We’re now at an inflection point in which symbolic technology will automate physical modeling and equation generation through tools like MapleSim. As a recent webinar hosted by Maplesoft and the Society of Automotive Engineers proved, engineers are fascinated by the application of the technology, and the technology itself.
We’ve certainly come a long way since I first applied Maple to an engineering problem.
While studying Chemical Engineering at The University of Nottingham (located in a fine East Midlands city in the UK), I and my fellow students were asked to solve the differential equations that described the heat transfer through a series of stirred tanks. Having just taken a numerical methods course, my fellow students brute-forced the solution using a Runge-Kutta method coded in FORTRAN. I, however, had several friends studying mathematics, who upon discovering my task, introduced me to Maple.
With my new-found knowledge, I solved the DEs analytically and impressed the lecturer so much that I was given a big fat zero for the assignment (he really wanted a numerical solution!)
That was the application of symbolic technology to an engineering problem in 1995; I used Maple to solve DEs that had already been derived by hand. MapleSim, however, addresses the first part of the problem by automating physical modeling and equation generation.
As I remarked earlier, we recently hosted a webinar delivered by three industry experts (Dr. Laurent Bernardin from Maplesoft, Dr. Ken Butts from Toyota Technical Centre and Dr. Chad Schmitke from Maplesoft) under the auspices of the Society of Automotive Engineers. The webinar centered on physical modeling through MapleSim and HLMT (High Level Modeling Tool – software that we developed specifically for Toyota).
The star attraction was Dr. Butts’ presentation that described how he and his team modeled a VVT Actuator with MapleSim and HLMT. Given the novel nature of the webinar, we were inundated with so many questions that we had to answer the bulk of them offline.
One delegate asked how Maplesoft’s symbolic technology benefited the physical modeling process. Dr. Bernardin replied that symbolic technology was used to preprocess models, automatically generate symbolic equations, and make large equations systems more numerically efficient to solve (by, for example, reducing the index of high-index differential algebraic equations).
Another delegate wanted to discover how MapleSim knows which variables to eliminate during the symbolic simplification step for multibody systems. Dr Schmitke (the senior architect of the multibody library) replied that MapleSim has heuristics for determining the optimum variables to eliminate, but users could override this process if desired.
If you missed out on the webinar, you can watch a recording at http://www.maplesoft.com/view.aspx?SL=6882