An official website of the United States government
The widespread integration of deep neural networks in developing data-driven surrogate models for high-fidelity simulations of complex physical systems highlights the critical necessity for robust uncertainty quantification techniques and credibility assessment methodologies, ensuring the reliable deployment of surrogate models in consequential decision-making. This study presents the Occam Plausibility Algorithm for surrogate models (OPAL-surrogate), providing a systematic framework to uncover predictive neural network-based surrogate models within the large space of potential models, including various neural network classes and choices of architecture and hyperparameters. The framework is grounded in hierarchical Bayesian inferences and employs model validation tests to evaluate the credibility and prediction reliability of the surrogate models under uncertainty. Leveraging these principles, OPAL- surrogate introduces a systematic and efficient strategy for balancing the trade-off between model complexity, accuracy, and prediction uncertainty. The effectiveness of OPAL-surrogate is demonstrated through two modeling problems, including the deformation of porous materials for building insulation and turbulent combustion flow for ablation of solid fuels within hybrid rocket motors.
more »
« less
Reconceptualizing the Prognostics Digital Twin for Smart Manufacturing with Data-Driven Evolutionary Models and Adaptive Uncertainty Quantification
This work presents an integrated architecture for a prognostic digital twin for smart manufacturing subsystems. The specific case of cutting tool wear (flank wear) in a CNC machine is considered, using benchmark data sets provided by the Prognostics and Health Management (PHM) Society. This paper emphasizes the role of robust uncertainty quantification, especially in the presence of data-driven black- and gray-box dynamic models. A surrogate dynamic model is constructed to track the evolution of flank wear using a reduced set of features extracted from multi-modal sensor time series data. The digital twin's uncertainty quantification engine integrates with this dynamic model along with a machine emulator that is tasked with generating future operating scenarios for the machine. The surrogate dynamic model and emulator are combined in a closed-loop architecture with an adaptive Monte Carlo uncertainty forecasting framework that allows prediction of quantities of interest critical to prognostics within user-prescribed bounds. Numerical results using the PHM dataset are shown illustrating how the adaptive uncertainty forecasting tools deliver a trustworthy forecast by maintaining predictive error within the prescribed tolerance.
more »
« less
- Award ID(s):
- 2317579
- PAR ID:
- 10557953
- Publisher / Repository:
- Prognostics and Health Management Society
- Date Published:
- Journal Name:
- Annual Conference of the PHM Society
- Volume:
- 15
- Issue:
- 1
- ISSN:
- 2325-0178
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Brehm, Christoph; Pandya, Shishir (Ed.)Computational fluid dynamics (CFD) and its uncertainty quantification are computationally expensive. We use Gaussian Process (GP) methods to demonstrate that machine learning can build efficient and accurate surrogate models to replace CFD simulations with significantly reduced computational cost without compromising the physical accuracy. We also demonstrate that both epistemic uncertainty (machine learning model uncertainty) and aleatory uncertainty (randomness in the inputs of CFD) can be accommodated when the machine learning model is used to reveal fluid dynamics. The demonstration is performed by applying simulation of Hagen-Poiseuille and Womersley flows that involve spatial and spatial-tempo responses, respectively. Training points are generated by using the analytical solutions with evenly discretized spatial or spatial-temporal variables. Then GP surrogate models are built using supervised machine learning regression. The error of the GP model is quantified by the estimated epistemic uncertainty. The results are compared with those from GPU-accelerated volumetric lattice Boltzmann simulations. The results indicate that surrogate models can produce accurate fluid dynamics (without CFD simulations) with quantified uncertainty when both epistemic and aleatory uncertainties exist.more » « less
-
Abstract. Plant transpiration downregulation in the presence of soil water stress is a critical mechanism for predicting global water, carbon, and energy cycles. Currently, many terrestrial biosphere models (TBMs) represent this mechanism with an empirical correction function (β) of soil moisture – a convenient approach that can produce large prediction uncertainties. To reduce this uncertainty, TBMs have increasingly incorporated physically based plant hydraulic models (PHMs). However, PHMs introduce additional parameter uncertainty and computational demands. Therefore, understanding why and when PHM and β predictions diverge would usefully inform model selection within TBMs. Here, we use a minimalist PHM to demonstrate that coupling the effects of soil water stress and atmospheric moisture demand leads to a spectrum of transpiration responses controlled by soil–plant hydraulic transport (conductance). Within this transport-limitation spectrum, β emerges as an end-member scenario of PHMs with infinite conductance, completely decoupling the effects of soil water stress and atmospheric moisture demand on transpiration. As a result, PHM and β transpiration predictions diverge most for soil–plant systems with low hydraulic conductance (transport-limited) that experience high variation in atmospheric moisture demand and have moderate soil moisture supply for plants. We test these minimalist model results by using a land surface model at an AmeriFlux site. At this transport-limited site, a PHM downregulation scheme outperforms the β scheme due to its sensitivity to variations in atmospheric moisture demand. Based on this observation, we develop a new “dynamic β” that varies with atmospheric moisture demand – an approach that overcomes existing biases within β schemes and has potential to simplify existing PHM parameterization and implementation.more » « less
-
Manufacturers have faced an increasing need for the development of predictive models that predict mechanical failures and the remaining useful life (RUL) of manufacturing systems or components. Classical model-based or physics-based prognostics often require an in-depth physical understanding of the system of interest to develop closed-form mathematical models. However, prior knowledge of system behavior is not always available, especially for complex manufacturing systems and processes. To complement model-based prognostics, data-driven methods have been increasingly applied to machinery prognostics and maintenance management, transforming legacy manufacturing systems into smart manufacturing systems with artificial intelligence. While previous research has demonstrated the effectiveness of data-driven methods, most of these prognostic methods are based on classical machine learning techniques, such as artificial neural networks (ANNs) and support vector regression (SVR). With the rapid advancement in artificial intelligence, various machine learning algorithms have been developed and widely applied in many engineering fields. The objective of this research is to introduce a random forests (RFs)-based prognostic method for tool wear prediction as well as compare the performance of RFs with feed-forward back propagation (FFBP) ANNs and SVR. Specifically, the performance of FFBP ANNs, SVR, and RFs are compared using an experimental data collected from 315 milling tests. Experimental results have shown that RFs can generate more accurate predictions than FFBP ANNs with a single hidden layer and SVR.more » « less
-
Agent-based models (ABMs) are used to simulate human-subject experiments. A comprehensive understanding of these human systems often requires executing large numbers of simulations, but these requirements are constrained by computational and other resources. In this work, we build a framework of digital twins for modeling human-subject experiments. The framework has three modules: ABMs of player behaviors built from game data; extensions of these models to represent virtual assistants (agents that are exogenously manipulated to create controlled environments for human agents); and an uncertainty quantification module composed of functional ANOVA and a Gaussian process-based emulator. The emulator is built from the extended ABM; we focus on emulator validation. By incorporating experimental data and agent-based simulation data, our proposed framework enhances the virtual representation of the dynamics in human-subject word formation experiments, which we consider a digital twin. Networked anagram experiments are used as an exemplar to demonstrate the methods.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- https://doi.org/10.36001/phmconf.2023.v15i1.3484
