Xun (Ryan) Huan is an Assistant Professor of Mechanical Engineering at the University of Michigan, and affiliated faculty to the Michigan Institute for Computational Discovery & Engineering (MICDE), the Michigan Institute for Data Science (MIDAS), and the U-M Applied Physics Program. Prior to U-M, he was a postdoctoral researcher at Sandia National Laboratories in Livermore, California. His research broadly revolves around uncertainty quantification, machine learning, and numerical optimization. He is particularly interested in optimal experimental design, Bayesian analysis, multilevel and multifidelity methods, model misspecification, and scientific computing. Outside work, Xun is passionate about aviation and holds a private pilot certificate.

CV (PDF)

News

• (2019/08) Xun Huan is giving a presentation “Uncertainty Quantification via Optimal Experimental Design and Bayesian Neural Networks for Aerospace Applications” at the National Institute of Aerospace Computational Fluid Dynamics Seminar.

• (2019/08) Xun Huan gave a tutorial lecture “Introduction to Optimal Experimental Design” at the Uncertainty Quantification Summer School. See [slides].

• (2019/07) U.S. National Congress on Computational Mechanics (USNCCM)

  • Xun Huan gave a presentation “Linear Experimental Design for Optimal Control Variate Based Surrogate Models”.
  • Krishna Garikipati gave a presentation “Variational System Identification of the Partial Differential Equations Governing Pattern-forming Physics: Inference under Varying Fidelity and Noise”.
  • Habib Najm gave a presentation “Uncertainty Quantification in Computational Models of Large Scale Physical Systems”.

• (2019/07) Xun Huan gave a presentation “Sequential Optimal Experimental Design via Reinforcement Learning” at the Workshop on Machine Learning and Uncertainty Quantification.

• News archive

Optimal Experimental Design

Experiments are often expensive, time-consuming, and perhaps even dangerous. However, not all experiments are equal, and some can produce more valuable data than others. Identifying and executing these “good” experiments can lead to substantial resource savings.

We develop mathematical frameworks and computational algorithms for optimal experimental design (OED), and seek to address questions such as:

• Under what condition should we conduct an experiment?
• What quantities should we probe?

• Where and when should we take measurements?

• Have we gathered enough data and should stop, or continue with more experiments?

OED is a form of decision-making under uncertainty. For example, when our experimental goal is to learn about model parameters from noisy measurements, a possible approach involves finding the design decision maximizing the expected information gain (mutual information).

There are many challenges associated with finding the optimal design, including the need for optimization routines to accommodate noisy responses, repeated Bayesian inferences under different scenarios, and all while wrapping around an often computationally intensive and highly nonlinear physics-based forward model with high-dimensional input and output spaces.

Example: hydrogen-oxygen ignition

• Design: initial temperature and equivalence ratio
• Measure: ignition delay times
• Infer: Arrhenius kinetic parameters

Below are expected utility contours approximated from using the full physical model (left), and a polynomial chaos surrogate model that greatly accelerate the computations (right); they show excellent agreement.

With full model	With polynomial chaos surrogate model

Three experiments are simulated (at designs A, B, and C in figure above). Bayesian inference is performed and posterior densities are shown below. Designs A (which has the highest expected utility value) produces the “tightest” posterior (left), reflecting a larger reduction in uncertainty and increased confidence for estimating these parameters.

Design A	Design B	Design C

Sequential Experimental Design

Coming soon!

Uncertainty Quantification for Machine Learning

Coming soon!

Uncertainty Quantification in Engineering Applications

Uncertainty quantification (UQ) plays a crucial role in engineering and science. It enters research and production workflows at various points, such as the highlighted yellow fields below. UQ is particularly important for the learning process, by providing a principled framework for interactions among theory, models, and data.

Example: reactive flows inside a scramjet combustor

Scramjets are propulsive systems that can enable efficient and stable operations under hypersonic flight conditions (Mach 5+). For instance, the figure below shows the flight test payload from the Hypersonic International Flight Research Experimentation (HIFiRE) program.

Designing scramjet engines requires both accurate flow simulations as well as UQ. Enabling UQ for expensive simulations of turbulent supersonic reactive flows faces many computational challenges. Techniques such as global sensitivity analysis, surrogate modeling, compressive sensing, efficient Bayesian calibration, and multilevel multifidelity sampling are needed to make UQ analysis possible. Below is an illustration of the predictive uncertainty (blue contours) of Mach number at select vertical stations of the combustor section. Such information would be valuable for finding optimal scramjet designs that are also robust and reliable.

Current and Recent Teaching

• MECHENG 305: Introduction to Finite Elements in Mechanical Engineering (Fall 2018, Fall 2019)
• MECHENG 502: Methods of Differential Equations in Mechanics (Winter 2019)

Graduate Students

Jeremiah Hauth	Wanggang Shen

• +1 (734) 647-5338
• 2033 AL (W. E. Lay Auto Lab), 1231 Beal Ave, Ann Arbor, MI 48109-2133