Dynamic probability modeling-based inspection intervals optimization for civil aircraft composite structures using Bayesian updating

Abstract

The evaluation of damage tolerance in composite materials is essential for ensuring the safety of aircraft structures. One of the most challenging aspects of applying probability modeling-based methods to evaluate damage tolerance is determining the actual damage size distributions for in-service aircraft structures. Although existing nondeterministic approaches have been used to optimize inspection intervals of composite structures, few studies have investigated the effects of updates on the actual damage size distribution and its impact on both the probability of structural failure and inspection intervals. This paper proposes a dynamic optimization method for inspection intervals of composite structures based on Bayesian updating. The damage size distribution of the composite structure is characterized by a general stochastic distribution. A Bayesian updating methodology is presented to iteratively update the actual damage size distribution whenever new data becomes available. Based on the constructed probability model, the inspection intervals of composite structures are determined under the objectives of optimal safety and economy for civil aircraft using a Monte Carlo approach. Compared to prior distribution models, the proposed method achieves higher safety for structures during a single inspection, reduces the failure probability of structures throughout their entire service life, and incurs lower maintenance costs. It also enables maintenance personnel to flexibly adjust inspection intervals while facilitating quantitative evaluation of both failure probabilities and maintenance costs associated with these intervals. These findings suggest that the proposed method holds great potential in enabling maintenance personnel to make informed decisions regarding inspection intervals for improved safety and economic performance.