KRISS UQ LAB
  • Jaebeom Lee
  • Members
  • Research Projects
  • Publications
  • Research Interest
  • Invited Talks

Main Research Topic: Uncertainty Quantification (UQ)

Picture
Picture

Why do we need to consider UNCERTAINTIES ?

Picture
Picture
Picture

Topic 1: Deterministic Input-Output Model Construction
​(Data-Driven & Physics-Based) 

Picture
Picture
Picture
Picture

Topic 2-1: Forward Uncertainty Quantification
​
(Forward UQ; Forward Uncertainty Propagation)

Picture
Picture
Picture
Picture
Picture

Topic 2-2: Inverse Uncertainty Quantification
​
(Inverse UQ; Inverse Uncertainty Propagation)

Picture
Picture
Picture

Topic 2-3: Model Uncertainty Quantification
​
(Model UQ)

Picture
Picture
Picture

#1. Finite Element (FE) Model Updating ​using Deep Reinforcement Learning Agent

Picture
  • Finite element models may have errors, due to various reasons (Sehgal and Kumar 2016):
  1. Wrong inputs of material properties (e.g., mass density, Young's modulus, etc.)
  2. Geometric simplification in modeling complex shapes
  3. Non-consideration of non-linear properties
  4. Incorrect modeling of boundary conditions and joints
  5. Poor quality mesh design
  6. Computational approximation (e.g., rounding off)
  • These factors need to be carefully corrected for diminishing the model errors. Here, this correction process is an optimization problem with multiple design variables, which may generate expensive computational cost.
  • I'm interested in solving this expensive optimization problem using an OPTIMIZATION ROBOT, well-trained using the deep reinforcement learning algorithm.

#2. Finite Element Reliability Analysis

Picture
  • It will be updated soon.

#3. Structural Condition Monitoring by Introducing Bayesian Inference

Picture
  • -​

#4. Probabilistic Finite Element Analysis

Picture
  • -​
Powered by Create your own unique website with customizable templates.
  • Jaebeom Lee
  • Members
  • Research Projects
  • Publications
  • Research Interest
  • Invited Talks