Quick Links

 

ali_knaak

Yong MA

Country: China

Degree Program: Ph.D., Geophysics

E-mail: yma@mines.edu

CV

Curriculum Vitae

 

Yong received his Bachelor of Science and Master of Science degrees in acoustics in 2004 and 2007, respectively, from the Institute of Acoustics, Nanjing University (NJU), China. His thesis work at NJU focused on nonlinear ultrasound propagation in biologic tissue and applications of nonlinear ultrasound to medical imaging and therapy.

Yong joined CWP in August 2007 and has worked on multi-scale velocity heterogeneity effects on wave-equation migration with Paul Sava. He is working on image-guided full waveform inversion under his current advisor Dave Hale.

In summer 2009, 2010 and 2011, Yong interned at ConocoPhillips in Houston, Texas, to work with geophysicists in the Seismic Velocity Modeling group on a full waveform inversion project.

 

Research

Yong's thesis work focuses on image-guided sparse-model full waveform inversion (IGFWI), the purpose of which is to solve some problems in conventional full waveform inversion (FWI), such as high computational cost, local minima, and solutions with no geologic sense. With image-guided interpolation (IGI) and its adjoint operators, we can represent FWI as a sparse optimization problem, where we only invert for parameters in a sparse model space. Common nonlinear conjugate-gradient methods can be adopted to solve IGFWI. Moreover, with respect to much fewer parameters in the sparse model, we develop a quasi-Newton method, which is based on the projected Hessian matrix, to solve the sparse optimization problem.

 

Wave-equation reflection traveltime inversion

       

 

RTMimage

 

RTMimage

       

 

RTMimage

 

RTMimage

       

 

Projected Hessian and quasi-Newton FWI

       

 

RTMimage

Sparse model space for projected Hessian

 

SigsbeeVelUpdate

Eigenvector (largest eigenvalue)

       

 

RTMimage

Quasi-Newton FWI

 

SigsbeeVelUpdate

Eigenvector (smallest eigenvalue)

       

 

 

Back to Top