Research On Vibration Isolation Prediction Method Of Environmental Vibration Caused By High-speed Train Based On BP Neural Network

With the development of high-speed railways in China,the environmental vibration caused by train operation has a serious impact on the buildings along the line and the lives of residents.Therefore,it is of practical significance to study environmental vibration problems and how to reduce the harm of vibration.At present,the numerical simulation method is mostly used for the vibration isolation of environmental vibration,but a lot of simplifications are made in the numerical simulation process,and the result is often not accurate enough and takes a long time.This article proposes a new research idea,using artificial intelligence algorithms to carry out vibration isolation prediction research on the vibration of buildings caused by train operation.The main research contents are as follows:(1)The numerical model of train-track-soil-isolation ditch(empty trench/infilled trench)-building is established.The model is divided into two subsystem models: train-track model and soil-isolation trench-building model.Through the former,the orbital dynamic reaction force is solved and applied to the latter for numerical analysis.(2)By modifying the model parameters,the changes in parameters such as train speed,vibration isolation trench width and depth,vibration isolation trench distance,building distance,foundation soil properties,building height,filling materials(for the filling trench)are analyzed for the influence of building vibration,research on the control parameters that affect building vibration.(3)On the basis of numerical simulation data,with different parameters as input and building vibration response as output,BP neural network is used to establish a preliminary model for building vibration prediction caused by high-speed trains.(4)Based on the measured data,the prediction model is deep-learned and continuously optimized to improve the prediction accuracy of the model.The predicted data is compared with the measured data to verify the feasibility of the neural network model.In this paper,the intelligent algorithm is applied to the vibration isolation prediction study of the environmental vibration caused by the train.For the prediction problem of the environmental vibration isolation caused by the high-speed train,a fast and efficient analysis method is proposed,which provides a new idea for the research of engineering problems.