Geometric size probability statistics characteristic analysis method used in turbine disc probability reliability analysis

Abstract

The invention relates to a geometric size probability statistics characteristic analysis method used in turbine disc probability reliability analysis. The geometric size probability statistics characteristic analysis method comprises the steps that a geometric model of a turbine disc is simplified, wherein a periodic sector is taken, and finite element calculation is performed according to periodic symmetric boundary conditions, and model simplification is performed in the calculation process; 2, parametric modeling of the model is performed, wherein geometrical fully-constrained parametric analysis is conducted on the simplified model to form a full-parametric model; 3, geometric dimensions are grouped, wherein parameters are divided into N groups during sensitivity calculation, and eachgroup includes about 10 parameters; 4, sensitivity analysis is performed, wherein an experiment design (DOE) method is adopted to conduct hypercube sampling on random geometric parameters, sensitivityanalysis is performed, and factors having the maximum counter stress influence are screened out; 5, dimension statistics and hypothesis testing are performed, wherein real turbine disc data is counted according to the screened key geometric dimensions, a single sample K-S checking method is utilized to check the distribution type of the key dimensions, and geometric size probability statistics distribution is obtained.

Description

technical field [0001] The invention is an analysis method for the probability and statistical characteristics of geometric dimensions in the probabilistic reliability analysis of an aeroengine turbine disk, which is a probability and statistical characteristic analysis method capable of considering complex structures and multi-dimensional variables, and belongs to the technical field of aerospace engines. Background technique [0002] Aeroengine is an extreme product, working under complex loads / environments such as high temperature, high pressure, and high speed; the improvement of engine performance and safety indicators requires the engine to be light in weight, long in life, and high in reliability (for example, for safe flight Engine structural parts require a low probability of failure, up to 10 -5 -10 -7 times / flight hour). In the reliability analysis process, a large number of size random variables are included. In the traditional geometric size statistics, all va...

AI Technical Summary

Problems solved by technology

[0004] The technical solution of the present invention is to overcome the shortcomings of the prior art that uniform sampling can easily lead to the collapse of the geometric model, and introduce a partition method in the sensitivity analysis of the geometric size, which can perform the sensitivity analysis and the subsequent probability and statistical feature analysis of the geometric size more accurately and efficiently

Images