TY - GEN
T1 - Analysis of degradation process with measurement errors
AU - Pan, Rong
AU - Wang, Wendai
PY - 2014
Y1 - 2014
N2 - Degradation tests are often applied on highly reliable products when the product performance can be repeatedly measured. In this paper, we compare two common types of degradation models - a nonlinear regression model and a stochastic process model. Particularly, we discuss the effects of measurement error on model parameter estimation and model selection. Using an example of photovoltaic product degradation, we (1) demonstrate the use of linear models for estimating model parameters, and (2) provide a hypothesis test for the statistical significance of product performance degradation. It shows that some forms of measurement errors, such as the drifting error of tester, can be easily incorporated into the analysis of stochastic degradation models.
AB - Degradation tests are often applied on highly reliable products when the product performance can be repeatedly measured. In this paper, we compare two common types of degradation models - a nonlinear regression model and a stochastic process model. Particularly, we discuss the effects of measurement error on model parameter estimation and model selection. Using an example of photovoltaic product degradation, we (1) demonstrate the use of linear models for estimating model parameters, and (2) provide a hypothesis test for the statistical significance of product performance degradation. It shows that some forms of measurement errors, such as the drifting error of tester, can be easily incorporated into the analysis of stochastic degradation models.
KW - Degradation modeling
KW - Hypothesis test
KW - Linear models
KW - Wiener process
UR - http://www.scopus.com/inward/record.url?scp=84900024021&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84900024021&partnerID=8YFLogxK
U2 - 10.1109/RAMS.2014.6798513
DO - 10.1109/RAMS.2014.6798513
M3 - Conference contribution
AN - SCOPUS:84900024021
SN - 9781479928477
T3 - Proceedings - Annual Reliability and Maintainability Symposium
BT - RAMS 2014 - Proceedings 2014
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 60th Annual Reliability and Maintainability Symposium, RAMS 2014
Y2 - 27 January 2014 through 30 January 2014
ER -