Inference And Calibration Of Rail Vehicles Load Spectrum Based On Mathematical Statistics Methods

State of loads which load on mechanical structures has a direct impact on the structural life. The load spectrum of structures is the basis for the structural reliability test evaluation standards and design specifications.Combine Weibull distribution and lognormal distribution as a new distribution to describe the distribution of measured rail vehicle load spectrum distribution, on the basis of reaching track vehicle line measured load spectrum distribution and its laws. This paper verified its effectiveness with mathematical distribution test methods. Calculate and compare maximum, spread spectrum, single-stage and spectrum damage of different distributions, to show combined distribution advantages on load spectrum estimation.Distinguish mass line measured load data with the running cross-road, set the single cross road daily test data as the basic unit, and get a group of independent data. Introduce the concept of equivalent life, on the basis of researching of aircraft the median load spectrum method. Calculate the minimum number of data unit with T-distribution test criterion on the basis of the given confidence level, and then select the most representative data in the range to establish load spectrum with high degree of confidence.Damage of the measured load and the stress should be consistent, validated and improved the existing damage consistent calculation method, proposed a new load calibration method with the principle of damage coherence.This paper developed a measured data processing system which includes corresponding time-history data corresponding part and load spectrum optimization part with Delphi and Matlab. Realize the optimizing of load spectrum quickly and easily.