Abstract

The main objective of modeling the degradation path of a device is to predict its eventual time-to-failure. When a system is operated on-field, abrupt and unexpected changes in ambient conditions could potentially cause deviations from the expected degradation path, such as an acceleration to a state of failure. Previous estimates of lifetime distributions become inaccurate because the fitted model may no longer be a satisfactory representation of the degradation path. This paper shows the application of a Cuscore statistic to detect a downward shift in the gradient of a deterministic trend buried in autocorrelated noise. The proposed diagnostic methodology finds an application in monitoring the natural degradation of solar photovoltaic modules installed on-field.

Original languageEnglish (US)
Title of host publicationProceedings - 18th ISSAT International Conference on Reliability and Quality in Design
Pages121-125
Number of pages5
StatePublished - Dec 1 2012
Event18th ISSAT International Conference on Reliability and Quality in Design - Boston, MA, United States
Duration: Jul 26 2012Jul 28 2012

Publication series

NameProceedings - 18th ISSAT International Conference on Reliability and Quality in Design

Other

Other18th ISSAT International Conference on Reliability and Quality in Design
CountryUnited States
CityBoston, MA
Period7/26/127/28/12

    Fingerprint

Keywords

  • Degradation
  • Statistical process control

ASJC Scopus subject areas

  • Safety, Risk, Reliability and Quality

Cite this

Mancenido, M., & Pan, R. (2012). A cuscore statistic for monitoring degradation paths. In Proceedings - 18th ISSAT International Conference on Reliability and Quality in Design (pp. 121-125). (Proceedings - 18th ISSAT International Conference on Reliability and Quality in Design).