LED照明灯具的色漂退化机理及其可靠性建模研究

色漂是LED照明灯具的关键失效模式之一，已成为系统评估LED灯具可靠性的研究重点。本项目针对LED照明产业的重大需求，开展LED灯具色漂退化机理及其可靠性建模研究。首先采用高加速极限应力退化试验，进行灯具的色漂极限应力研究，获得可保证色漂机理一致的加速试验极限应力；结合子系统隔离加速的步进应力加速衰减试验模型，试验研究引起灯具色漂的关键因素及其相互作用机制，建立灯具色漂的步进应力加速试验理论分析方法。然后分析灯具系统色漂轨迹的漂移规律，建立基于色漂退化机理的色漂轨迹统计分析模型，建立灯具系统色漂的可靠性评估模型，分析灯具色漂失效的寿命特征；进而深入研究灯具色漂关键因素的色漂诱导机理及其主要表征参数，阐明关键因素诱导灯具色漂的作用规律，揭示关键色漂机理与环境应力的关联机制，建立基于关键色漂机理的LED灯具可靠性预测模型。研究成果将对LED产品设计、制造及可靠性评估具有重要科学意义和应用价值。

Color shift (CS), as one of the critical failure modes of LED luminaires, has become a research focus on systematically evaluating the reliability of LED luminaires. This project aims at the major demand of LED lighting industry, and carries out the research on the CS degradation mechanisms and their reliability modeling for LED luminaires. First, adopting highly accelerated stress-limit decay testing, the SC stress limit of LED luminaires is investigated in order to obtain an accelerated testing stress limit that ensured with the consistency of CS mechanisms; the key factors that cause CS of LED luminaires and their interaction mechanisms are studied experimentally based on the step-stress accelerated decay testing models with a subsystem isolated acceleration pattern; and the theoretical analysis method of the step-stress accelerated decay testing for the color shift process of LED luminaire is established. Second, statistics analysis models for CS paths are established by analyzing experimental CS laws of lamp system; the statistics models are associated with the CS decay mechanisms under different testing stresses; consequently, a reliability evaluation model for lamp system CS procedure is established, and life-span characteristics of LED lamp CS failure are analyzed. Furthermore, the CS induced mechanisms and accordingly main characteristic parameters for the key CS factors are explored deeply; the effect rules of lamp CS induced by key CS factors are then elucidated, and the correlation mechanisms between characteristic parameters of key CS mechanisms and environmental stresses are illustrated; consequently, a Reliability prediction model focused on the key CS mechanisms of LED luminaires is developed. The project study results are of important scientific meaning and application value on LED product design, manufacturing and reliability evaluation.