Reliability based optimization of a stiffened aircraft fuselage panel and the number of structural tests

Aircraft structural reliability can be improved by design and post-design safety measures that reduce uncertainty. The design uncertainty reduction measures include the use of safety factor, the allowable material properties and structural testing.Quality control, inspection, health inspection, condition monitoring, maintenance measures are the examples of uncertainty reduction measures implemented after the design process. In this study, design of a stiffened fuselage panel of a typical passenger aircraft against fracture is considered as an application problem, and simultaneous optimization of the panel and the number of structural tests, which is an example of uncertainty reduction measures, is performed. The number of coupon and element test along with company knockdown factors are used as design variables. The objective function is taken as the life cycle cost and probability of passing certification test is used as a constraint. Certification test evaluates the capability of the panel to sustain a two bay skin crack with broken central stiffener and the probability of passing this test is calculated by Monte Carlo Simulation. Uncertainty in the fracture toughness prediction, variability due to the finite number of coupon tests, and uncertainties in geometry, loading conditions and optimization for minimum cost are investigated. The effect of element test on the reliability is calculated with Bayesian updating. Response surfaces for the reliability index and structural weight are created to be used in optimization. The effect of weight penalty on the optimization results is investigated. It is found that the number of coupon tests is only marginally sensitive to the weight penalty parameter, while the number of the element test is strongly dependent on it. When the weight penalty parameter increases, each pound of structural weight becomes more valuable, the designer tend to increase the additional knockdown factors to save weight and the number of element tests increases to compensate for the increased knockdown factor. Lastly, the cost saving increases as weight penalty parameter increases, as expected.