基于 Johnson 分布体系的随机振动环境测量数据归纳

为了解决现有随机振动环境测量数据归纳方法不能处理非正态分布数据的难题,提出了利用John-son分布体系转换非正态分布数据进行处理的数据归纳方法。此方法基于Johnson分布体系理论,对Johnson分布选择准则中百分位点z值的确定进行了讨论,并给出了优化z值的选择流程,从而获得合适的Johnson分布转换非正态分布数据,并进一步确定出非正态分布情形一定概率下的随机振动环境测量数据上限。工程实例验证了该方法的有效性和实用性。

Induction for random vibration environmental measured data using Johnson system distributions

To solve the existing traditional inductive method for measured random vibration environmental data could not apply to non-normal distribution data problem,an inductive method by transforming non-normal distribution data is presented.Based on Johnson system distributions,the method discusses how to find the sample percentile point z for Johnson system distributions,and the procedure for selecting the suitable z in Johnson curves is introduced.Then probability limits of measured random vibration environmental data with non-normal distribution is researched.Finally,the case study verifies that the proposed method is valid and practical.