Research On Evaluation And Analysis Of Passenger Comfort In Relation To Vibration In Railway Vehicles

Railway vehicles as the main means of transport in China, plays an important role in the national economy. Recent years, with the speedup of the train in our country, people put forward higher requirements in ride comfort. Therefore, it is necessary to research and analyze ride comfort of railway passenger train.On the basis of previous studies, this paper focused on analysis of relationship between different evaluation methods and put forward a unified evaluation method. Because, so far, there have been many different evaluation methods and there is no unified criterion in global, so user should carefully choose a ride comfort evaluation method and must examine whether it can be applied to their railway vehicles or not, and sometimes the evaluation results are different with different method. The author made the following main aspects of work.Firstly, the ADAMS/Rail simulation software has been utilized to set up the model of train vibration system. Enter the five track irregularity power spectrum respectively, generated a lot of vibration acceleration data. According to specific calculation process of each evaluation method, comfort index value is calculated with the help of MATLAB programming.Secondly, based on Genetic Algorithm and Levenberg-Marquardt Algorithm, a GA-LMBP neural network model was established. With the network model, the relationships between various ride comfort indices of railway vehicles are analyzed, resulting in reliable relations between the ride comfort indices. If raw data are not available or are missing and a ride comfort index is known, the comfort index of another method can get with the network model. The result of this study can convert one comfort index to another immediately.Lastly, on the basis of the existing evaluation methods, fuzzy AHP is analyzed thoroughly. A unified evaluation model based on FAHP is put forward. According to the calculated results of the model, divided new comfort level. Then it is applied in a example to demonstrate the feasibility and effectiveness of the model.