Gas station without pumps

2014 January 18

Engineering Encounters Bridge Design Contest 2014

Filed under: Uncategorized — gasstationwithoutpumps @ 23:23
Tags: , , , , ,

Last year’s bridge design did not work well for the 2014 Engineering Encounters Bridge Design Contest (formerly the West Point Bridge Design Contest):

Bridge design costing about $169.9k in the 2013 contest.  Note: I've deliberately distorted the picture to make it difficult to blindly copy the design, as I had problems with middle-school students using my published designs to cheat on their homework.  The truss design I have here can be used as inspiration, but not copied directly.

Bridge design costing about $169.9k in the 2013 contest.

When I tried a similar design in the West Point Bridge Designer 2014, I couldn’t get the cost below about $172k, but a simpler design was cheaper:

$167.3k bridge design for West Point Bridge Designer 2014.

$167.3k bridge design for West Point Bridge Designer 2014.

This design is currently 12 of 41 in the open contest, so clearly one can do better. I don’t expect it to stay high on the leaderboard for long.  It would already be much worse than that on the consolidated board, since the top 10 on the open board only fall in the top 50 on the consolidated one.

I think that the contest would be more interesting to me if they had provided an API for testing bridges.  Then the challenge would be to write bridge optimization software that explored the design space much more thoroughly and tweaked the designs.  It might be possible to do that this year, as the source code is available from sourceforge.  I’m not interested enough in the optimization problem to try to interface to their Java code, but it might be a good way to make a college-level version of the Bridge Designer Contest.

Related articles

2013 January 16

West Point Bridge Design Contest 2013

Filed under: Uncategorized — gasstationwithoutpumps @ 21:54
Tags: , , , , ,

I modified last year’s bridge to meet the constraints of this year’s West Point Bridge Design Challenge:

Bridge design costing about $169.9k in the 2013 contest.  Note: I've deliberately distorted the picture to make it difficult to blindly copy the design, as I had problems with middle-school students using my published designs to cheat on their homework.  The truss design I have here can be used as inspiration, but not copied directly.

Bridge design costing about $169.9k in the 2013 contest. Note: I’ve deliberately distorted the picture to make it difficult to blindly copy the design, as I had problems with middle-school students using my published designs to cheat on their homework. The truss design I have here can be used as inspiration, but not copied directly.
As of 2013 Jan 16, this design is number 3 in the “open competition”, but I’m sure it will slip a long way, as I don’t plan to do much (if any) tweaking.

I found it interesting that from this year’s version of the code they removed the options to change the viewpoint from which the bridge was viewed, though the included help system still describes the now-missing controls.

2012 December 20

2012 October 6

QB3 competition

Filed under: Uncategorized — gasstationwithoutpumps @ 21:09
Tags: , , ,

The Deloitte QB3 Award for Innovation is an annual contest for research which “represents the best life science on the campuses of UC Berkeley, UC Santa Cruz, and UCSF.”  This year one of the finalists is a first-year grad student in our department,  Michelle Maalouf, who did her work as an undergrad in Nader Pourmand’s lab.  She kindly gave me this description of what she did (which I resisted editing, though the temptation is always strong when student work crosses my desk):

I utilize an injection system based on nanopipette technology, which was pioneered by Professor Pourmand to introduce defined quantities of molecules, such as RNAs, proteins and small molecules, into cells and alter their fate. The fully electrical operation control as well as the ease and low cost of fabrication are unique features that give nanopipette technology enormous potential to alter cell fate. I, as a part of Prof. Pourmand’s team, co-developed a single-cell manipulation platform based on quartz nanopipettes (~50nm) which is fitted with electrodes to mediate voltage-dependent injection into individual cells. Due to the nanostructure/size (<100nm), nanopipette tips cause less disruption to the cell membrane and allows single cell penetration multiple times without compromising cell viability. Furthermore, the use of double-barrel nanopipettes allows independent injection of two separate molecules, one from each barrel.
Nanopipettes improve current injection methods due to its high controllability and high viability of cells post injection. We have shown successful injections into mammalian cells, a technique that is a historically difficult task when using a micropipette. We will use single cell injections to reprogram human skin cells into artificial stem cells known as induced pluripotent stem cells (iPSCs). Using the nanopipette for injections to create patient-derived iPSCs offers a customized technique to address regenerative medicine by replacing damaged cells with cell therapy. iPSCs have the potential to deliver cell replacement therapy to support regenerative medicine. iPSCs can by-pass the issues of immune rejections when used for a source of tissue for transplantation because the cells can be biopsied from within the same patient (Stadtfeld et al. 2010). Regenerative therapies have the potential to treat many diseases including Alzheimer’s disease, cardiovascular disease, and Parkinson’s disease. Reprogramming of adult cells into stem cells avoids the need to use human embryos to derive stem cells and therefore negates moral and ethical issues connected with this source (Robertson 1999). Present-day methods to reprogram adult cells into iPSCs are inefficient with a success rate of 1-5% of cells within a population (Warren et al. 2010). Nanopipette technology potentially can have a great impact in stem cell research where there is a need for technologies able to inject molecules of reprogramming factors into cells with a great precision in terms of ratio and concentration.

Double-tip nanopipette.

Differentially injected cells.

If you have a University of California e-mail address, you can vote for her project (or one of the other 4 finalists, if you really think that they are better), on the public Facebook page for the contest, but voting only runs until 2012 Oct 12.

The nanopipette work in Pourmand’s lab is pretty cool, and they do a lot with nanopipettes (sensing, injecting, and extracting). For such a cheap technology, it is surprisingly powerful.

2012 September 19

USA Computing Olympiad

Filed under: home school — gasstationwithoutpumps @ 20:25
Tags: , ,

I’m thinking of encouraging my son to try the USA Computing Olympiad this year.  They have 5 web-based programming contests, each lasting 3–5 contiguous hours in a 4-day (Friday–Monday) window, followed by a proctored exam in April.  They don’t seem to have posted this year’s schedule yet, as they still have up the 2011–12 schedule—probably because the international competition is in September in Italy this year, so they are waiting to get the results from that before starting the next year.

I can’t find out much about the rules, but I believe that students can program in any language that they are proficient in.  Students all start in the “bronze” division, but students who score well are promoted to silver and gold divisions.  I suspect that my son will move quickly into the “silver” division with the correspondingly harder problems, but needs a good deal more learning to get into the “gold” division.

Next Page »

The Rubric Theme. Create a free website or blog at WordPress.com.

Follow

Get every new post delivered to your Inbox.

Join 272 other followers

%d bloggers like this: