User:Dwhan89

Dongwon 'DW' Han

I attend University of Massachusetts at Amherst. I am majoring in mathematics with minors in physics and computer science.

Simulation of Ocean Motions using Kinematic Model

Goals and Milestones

Week1  : Familiarize myself with the project. Do background research

Week2  : Study kinematic model and previous simulations.

Week3-4 : Perform twin experiments to check the validity of the kinematic model.

Week5-7 : Implement data assimilation methods into the kinematic model

Week8-9 : Quantify error in new model

Week10  : Wrap up and prepare for a presentation

Research Journal

Week One [05/30 - 06/01]

Monday

• No work: Holiday

Tuesday

• Orientation
• Campus Tour
• Pre-Survey

Wednesday

• Library tour
• Meeting with Dr. Elaine Spiller
  • research object
  • tentative timeline
  • reading assignment
• start reading chapter 6 in Numerical Analysis by Timothy Sauer

Thursday

• Talk: Good Research Practices
• Still reading chapter 6 in Numerical Analysis
  • wrote matlab scripts for ODE solvers : Euler, Trapzoid, and RK4
• Read Kinematic Representation of the velocity field
• Start reading a note, The Rotating Can Problem by Larry Pratt.

Friday

• Start reading A Bayesian tutorial for data assimilation by Berlinear et al.
• Start reading a lecture note on Brief introduction to Probability and Simulation.
• Had a brief meeting with Dr. Spiller

Week Two [06/04 - 06/08]

Monday

• Read first few chapters of Data Assimilation: The Ensemble Kalman Filter by Evensen
• Read few chapters of Simulation by Ross
• Discussed about Data Assimilation with Dr. Spiller

Tuesday

• Read chapter 4 of Data Assimilation: The Ensemble Kalman Filter by Evensen
• Made few test runs of Kinematic Model with different initial conditions.

Wednesday

• More test runs and data analysis.
• Wrote few matlab scripts to help analyzing the data

Thursday

• More data analysis
• Modified the original code, such that it can take any initial state as an input
• Wrote more matlab scripts
• Working lunch
• Meeting with Dr. Spiller
  • presented what I learned from simulations so far
  • made a plan for next two weeks

Friday

• Started to read A Framework for Validation of Computer Models by Sacks et al.
• Started to analyze the kinemetic using the framework above
• Wrote a script for twin experiment

Week Three [06/11 - 06/15]

Monday

• Read first nine chapters of MATLAB Primer, Eighth Edition by Timothy A. Davis
• Trying to rewrite code the original Kinemetic model code in oop style.
• Note: Dr. Spiller is in conferences for next two weeks

Tuesday

• Talk: Good Presentation, Good Technical Writing and the Difference Between the Two
• Still rewriting the code

Wednesday

• finish writing main code
  • need to verify whether the outcome from new code is consistent with that of original code.
  • need to debug plotting functions

Thursday

• Working lunch
• Debugging the code [Current outcome is not consistent with that of original code]
  • Fixed plotting functions

Friday

• Still debugging the code
  • I cannot pin down the source of the error. The error must occur in Particle Filter/Bootstrap step.
  • It might be better idea to abandon new code.

Week Four [06/18 - 06/22]

Monday

• Finished debugging the code.
  • Wrote a script for twin experiment
  • Wrote more matlab functions for data analysis

Tuesday

• Wrote a script for sensitivity analysis
• Added more functions for data analysis

Wednesday

• LaTeX Introduction
• Read A Framework for Validation of Computer Models by Sacks et al.
• Organizing data
• Started to prepare for mini-presentation.

Thursday

• Working lunch
• Trying to understand A Framework for Validation of Computer Models by Sacks et al.
• Made a few images for mini-presentation

Friday

• Brainstorming about error quantification
• Read about Bayesian data analysis

Week Five [06/25 - 06/29]

Monday

• Modifying Matlab scripts
• Preparing for presentation
• Meeting with Dr. Spiller

Tuesday

• mini-presentation
• Read about Bayesian data analysis

Wednesday

• Read first four chapters of Principle of Statistics by Blumer
• Lost most of matlab codes( Luckily, I had a backup)
  • Recovering data

Thursday

• Meeting with Dr. Spiller
• More matlab coding
• Analyzed new trajectory data

Friday

• Read few chapters of Principle of Statistics by Blumer

Week Six [07/02 - 07/06]

Monday

• Studied about Bayesian Interference
• Studied about Data Assimilation

Tuesday

• Talk: What makes a good poster
• Studied about Bayesian Interference

Wednesday

• Holiday - No Work

Thursday

• Working Lunch
• Read few chapters of Principle of Statistics by Blumer

Friday

• Start writing the paper
  • Organized notes, data, and graphics
• Meeting with Dr. Spiller

Week Seven [07/09 - 07/13]

Monday

• Meeting with Dr. Spiller
  • New Task is given
  • Reading is assigned.
• Watched a lecture,Introduction To Bayesian Inference part 1, by Christopher Bishop

Tuesday

• Watched a lecture,Introduction To Bayesian Inference part 2, by Christopher Bishop
• Watched a lecture,Markov Chain Monte Carlo, by Iain Murray

Wednesday

• Worked on outline of the paper
• Organized data and notes

Thursday

• Working Lunch
• Worked on the paper
• Studied about Gaussian Procedures.

Friday

• Learned about Random Function methods
• Started to implement Random Function method.

Week Eight [07/16 - 07/20]

Monday

• Worked on implementation of Random Function method.

Tuesday

• Couldn't log into MSCS server (No computer access..)
• Worked on my paper

Wednesday

• Doesn't have an access to MSCS server again!
  • Server was working again in the afternoon.
• Worked on paper
• Implemented Randome Function method

Thursday

• Working lunch.
• Updated 4th Runge Kunta order
• Generated images for Dr. Spiller
• Spent most time debugging and writing test cases.

Friday

• Cleaned codes
• Worked on poster

Week Nine [07/23 - 07/27]

Monday

• Debugged Matlab codes
• Meeting with Dr. Spiller

Tuesday

• Prepared poster

Wednesday

• Played with Random Function
• Tried to fix the boundary problem

Thursday

• Meeting with Dr. Spiller
• Implemented boundary sampling routine
  • still has the boundary problem

Friday

• Implemented weighted averaging
  • boundary problem is fixed...
  • simulation is broken near the wall
• Sent Dr. Spiller the code

Week Ten [07/30 - 08/03]

Monday

• Still working on the boundary problem
• Organized data for paper and presentation

Tuesday

• Poster Session
• Worked on presentation

Wednesday

• Worked on presentation

Thursday

• formal presentation
• Worked on paper

Friday

• Paper is done