Comparison of statistical models for the lumen lifetime distribution of high power white LEDs

Jiajie Fan, Kam Chuen Yung, Michael Pecht

Research output: Chapter in book / Conference proceedingConference article published in proceeding or bookAcademic researchpeer-review

7 Citations (Scopus)

Abstract

Compared to conventional light sources, high power white LED (HPWLED) possesses superior benefits in terms of efficiency, power consumption, environmental friendliness, and lifetime. Therefore, the market of HPWLED is growing rapidly in the application of general lighting, LCD-TVs backlighting, motor vehicle lighting. However, traditional reliability assessment techniques have several limitations on this highly reliable electronic device with little failure during life test. This paper uses the general degradation path model to analyze the lumen maintenance data of HPWLED with two approaches (Approximation approach and Analytical approach). And three statistical models (Weibull, Lognormal, and Normal) were utilized to predict the lumen lifetime of HPWLED and finally the prediction results were verified by the Akaike Information Criterion (AIC). Results show that Weibull model is the best-fitting one to the "pseudo failure time" data in the approximate approach, however, Lognormal is the most suitable fitting model for the random effect parameter, β, in analytical approach.
Original languageEnglish
Title of host publicationProceedings of IEEE 2012 Prognostics and System Health Management Conference, PHM-2012
DOIs
Publication statusPublished - 18 Sep 2012
Event2012 3rd Annual IEEE Prognostics and System Health Management Conference, PHM-2012 - Beijing, China
Duration: 23 May 201225 May 2012

Conference

Conference2012 3rd Annual IEEE Prognostics and System Health Management Conference, PHM-2012
CountryChina
CityBeijing
Period23/05/1225/05/12

Keywords

  • Akaike Information Criterion (AIC)
  • High Power White LEDs
  • Lognormal Distribution
  • Lumen Lifetime Distribution
  • Normal Distribution
  • Statistical Models
  • Weibull Distribution

ASJC Scopus subject areas

  • Computer Science Applications
  • Software
  • Biomedical Engineering
  • Health Information Management

Cite this