中国惯性技术学报
中國慣性技術學報
중국관성기술학보
JOURNAL OF CHINESE INERTIAL TECHNOLOGY
2013年
6期
828-833
,共6页
故障物理%可靠性预计%多任务剖面%惯导电路
故障物理%可靠性預計%多任務剖麵%慣導電路
고장물리%가고성예계%다임무부면%관도전로
physics of failure%reliability prediction%multiple mission profile%inertial navigation circuits
传统的基于故障率经验数据的可靠性预计方法不能满足具有高可靠长寿命特征的惯导电路的设计分析需求,而现有的基于故障物理的方法只能分析产品在寿命周期内经历单一典型任务剖面下的故障前时间。以某型飞机惯导系统内的电源电路模块为研究对象,在分析并确定其在寿命周期内可能经历的多个任务剖面和产品在每个任务剖面下的潜在故障机理及其对应的故障物理模型的基础上,通过应力分析模拟载荷施加过程,利用累积损伤和多故障时间竞争理论模拟故障发生过程,最后统计分析得到产品在寿命周期内经历多任务剖面下的平均故障前时间,获得其故障率和可靠度函数。结果表明,该新方法可获得比现有方法更接近产品实际使用情况的可靠性预计结果。
傳統的基于故障率經驗數據的可靠性預計方法不能滿足具有高可靠長壽命特徵的慣導電路的設計分析需求,而現有的基于故障物理的方法隻能分析產品在壽命週期內經歷單一典型任務剖麵下的故障前時間。以某型飛機慣導繫統內的電源電路模塊為研究對象,在分析併確定其在壽命週期內可能經歷的多箇任務剖麵和產品在每箇任務剖麵下的潛在故障機理及其對應的故障物理模型的基礎上,通過應力分析模擬載荷施加過程,利用纍積損傷和多故障時間競爭理論模擬故障髮生過程,最後統計分析得到產品在壽命週期內經歷多任務剖麵下的平均故障前時間,穫得其故障率和可靠度函數。結果錶明,該新方法可穫得比現有方法更接近產品實際使用情況的可靠性預計結果。
전통적기우고장솔경험수거적가고성예계방법불능만족구유고가고장수명특정적관도전로적설계분석수구,이현유적기우고장물리적방법지능분석산품재수명주기내경력단일전형임무부면하적고장전시간。이모형비궤관도계통내적전원전로모괴위연구대상,재분석병학정기재수명주기내가능경력적다개임무부면화산품재매개임무부면하적잠재고장궤리급기대응적고장물리모형적기출상,통과응력분석모의재하시가과정,이용루적손상화다고장시간경쟁이론모의고장발생과정,최후통계분석득도산품재수명주기내경력다임무부면하적평균고장전시간,획득기고장솔화가고도함수。결과표명,해신방법가획득비현유방법경접근산품실제사용정황적가고성예계결과。
Traditional reliability prediction methods based on empirical data of failure rate cannot meet the analysis requirements in the design of inertial navigation circuits with high reliability and long life characteristics. The existing method based on physics of failure(PoF) merely predicts time-to-failure of electronic products suffering from one typical single mission profile. This paper analyzed the multiple mission profiles in life cycle, potential failure mechanisms and corresponding PoF models under each profile of a power circuit module of some airplane’s inertial navigation system. Real procedure of load application was simulated with stress analysis. By using accumulated damage and multiple failure time competition theories, the failure occurrence process was imitated. The mean time-to-failure, failure rate and reliability of the circuit under multiple mission profiles in the whole life cycle were calculated. The results show that the reliability prediction results by the proposed method are closer to those by the previous methods under practical application condition.