Optimal sensor placement methodology for uncertainty reduction in the assessment of structural condition

This paper introduces a novel approach for selecting sensor positions for uncertainty reduction in the assessment of structural condition, where the main difficulty is how to quantify the uncertainty. In order to tackle this problem, a condition index, which is a linear combination of the finite-element model parameters, is defined. By taking the multiplying coefficient equal to the vulnerability index corresponding to each model parameter, the linearized condition index is able to reflect the influence of local damage on the global damage condition. The uncertainty in the estimate of this linearized condition index can be readily quantified from the uncertainty in the updated model parameters. Bayesian finite-element model updating is applied for uncertainty quantification in the model parameters. The procedure of the proposed method is illustrated by designing the optimal sensor configuration for a truss structure model. The simulated damage and condition assessment of the truss structure shows that the proposed method is effective in reducing the uncertainty in the condition assessment. Furthermore, it is demonstrated that the proposed method is useful for a more important reason: it can reduce the uncertainty in the damage assessment of vulnerable substructure. Copyright (c) 2016 John Wiley & Sons, Ltd.