General Motors leverages Simcenter simulation to optimize performance and reduce development.
GM was founded in 1908 by William C. Durant and has been at the forefront of the automotive revolution for more than 100 years.
General Motors has been working on a vision of a world with zero crashes, zero emissions and zero congestion through the creation of electric, self-driving, connected vehicles and shared mobility services that will transform how we get around. Its diverse team of 155,000 employees brings a collective passion for engineering, technology and design to deliver on this ambitious vision.
In the Propulsion System Global Noise and Vibration (N&V) department, Michael J. Grimmer, GM technical fellow and his team are working on all products related to internal combustion engines, hybrid drive units, transmissions and electric drive units.
Like most automotive manufacturers, the critical business issues for GM today are reducing costly physical testing with virtual simulation twins and therefore accelerating the product development timeline. In addition, the transition of the transportation industry from internal combustion propulsion to electric propulsion is also an important consideration.
Improving driveline quality with simulation
Modeling the propulsion system and understanding the impact of torque fluctuations on the drive quality and seat acceleration experienced by the driver and passengers is a challenge that GM and many other original equipment manufacturers (OEMs) face today. The propulsion system torque must operate smoothly when the driver or the operator of the vehicle accelerates and decelerates the vehicle.
To address this challenge, GM uses SimcenterTM software, which is part of the Xcelerator portfolio, the comprehensive and integrated portfolio of software and services from Siemens Digital Industries Software, to perform extensive simulation on vehicle disturbances that come from the driveline and to investigate techniques to mitigate the disturbances going into the cabin.
With Simcenter solutions, Grimmer and his team have shortened the product development cycle by significantly reducing the time spent on development and physical testing of the prototypes. Building a comprehensive digital twins of the vehicle enables the team to perform many what-if analyses.
For example, to alleviate driveline noise and vibration, GM engineered several types of vibration isolators virtually using model-based systems engineering (MBSE) techniques and performed system-level analysis to understand the performance and com- patibility of the complete system before developing a physical prototype. “We typically had to design and fabricate the prototype of driveline components such as isolators to perform each test and realize their impact on the upstream and downstream driveline frequencies,” says Grimmer. “This is a process that can take many months and can cost up to $50,000. Using simulation in Simcenter Amesim, it now takes only days.”
Simulation can also be used in the concept selection phase. When a new product is designed, the engi- neering team proposes several concepts. Engineers collaborate and select the best concepts based on the calculated expected performance. “In this case, we can do concept selection based on simulation of how the products will perform,” Grimmer says. “This greatly expedites the number of options we can consider and the correctness of the ones that we choose.”
Before simulation technology existed, companies used engineering experience, best practices, physical testing and analysis results from previous similar products, but this approach had little fidelity to actual performance. “The wonderful thing about simula- tion software, especially good simulation software like Simcenter Amesim, is the ability to innovate and to demonstrate your innovation skills,” Grimmer explains. “It allows you to put in place improved engineering processes. I can say that I’ve had good personal experiences using Simcenter Amesim.”
The Simcenter Amesim models that we build are easy to operate by others and can be integrated into the rest of the system and with the controls. Another user from a different engineering team can receive the subsystem model from us or even from a supplier as a black box and integrate it in their workflows quite effectively.
Michael J. Grimmer, Technical Fellow, Propulsion System Global N&V General Motors
Fully integrating system simulation into the engineering process
The integration of Simcenter solutions and methodologies into General Motors’ development processes has enabled new standardized methods that better balance various performance areas.
Specifically, in the realm of driveline noise and vibration, GM has established new simulation-based standard work practices. The critical part of this work is balancing the noise and vibration requirements with energy requirements for the vehicle. There is a natural tradeoff between noise and vibration performance and, for example, fuel economy. The ability to simulate driveline noise and vibration enables the company to establish best practices and to determine the optimum (or acceptable) level of noise and vibration within the energy requirements for the vehicle that fulfill fuel economy requirements.
“It’s not simply quicker and less expensive, it’s simply more available,” Grimmer says. “As a consequence, we could say we have products that achieve a much better balance than our previous products. Because in the past, without the ability to do this with simulation, that simply would not be done. The products would come out with less optimization, less balancing of these different performance areas.”
In addition, the Simcenter portfolio helps different functional groups collaborate more effectively and makes the engineering process more efficient. “The Simcenter Amesim models that we build are easy to operate by others and can be integrated into the rest of the system and with the controls,” Grimmer says. “Another user from a different engineering team can receive the subsystem model from us or even from a supplier as a black box and integrate it in their workflows quite effectively. For example, we can have one engineer building a model of the propulsion system and passing that model on to another engineer who has a multibody dynamics model of the entire vehicle and they’ll take the propulsion model and run their vehicle model with it. The nature of the Simcenter Amesim solution makes model integration and collaboration possible.”
Meeting industry challenges with evolving simulation solutions
Grimmer’s team strives to keep up with the latest tools for faster and cheaper development and to remain focused on their main concern: energy consumption and vehicle refinement. To do so, they often rely on Simcenter Engineering and Consulting services.
Grimmer and his team periodically uncover various features that they would like to see added to Simcenter AmesimTM and are helped by the product development team to develop special features or improve existing solutions. “Modeling torsional isolators within the transmission involve arc springs that have unique physics and Simcenter Amesim has an out-of-the-box submodel for modeling those physics, which was improved to achieve much quicker and accurate simulation,” Grimmer says. “Another example is what’s called a viscoelastic spring and its interface was improved for a more intuitive version based on our needs. That was extraordinary product support. We’re very pleased with the product support we get with Simcenter Amesim.”
The power of simulation in the engineering process
The power of simulation in engineering first relies on combining various simulation solutions. Grimmer’s team occasionally exports simulation data from Simcenter Amesim into SimcenterTM TestlabTM software to do signal processing and compare simulation results directly with the test data. GM also plans to use functional mockup units (FMUs) to integrate multiple subsystems and couple test data and simulation. One of the challenges to doing this work virtually, without physical prototype testing, is integrating several subsystem simulation models with the controls.
On the other hand, this enables validation of controls, testing of new designs concepts and performing virtual calibration, thus providing virtual models that are available throughout the development cycle of their vehicle systems. The team greatly values the ability to perform simulations with software-in-the- loop (SiL) and to conduct co-simulations with other simulation tools.
For Grimmer, there is a clear paradigm shift towards exclusively using simulation where possible. GM believes that simulation accelerates product performance prediction, allowing teams to make well-informed decisions in earlier development stages. It is now possible to model and simulate a much broader range of products and not only product designs, but also production, manufacturing variation and what-if analyses. Engineering teams can use data from actual duty cycles to realize the performance and quickly run design of experiments (DOE) to evaluate how parametric variations influence the performance of the entire product.