Method for converting different normal matrix test data

Abstract

The invention discloses a method for converting different normal matrix test data. The method comprises the following steps of: randomizing a function related to test data, obtaining massive simulation test data under each normal matrix by utilizing a Monte-Carlo simulation technology, and then estimating normal environmental factors used for conversion on different normal matrix test data by utilizing the simulation test data to complete conversion between small sample test data among different normal matrixes. The method has the beneficial effect that the normal environmental factor used for conversion on different normal matrix test data is obtained by utilizing the Monte-Carlo simulation technology, on one hand, enough test data can be obtained to achieve the performance parametric reliability evaluation requirement, on the other hand, the number of system tests on a product is increased to obtain enough reliability evaluation data, thus development expenditure and development time are greatly saved.

Description

Technical field [0001] The invention relates to a method for converting test data from different normal parent bodies into test data of the same normal parent body. Background technique [0002] At present, the commonly used conversion methods between different normal population test data are: [0003] Suppose the normal performance parameters of a launching device are tested under environmental conditions A {x 1 , X 2 ,..., x n }From the normal population V 1 ~N(μ 1 , Σ 1 2 ), the test data under environmental conditions B {y 1 , Y 2 ,..., y m }From the normal population V 2 ~N(μ 2 , Σ 2 2 ), and V 1 And V 2 Independent; define the normal environmental factor under environmental condition A and environmental condition B as (k, b), where Is the expansion factor, Is the translation factor; then by {x 1 , X 2 ,..., x n } And {y 1 , Y 2 ,..., y m } Two sets of experimental data can be obtained to estimate the points of k and b as: among them [0004] x ‾ = 1 n X ...

AI Technical Summary

Problems solved by technology

[0008] The technical problem to be solved by the present invention is that the test data of the performance parameters of the product under various environmental conditions is less or even a single test data, so that the estimation accuracy of the normal environmental factors used to convert the test data is poor or even impossible to estimate. Provide A method for transforming different normal parent test data based on Monte-Carlo simulation technology

Images