Computational & Technology Resources
an online resource for computational,
engineering & technology publications
PROCEEDINGS OF THE EIGHTH INTERNATIONAL CONFERENCE ON COMPUTATIONAL STRUCTURES TECHNOLOGY
Edited by: B.H.V. Topping, G. Montero and R. Montenegro
Damage Tolerant Design
Institute of Structural Mechanics, Bauhaus-Universität Weimar, Germany
, "Damage Tolerant Design", in B.H.V. Topping, G. Montero, R. Montenegro, (Editors), "Proceedings of the Eighth International Conference on Computational Structures Technology", Civil-Comp Press, Stirlingshire, UK, Paper 3, 2006. doi:10.4203/ccp.83.3
Keywords: damage tolerant design, risk assessment, probabilistic life cycle simulation, inspection program management.
Extending the life time of already existing engineering structures or performing risk assessments for aging structures are defining new challenges for design and maintenance management engineers. A design concept which has become popular in mechanical engineering within the last twenty years is the damage tolerant design concept, which is based on safety of damage detection during scheduled inspections and follow-up repairs. Accepting a certain maximum risk level demands the definition of an inspection and maintenance program with specific inspection intervals and inspection techniques. Fewer inspections increase the probability of failure while over-inspection will lead to an increase in life-cycle costs and reduced operation times. All governing variables describing aging structures, such as crack initiation, crack growth, damage detection and damage tolerance are of a probabilistic nature. This paper presents a simulation framework, based on Monte-Carlo techniques, which assesses the failure risk of generic engineering structures taking into account scheduled inspection and repair programs.
The deterioration of a structure is associated with both the initial manufacturing quality and the service usage. A slight variation in the manufacturing quality of new structures can lead to substantial differences in their life spans. In service, the hostile environment, operational as well as overloading or accidental loading and careless usage are responsible for degradation. Prediction of the exact limit state in terms of capacity and life is improbable. On the other side, the damage detection is subjective to the inspection program, i.e., inspection scheduling, detection technique and the inspector quality. Whatever the inspection tool may be, there is a probability of detection (POD) associated with damage type, size, location, orientation, and so on. There is a need to synthesize these seemingly deterministic probability distributions.
The damage tolerant concept taking into account stochastic scatter of input variables can be seen in a graphical representation in Figure 1. Here bands describing the stochastic scatter of data, characterize the damage evolution, the probability of detection and the probability of failure. The intersections therewith become intersection areas with complex probabilistic distributions. We can now distinguish between the minimum and the maximum damage growth life, with the first defined as the minimum time period between latest possible detection and first possible structural failure, the latter defined as the maximum time period between first possible detection of failure and latest possible structural failure.
The presented approach of a software framework for the risk-assessment of engineering structures, named CORAL [1,2] allows design and maintenance management engineers to quantitatively judge on inspection and maintenance programs for their structures. Damage initiation and evolution can be included in the framework in a very flexible manner, by external special purpose software packages or by experimental results. Costs associated with downtime and repair as well as earnings associated with operation of investigated structures can be taken into account. The proposed framework can be conveniently applied to predict the outcome of maintenance management plans in terms of probability of failure, fatal accident rate, monetary expense, downtime periods and costs, availability, reliability and so on. The ability to correlate the cost and estimate the impact of a maintenance management plan has not been attempted before. When coupled with an optimization algorithm, CORAL can help managers arrive at the most economic maintenance management plan for a desired safety level.
purchase the full-text of this paper (price £20)