Reliability assessment based on time waveform characteristics with small sample: A practice inspired by few-shot learnings in metric space

Based on reliability analysis theory, traditional interval reliability analysis method is improved by take the smaller value of probability reliability and interval reliability as the results of reliability assessment, which makes the reliability assessment results more fit engineering cases. The improved model is applied to reliability assessment of small sample products with degradation characteristics, and results of assessment are compared with existing methods. A numerical example is considered, the results show that the assessment results by the method proposed in the paper is reasonable and believable.

Download Full-text

A Method of Small Sample Reliability Assessment Based on Bayesian Theory

Cloud Computing and Security - Lecture Notes in Computer Science ◽

10.1007/978-3-030-00006-6_11 ◽

2018 ◽

pp. 120-131

Author(s):

Hongbin Wang ◽

Nianbin Wang ◽

Lianke Zhou ◽

Zhenbei Gu ◽

Ruowen Dang

Keyword(s):

Reliability Assessment ◽

Small Sample ◽

Bayesian Theory

Download Full-text

Research on reliability assessment of success or failure product based on small sample

Civil, Architecture and Environmental Engineering ◽

10.1201/9781315226187-179 ◽

2017 ◽

pp. 986-991

Keyword(s):

Reliability Assessment ◽

Small Sample

Download Full-text

Research on Reliability Assessment for the Armored Equipment Based on Multi-Source Information Fusion

Advanced Materials Research ◽

10.4028/www.scientific.net/amr.760-762.2091 ◽

2013 ◽

Vol 760-762 ◽

pp. 2091-2094

Author(s):

Jian Du ◽

Bao Jun Fei ◽

Ying Liu ◽

Guo Zheng Yao

Keyword(s):

Field Test ◽

Test Data ◽

Information Fusion ◽

Reliability Assessment ◽

Small Sample ◽

Test Results ◽

Fuzzy Information ◽

Source Information ◽

Sample Test ◽

Fusion Theory

In order to solve the problem of reliability evaluation for armored equipment, used Bayes fusion theory, combined with minimal amounts of reliability information on the field test, the paper fusioned reliability information of the same model for armored equipment, and completed the reliability assessment for a certain type of armored equipment. The test results show that the technology of multi-source information fusion can effectively solve the fusion problem of the prior fuzzy information and small sample test data, improve the accuracy of the reliability assessment, prove the feasibility and effectiveness for the multi-source information fusion in the armored equipment assessment.

Download Full-text

Application of Bayesian Method in Assessing Reliability Parameters of Mechanical Product

Advanced Materials Research ◽

10.4028/www.scientific.net/amr.199-200.587 ◽

2011 ◽

Vol 199-200 ◽

pp. 587-590

Author(s):

Yu Lian Cui ◽

Wei Wu

Keyword(s):

Normal Distribution ◽

Prior Distribution ◽

Traditional Method ◽

Bayesian Method ◽

Small Sample Size ◽

Reliability Assessment ◽

Small Sample ◽

Time To Failure ◽

Reliability Parameters

When assessing the reliability parameters, the traditional method is to deal with a large quantity of data obtained through test and to get the result with high confidence. But small sample size and the lack of test data of mechanical items make the traditional method have weakness. If the Bayesian method can be used properly, accurate assessment will be derived even if it is lack of data. What is discussed in this paper are: 1) determination of prior distribution, 2) reliability assessment based on pass/fail data with Beta distribution, Gamma distribution and cut-tail normal distribution as prior distribution, and 3) reliability assessment based on time-to-failure with Weibull distribution and normal distribution as prior distribution.

Download Full-text