Estimation Of Extreme Vehicle Load Effect Based On Generalized Pareto Distribution

Estimation of extreme vehicle load effect has a significant impact on bridge safety evaluation. Due to the vehicle load in different regions have distinct forms of distribution, it is difficult to make the vehicle load effect with a certain distribution fitting. So the research on the effect of vehicle load of bridge specific area has great practical significance. In the estimation of extreme traffic load effect, this paper mainly studied the following aspects:(1) Vehicle load characteristics of the measured statistics were studied to obtain:vehicle wheelbase is difficult to fit with a specific distribution; the total weight of the vehicle axle ratio can be fitted with a normal distribution; vehicle weight can be proposed multimodal distribution; when generating random traffic, select the vehicle spacing by generating random vehicle weight.(2) The peak over threshold method(POT) based on the generalized Pareto distribution was introduced. For small and medium span bridges, the load effect sample is an un-independent sample. It is difficult to find a simple way to get accurate threshold.(3) The method of independent storms(MIS) is presented for estimating the vehicle load effect. The method of independent storms only considering the maximum independent storm, on the threshold requirement of accuracy is low. In order to obtain independent sample, divide the independent storms by the influence of load effect(4) The modified method of independent storms(MMIS) was put forward. The method of analysis data not only contains the data of the method of independent storms, also includes the larger independent storms. By calculating the location of the selected data to ensure data samples meet the independence requirements.(5) A triple class threshold method is proposed to obtain the threshold by primary data analysis. This method can be applied to the theory of POT, MIS and MMIS.(6) Finally, the POT, MIS and MMIS applied to the simply supported beam were analyzed. The results show that within the short estimation period, all the three methods can better estimate the extreme load effect, while within the middle or long estimation period, the estimation of the modified method of independent storms is higher and safer.