Estimation of material parameter uncertainties using probabilistic and interval approaches

TL;DR

This paper introduces a novel interval-based approach for estimating uncertainties in material parameters, offering an alternative to probabilistic methods by using constrained optimization to identify feasible parameter domains from measurement bounds.

Contribution

It presents an efficient line-search method to determine the feasible parameter domain considering measurement intervals, enhancing uncertainty quantification in material model calibration.

Findings

Successfully identified five fracture parameters of concrete.

Demonstrated the method's ability to quantify parameter interaction.

Provided a new approach for uncertainty estimation without Gaussian assumptions.

Abstract

Within the calibration of material models, often the numerical results of a simulation model $y$ are compared with the experimental measurements $y^*$. Usually, the differences between measurements and simulation are minimized using least squares approaches including global and local optimization techniques. In this paper, the resulting scatter or uncertainty of the identified material parameters p are investigated by assuming the measurement curves as non-deterministic. Based on classical probabilistic approaches as the Markov estimator or the Bayesian updating procedure, the scatter of the identified parameters can be estimated as a multi-variate probability density function. Both procedures require a sufficient accurate knowledge or estimate of the scatter of the measurement points, often modeled by a Gaussian covariance matrix. In this paper, we present a different idea by assuming the scatter of the measurements not as correlated random numbers but just as individual intervals with known minimum and maximum values. The corresponding possible minimum and maximum values for each input parameter can be obtained for this assumption using a constrained optimization approach. However, the identification of the whole possible parameter domain for the given measurement bounds is not straight forward. Therefore, we introduce an efficient line-search method, where not only the bounds itself but also the shape of the feasible parameter domain can be identified. This enables the quantification of the interaction and the uniqueness of the input parameters. As a numerical example, we identify the fracture parameters of plain concrete with respect to a wedge splitting test, where five unknown material parameters could be identified.