A System for Monitoring Damage in Composite Materials Using Statistical Calibrations and Bayesian Model Selection

This chapter summarizes the results of a feasibility study exploring the development of a stochastic Dynamic Data-Driven Application System (DDDAS) for prediction and monitoring of material damage in composite materials common to many types of contemporary high-performance military aircraft. The methodology involves (1) acquiring data from mechanical experiments conducted on a composite material; (2) the use of continuum damage mechanics based material constitutive models; (3) developing a Bayesian framework for uncertainty quantification, calibration, validation, and selection of models; and (4) general Bayesian filtering algorithm. The Bayesian framework, enables statistical calibration of the computational models of material damage against experimental observations, along with quantifying the inherent uncertainties in the data, the model, and the numerical solution approach. Moreover, the real-time monitoring of the damage evolution with the proposed approach results in enhancement of predictive models which allows forecasting failure in the structural components given the near real time experimental measurements. K. Ravi-Chandar Department of Aerospace Engineering and Engineering Mechanics, The University of Texas at Austin, Austin TX 78712 e-mail: [email protected] D. Faghihi Institute for Computational Engineering and Sciences, The University of Texas at Austin, Austin TX 78712 e-mail: [email protected] J. T. Oden Institute for Computational Engineering and Sciences, The University of Texas at Austin, Austin TX 78712 e-mail: [email protected]