In order to achieve quick and accurate lifetime prediction of LED lighting products under the testing time of 2 000 h, a method of online testing of luminous flux is proposed under the condition of temperature stress....In order to achieve quick and accurate lifetime prediction of LED lighting products under the testing time of 2 000 h, a method of online testing of luminous flux is proposed under the condition of temperature stress.Exponential fitting of lumen maintenance, the Bayesian estimation of failure probability, the Weibull distribution of lifetime and the Arrhenius model of the decay rate are used in combination to acquire the distribution of failure probability over time at the ambient temperatures of 25 ℃. The lifetime test of the same lamps based on the Energy Star standard under the testing time of 6 000 h is also implemented to verify the effectiveness of the method. The errors of lifetimes acquired with the proposed method are 7%, 4%, 3% and 1% at the failure probabilities of 62. 3%, 10%, 5% and 1%,respectively.展开更多
In order to verify which of the distributions and established methods of reliability model are more suitable for the analysis of the accelerated aging of LED lamp, three established methods （approximate method, analy...In order to verify which of the distributions and established methods of reliability model are more suitable for the analysis of the accelerated aging of LED lamp, three established methods （approximate method, analytical method and two-stage method） of reliability model are used to analyze the experimental data under the condition of the Weibull distribution and Lognormal distribution, in this paper. Ten LED lamps are selected for the accelerated aging experiment and the luminous fluxes are measured at an accelerated aging temperature. AIC information criterion is adopted in the evaluation of the models. The results show that the accuracies of the analytical method and the two-stage method are higher than that of the approximation method, with the widths of confidence intervals of unknown parameters of the reliability model being the smallest for the two-stage method. In a comparison between the two types of distributions, the accuracies are nearly identical.展开更多
基金The Cui Can Project of Chinese Academy of Sciences(No.KZCC-EW-102)the National High Technology Research and Development Program of China(863 Program)(No.2015AA03A101,2013AA03A116)
文摘In order to achieve quick and accurate lifetime prediction of LED lighting products under the testing time of 2 000 h, a method of online testing of luminous flux is proposed under the condition of temperature stress.Exponential fitting of lumen maintenance, the Bayesian estimation of failure probability, the Weibull distribution of lifetime and the Arrhenius model of the decay rate are used in combination to acquire the distribution of failure probability over time at the ambient temperatures of 25 ℃. The lifetime test of the same lamps based on the Energy Star standard under the testing time of 6 000 h is also implemented to verify the effectiveness of the method. The errors of lifetimes acquired with the proposed method are 7%, 4%, 3% and 1% at the failure probabilities of 62. 3%, 10%, 5% and 1%,respectively.
基金Project supported by the National High Technology Research and Development Program of China(Nos.2015AA03A101,2013AA03A116)the Cuican Project of Chinese Academy of Sciences(No.KZCC-EW-102)the Jilin Province Science and Technology Development Plan Item(No.20130206018GX)
文摘In order to verify which of the distributions and established methods of reliability model are more suitable for the analysis of the accelerated aging of LED lamp, three established methods （approximate method, analytical method and two-stage method） of reliability model are used to analyze the experimental data under the condition of the Weibull distribution and Lognormal distribution, in this paper. Ten LED lamps are selected for the accelerated aging experiment and the luminous fluxes are measured at an accelerated aging temperature. AIC information criterion is adopted in the evaluation of the models. The results show that the accuracies of the analytical method and the two-stage method are higher than that of the approximation method, with the widths of confidence intervals of unknown parameters of the reliability model being the smallest for the two-stage method. In a comparison between the two types of distributions, the accuracies are nearly identical.
