Research On Dynamic Stress Monitoring And Load Identification Technology For Key Structures Of High-speed Trains

The key structure of high-speed trains,as represented by the crossbeam,plays an important role in carrying passenger/cargo weight,transmitting dynamics,and ensuring passenger safety.During service,the key structure bears the vehicle’s own weight and is subjected to a variety of alternating loads brought by train operation.It may easily result in stress concentration,overload,and fatigue and cause structural damage.Furthermore,the accumulation of structural damage may lead to structural failure,which poses a huge challenge to the long-term safe operation of high-speed trains.Dynamic stress and load monitoring is the basis for structural strength assessment and fatigue life prediction,which is important to promote the maintenance of key structures and to ensure the operation safety of high-speed trains.Therefore,research on dynamic stress monitoring and load identification technology for critical structures of high-speed trains is carried out.This thesis takes the crossbeam of high-speed trains as an example,and the main research contents include:(1)Research on load identification theory of high-speed train crossbeams.Firstly,load conditions of the crossbeam of high-speed trains in service are analyzed.Then,the equations of motion of the linear elastic multi-degree of freedom system are decoupled according to the structural dynamics and modal theory,and the mapping relationship between load and strain responses when the structure is loaded is studied.And a load identification method based on load-strain linear superposition is proposed.Finally,the ill-posedness of load identification is analyzed based on the relationship between load and strain responses,and the solutions to ill-posedness based on mathematical models and neural networks are discussed.(2)Construction of a dynamic stress monitoring system for the crossbeam of high-speed trains.Firstly,a three-dimensional structure model of the crossbeam is built based on the actual model of the crossbeam of high-speed trains by SolidWorks and Abaqus software,and the structure is finely meshed by HyperMesh.Then,crossbeam mechanical simulations are carried out according to its practical service condition,the load area of the structure is discretized,and the strain distribution pattern of the structure under different loads is analyzed.Finally,a fiber grating dynamic stress monitoring system based on FBG(Fiber Bragg Grating)sensors is designed to acquire,display,and store the structure’s dynamic stress/strain signals according to the crossbeam and its service conditions.(3)Research on crossbeam load identification method based on load-strain linear superposition.Firstly,the load identification method based on load-strain linear superposition of the crossbeam is studied,and the load-strain matrix acquisition method and load identification process are introduced.Then,the mechanical simulations of two-load and multi-load conditions under different amplitudes are carried out based on the finite element model of the crossbeam.Finally,an experimental platform is built based on the fiber grating dynamic stress monitoring system,and the loading conditions and experimental procedures are designed.The experimental results show that the load identification method based on load-strain linear superposition provides a good identification of the loads on the crossbeam.(4)Research on crossbeam load identification method based on neural network.Firstly,the mapping relationship between structural load and strain responses under the framework of the IELM(Improved Extreme Learning Machine)is studied based on the load-strain linear superposition load identification method,and a load identification method based on IELM is proposed.Furthermore,the proposed method is analyzed and compared with conventional methods for load identification at different noise levels in combination with finite element simulation techniques.The results show that the IELM-based load identification method has higher accuracy and stronger noise immunity.Finally,an experimental platform is built and an experimental procedure is designed.The experimental results show that the IELM-based load identification method achieves good identification of loads at any position on the crossbeam.This thesis designs a dynamic stress monitoring system based on fiber grating sensing for monitoring the mechanical status of critical structures of high-speed trains,and proposes a load identification method based on load-strain linear superposition and neural network.And both simulation and experimental results show the proposed method has good accuracy and noise immunity,which successfully realizes dynamic stress monitoring and load identification of critical structures of high-speed trains,and is of great engineering application value for supporting the intelligent operation and maintenance of high-speed trains.