This is what i think

I’m fairly far into the coding of my thesis project… but today I figured out a much faster/efficient way to import data. Before I had 4 Component databases and 4 Configuration files… and I had a LOT of information passing between functions (that wasn’t necessary, cause the data wasn’t changing from the files). So I decided it’d be better for me to have 1 Component database and 1 Configuration file… with the configuration file being separated into 4 areas.

Now the functions import both, and only pull the information they need out of the component database. DUH. This actually lets me get rid of 2 entire functions (whose sole purpose was to import ALL the data for that configuration type and pass it into other functions). I always thought those two functions were a waste… and they were. I just wasn’t smart enough to go that one extra step.

The timeline I outlined a post or two ago is pretty intimidating to me, because this might happen a couple more times where I have to go back and rewrite entire sections. I feel like I’m not going to have anything good to show Dr. Neu this week. I need to get the resistance code comparison done.

I did populate some of the major parts of the component database today (went into lab and took all the measurements).

Thursday I’ll probably make a vidcast that has a 2 minute overview of what my thesis is about, so look forward to that.

Paper Title: Inverse Problem in Determining the Normal and Tangential Drag Coefficients of Marine Cables
Referenced by: Investigation into the hydrodynamic performance of continuous plankton recorders (Matt Wedgwood)

Paper Title: Transient Behaviour of towed marine cables in two dimensions
(M.A.Vaz and M.H.Patel) Santa Ft Laboratory for Offshore Enginering, Department of Mechanical Engineering, University College London, Torrington Place, London
(sciencedirect.com via scholar.google.com)

IEEE: The State of the ARt in Simulation of Ocean Cable Systems (Meggitt, D. Ottsen, H.)

Dynamics of Cables, Towing Cables and Mooring Systems (M.S. Triantafyllou)
The Shock and Vibration Digest, Vol. 23, No. 7, 3-8 (1991)

Engineering Aspects of Manned and Remotely Controlled Vehicles [and Discussion] (R.F.Busby, C.Kuo, M.W.Thring) Philosophical Transactions of the Royal Society of London. Series A, Mathematical and Physical Sciences, Vol. 290, No. 1366, Sea Floor

Design and Simulation of a Towed Underwater Vehicle (Amy Linklater)

Wikipedia (http://en.wikipedia.org/wiki/Catenary)

High Speed Autonomous Underwater Vehicle Computational Fluid Dynamics Stuff

I’m using a program called gambit. Since I’ve never used it before, I’m going to take some notes on it.

The AUV is 4.75 inches wide = 0.12065 meters (Gambit uses metric units)

Radius = 0.060325 m

(gonna start with a circle, then move to a sphere)

I’m also going to use a symmetry line so only the upper half of the circle

I’m placing 1 diameter in front, 3 behind, and 2 above.

First I created the vertices for the test area, then the beginning, center and end of the circle, then the upper part. Then I created the ‘Edges’. 4 for the test area, then the arc for the circle. I’m going to go back and split up the bottom line. To select points hold down shift.

Scratch everything above, I’m going to do something similar to this first: http://www.mne.psu.edu/cimbala/Learning/Fluent/gambit_cylinder.htm

That is a great tutorial. When rendering the mesh in GAMBIT I get a warning that the Mesh size is smaller than the geometric tolerance (as noted above, my radius is quite small). I’m using a node count of 50 on each quarter of the circle, so this probably need to be decreased to something else. I ran a thing in gambit and it quite on iteration 120/200 because of a floating point error which i think is related to the same problem. Originally I was in standard 2D fluent and it wouldn’t get past the first iteration, but I switched to double precision and that’s where it quit at 120. The tutorial had the outer circle at 100 times the diameter of the inner circle. I originally used 10x instead, so i’m switching back to 100 to hopefully allow for larger geometry. (Not working, got the same error, and now gambit is taking longer to mesh.

I’m doing a few test runs. Turns out our senior design project design is optimum. The largest FEASIBLE design I’ve got has an OMOE of 0.731. Now I’m running an experiment…

Continue reading ‘GA Partial Results…’

Over the past few days I’ve been working on my own genetic optimization program. For our senior design project me and 5 other OEs were tasked to design a submarine for a competition. We’re using a Multiple-Objective Genetic Optimization (MOGO) method Dr. Brown has written two papers on it:

• http://www.aoe.vt.edu/~brown/Papers/ASNE2002Paper.pdf
• http://www.aoe.vt.edu/~brown/Papers/ASNE2003RiskPaperRevised.pdf

I need to check his references at the end of his papers.

What I’ve done…

Continue reading ‘Genetic Optimization…’