Chapter

Current Trends in High Performance Computing and Its Applications

pp 25-36

Dynamic Data-Driven Contaminant Simulation

  • Craig C. DouglasAffiliated withDepartment of Computer Science, University of KentuckyDepartment of Computer Science, Yale University
  • , Yalchin EfendievAffiliated withTexas A&M University
  • , Richard EwingAffiliated withTexas A&M University
  • , Victor GintingAffiliated withTexas A&M University
  • , Raytcho LazarovAffiliated withTexas A&M University
  • , Martin J. ColeAffiliated withScientific Computing and Imaging Institute, University of Utah
  • , Greg JonesAffiliated withScientific Computing and Imaging Institute, University of Utah
  • , Chris R. JohnsonAffiliated withScientific Computing and Imaging Institute, University of Utah

* Final gross prices may vary according to local VAT.

Get Access

Summary

In this paper we discuss a numerical procedure for performing dynamic data driven simulations (DDDAS). In dynamic data driven simulations our goal is to update the solution as well as input parameters involved in the simulation based on local measurements. The updates are performed in time. In the paper we discuss (1) updating the solution using multiscale interpolation technique (2) recovering as well as updating initial conditions based on least squares approach (3) updating the permeability field using Markov Chain Monte Carlo techniques. We test our method on various synthetic examples.