The Mathworks helps GM shift to model-based design

  • 08-Dec-2008 09:23 EST
aefgmSaturnVue2hybridCGI.jpg
The powertrain for GM hybrids, such as the Saturn Vue 2, are designed with a heavy reliance on modeling.

Though Congress and others paint automakers as stodgy companies in need of reorganization, automakers have been transforming their business for years. General Motors, for instance, has significantly changed its design methodologies, using more model-based design in powertrain development. The technique helped the company devise a universal architectural approach for all types of vehicles while also aiding in the struggle to slash operating expenses.

Employing models makes it easier for engineers to try out large-scale designs, since much testing can be done without building physical prototypes. That is especially helpful in hybrid vehicles, since there is far more interaction between components than with conventional drivetrains.

“The big challenge with hybrids is integration,” said Kent Helfrich, Director of Powertrain Software Engineering at GM. "The engine, transmission, and electric motors all have to work together. You can’t do the engine and transmission separately."

The level of complexity and the clean slate provided by hybrid projects made these designs an ideal application for model-based design. Hybrids do not have much legacy code, so models were written from the start for each of the ­design elements.

“Our hybrid team went into production with systems that were all model-based,” Helfrich said. “That was the first powertrain design that was purely model-based.”

Helfrich explained that modeling made it possible to create one approach so that all elements can work together. That concept will make it easier to create hybrid versions of cars first produced with gasoline engines, for example.

“We have the same powertrain architecture for gasoline, diesel, and hybrids,” Helfrich said. “With models, we could abstract out an architecture beforehand.”

Creating the architecture before any system elements were physically built saved both time and money. Simulations could be run in far less time than it takes to build a prototype, so far more design concepts could be tested. The expense of multiple prototypes was eliminated, with physical versions used mainly to ensure that the simulations were correct.

“Using model-based control software lets us do all our designs virtually. We only do a few confirmation tests on prototypes instead of using prototypes to make up our minds about an idea,” Helfrich said.

Using models also made it simpler to develop hardware and software at the same time. That is increasingly important as software grows in importance.

Modeling also helps trim development time and expense. Models can be used by autocoding tools so engineers no longer have to write software by hand. Automated code generation is gaining usage rapidly as the amount of software on a vehicle increases.

“When you’re going from 1 million lines of code to 10 million, it’s clear that the hand coding process is not scalable,” said Jon Friedman, Transportation Industry Manager at The MathWorks.

GM is using MathWorks tools throughout its powertrain development organizations, cutting support costs while letting engineers share files with other groups without worrying about compatibility.

The tool provider will play a key role in GM’s future. “Our destinies are linked,” Helfrich said.

Using tools from a single supplier also makes it possible for GM to get help when those tools do not meet automakers' complex needs. “We’ve broken some of The MathWorks tools because our models were too big. They had to go back and fix things,” Helfrich said.

Share
HTML for Linking to Page
Page URL
Grade
Rate It
4.50 Avg. Rating

Read More Articles On

2016-08-19
Cadillac and Mercedes-Benz show slinky new concept cars that suggest autonomous driving shouldn't be inevitable.
2016-08-18
FiatChrysler's 2017 model year brings bawdy new Dodge throwback models, more Jeep variants and fewer Fiats.
2016-08-17
Predicting more accurately what happens in the complex chemical reactions between air and fuel when they are compressed and ignited in an IC engine’s combustion chamber—and solving the challenge of exhaust particulates—may have taken a step forward with the creation of the Computational Chemistry Consortium (C3), driven by the auto industry’s need to meet increasingly stringent global emissions regulations.
2016-11-15
Freudenberg Sealing Technologies has expanded its LESS (Low Emission Sealing Solution) lineup to include new products designed to address challenges associated with powertrain friction, smaller spaces, lighter weight vehicles and growth in the electric mobility vehicle arena.

Related Items

Article
2016-11-15