<p>Many physical phenomena in science and engineering fields are too complex to be thoroughly understood. Therefore, computer modeling and simulation are being applied to get a better understanding of these physical behaviors. However, the resulting computer models of these systems inevitably lack certain first-principle aspects of the physical system behavior, leading to systematic deviations between model predictions and physical observations. One popular approach for addressing such deviations involves simultaneously calibrating the uncertain model input parameters and training an independent error model that reflects the inherent biases in model predictions. This formulation has been extensively implemented in a Bayesian context where the calibrated values or posteriors probabilities for the model input parameters were sought –&nbsp; as an aggregate representation – for the entire domain of applicability. However, an incomplete understanding of the underlying system behavior could lead to the state-dependency of the model parameters being omitted from the model. State-aware calibration permits learning such state-dependency from observational data. In this approach, the calibration parameters are presented as functions of the model input using a Gaussian process prior. This paper presents state-aware calibration as applied to multi-dimensional problems and demonstrates its application on a single mass oscillator system subject to passive and active vibration isolation and its finite element model.</p>

Multi-Dimensional, State-Aware Calibration as Applied to Models of Vibration Isolation

Paper Proceedings: https://asmedigitalcollection.asme.org/VVS/VVS2021/volume/84782

ASME 2021 Verification and Validation (VVS2021)

Supporting the need among researchers worldwide for additional online and multimedia content, ASME now offers the ASME Conference Video Collection. This platform will collect and provide long-term access to conference presentation videos, encompassing all technology areas within ASME's technical conference and events program. About 30,000-40,000 engineers across all mechanical engineering and related disciplines participate in ASME conferences each year. This includes authors of approximately 6,000 to 8,000 technical papers published by ASME, the majority of whom also submit video presentations for our virtual conferences. The video collection will be available as an add-on to existing product packages or as a separate subscription.