The reliability assessment of aircraft ejection separation systems is crucial for aviation safety,but traditional methods exhibit significant shortcomings in dynamic behavior modeling,rare event quantification,and bid...The reliability assessment of aircraft ejection separation systems is crucial for aviation safety,but traditional methods exhibit significant shortcomings in dynamic behavior modeling,rare event quantification,and bidirectional consistency verification of function and fault paths.This paper proposes an innovative GO-FTA-GERTS dual model that integrates goal-oriented(GO)methods,fault tree analysis(FTA),and graphical evaluation and review technique(GERT)networks.The core innovation lies in constructing a bidirectional logical interlocking mechanism between the GO success tree and the FTA failure tree,leveraging the dynamic state transfer characteristics of GERT to achieve mathematical equivalence verification of function paths and fault paths.Using a specific aircraft ejection subsystem as an example,the model demonstrates excellent accuracy in ultra-low failure probability calculations:the mean system failure probability calculated by the FTA-GERT network is approximately 5.26×10^(−10) per flight hour,meeting airworthiness standards.At the same time,Monte Carlo simulation and Welch’s t-test are also used to verify that the success probability of GO-GERT model is strictly complementary to the system failure probability calculated by FTA-GERT network.This novel model effectively addresses the limitations of static analysis,dynamic path quantification,and rare event evaluation,providing a new method for reliability analysis of critical aviation safety systems.展开更多
基金supported by Shanghai Central Guidance Science and Technology Development Fund(NO:YDZX20233100004008)National Natural Science Foundation of China(NO:T2441003).
文摘The reliability assessment of aircraft ejection separation systems is crucial for aviation safety,but traditional methods exhibit significant shortcomings in dynamic behavior modeling,rare event quantification,and bidirectional consistency verification of function and fault paths.This paper proposes an innovative GO-FTA-GERTS dual model that integrates goal-oriented(GO)methods,fault tree analysis(FTA),and graphical evaluation and review technique(GERT)networks.The core innovation lies in constructing a bidirectional logical interlocking mechanism between the GO success tree and the FTA failure tree,leveraging the dynamic state transfer characteristics of GERT to achieve mathematical equivalence verification of function paths and fault paths.Using a specific aircraft ejection subsystem as an example,the model demonstrates excellent accuracy in ultra-low failure probability calculations:the mean system failure probability calculated by the FTA-GERT network is approximately 5.26×10^(−10) per flight hour,meeting airworthiness standards.At the same time,Monte Carlo simulation and Welch’s t-test are also used to verify that the success probability of GO-GERT model is strictly complementary to the system failure probability calculated by FTA-GERT network.This novel model effectively addresses the limitations of static analysis,dynamic path quantification,and rare event evaluation,providing a new method for reliability analysis of critical aviation safety systems.
