An enhanced dynamic Gaussian mixture model–based damage monitoring method of aircraft structures under environmental and operational conditions

Abstract

Gaussian mixture model–based structural health monitoring methods have been studied in recent years to improve the reliability of damage monitoring under environmental and operational conditions. However, most of these methods only use the ordinary expectation maximization algorithm to construct the Gaussian mixture model but the expectation maximization algorithm can easily lead to a local optimal solution and a singular solution, which also results in unreliable and unstable damage monitoring especially for complex structures. This article proposes an enhanced dynamic Gaussian mixture model–based damage monitoring method. First, an enhanced Gaussian mixture model constructing algorithm based on a Gaussian mixture model merge-split operation and a singularity inhibition mechanism is developed to keep the stability of the Gaussian mixture model and to obtain a unique optimal solution. Then, a probability similarity–based damage detection index is proposed to realize a normalized and general damage detection. The method combined with guided wave structural health monitoring technique is validated by the hole-edge cracks monitoring of an aluminum plate and a real aircraft wing spar. The results indicate that the method is efficient to improve the reliability and the stability of damage detection under fatigue load and varying structural boundary conditions. The method is simple and reliable regarding aviation application. It is a data-driven statistical method which is model-independent and less experience-dependent. It can be used by combining with different kinds of structural health monitoring techniques.