Somehow I've ended up on the NAFEMS mailing list for their 'BENCHmark' magazine for engineering designers and analysts. FEA is probably the most interesting area of CAD, with math that makes my head spin.
The July issue's cover story is "Simulating Snow: reducing development risk for trains." The problem is that swirling snow can plug up air intake and exhaust ducts located underneath locomotives and railcars. The topic holds double-facsinating for me: I grew up in northern Canada, where we experienced snowfall accumulations of up to 40 feet per winter; and I was heavily into model railroading in my teen years.
Snow and trains. Ahhhh!
Anyhow, the article describes how researchers Trenker, Payer, Haider, and Mann of Austria developed a numerical simulation of snow entrainment [no pun intended?] underneath railroad cars. They determined there was three ways that particles of snow moved through the air: surface creep, saltation, and suspension.
Not to deride the research, which won the NAFEMS World Congress award for "Most Innovative Use of Simulation Technology," but I had a problem with the otherwise fascinating article: There are different kinds of snow. Snow that falls has different characteristics, ranging from near ice to what we called "quarter-pounder flakes," those huge fluffy ones. Once on the ground, snow takes on different characteristics, from drifts, to ice-hard surfaces, to watery slush.
I don't know that simulation can account for the different kinds of snow.
Think of the Eskimoes, who have distinct 200 words for snow, ranging from snow that falls in the morning, snow you remember from your childhood, snow that hits the cheek of your loved one, over to snow that goes horizontally and snow with that yellow shade that indicates the huskies has been there and which thou shalt not eat.
/Z
Posted by: Zap | Aug 22, 2005 at 11:54 PM