that the method achieves better prediction accuracy.On the other hand,this paper designs and implements the track irregularity prediction system according to actual needs.The system mainly includes functions such as detection data acquisition,data display,data annotation,mileage calibration,TQI calculation and trend prediction.The system can provide complete inspection data receiving and processing procedures for the railway staff,so that the railway staff can reasonably analyze and utilize the test data,and finally achieve the purpose of ensuring railway transportation safety and saving railway maintenance and repair costs.