Estimation Of Extreme Values By The Average Conditional Exceedance Rate Method

In actual cases,the spacecraft structures are often subjected to multiple random load combination effect. In this paper in detail introduces the using condition of average cross ratio method (ACER) to predict extreme value process.The method is specifically designed to account for statistical dependence between the sampled data points in a precise manner. In fact,if properly used,the new method will provide statistical estimates of the exact extreme value distribution provided by the data in most cases of practical interest. It avoids the problem of having to decluster the data to ensure independence,which is a requisite component in the application.The process which predicts the extreme value: firstly, much combination load process which is based on Monte Carlo method is generated; then, the mathematical statistics on each sample should be completed; at last, predicting the extreme value.In this article, we mainly describe average conditional exceedance rate method.Considering the neighboring data of the average conditional exceedance rate method,according to principle that at least the first two function value are less than threshold as once, calculating the mean upcrossing rate based on a given rate.Then, the extreme value is predicted according to the relationship between the mean upcrossing rate and the extreme value. Lastly, the average conditional exceedance rate method and the other three distribution method on the prediction results are compared.These methods of generalized extreme value distribution analysis method respectively, Gumbel extreme value distribution analysis method and utilization based on the high threshold of an analysis of the generalized pareto distribution of the above methods.