Digital data plays huge role in Chevy Trax safety

Every day, vast amounts of digital information is used to process and predict everything from flight patterns to calorie consumption and sleep habits. At General Motors, similar computing capabilities enable virtual crash testing to make vehicles – including the 2015 Chevrolet Trax – more road ready.
Eight decades after GM engineers developed the first barrier crash test, physical crash testing continues to provide engineers with great insight as to how vehicles perform. But it takes a lot of time.
Creating the tooling to build a physical prototype can take weeks or months. Setting up the cameras, test rigs, crash test dummies, sensors, and other hardware for a physical test also adds considerable time. Once a prototype vehicle is used in a crash test, there are limited possibilities for the vehicle to be used in another test. If engineers need to test for different variables or evaluate a revised part, a new batch of prototype vehicles would need to be built in order to continue testing.
Enter virtual crash testing. Advanced digital models and simulations allow engineers to address challenges early in the vehicle design process, and to do it over and over.
“The safety Trax offers is the direct result of exhaustive testing and analysis using both traditional physical tests and advanced computer simulation,” said Al Manzor, GM North America regional chief engineer for Small and Compact Vehicles. “The vehicles in the computer models are complete 3-D replicas of the physical vehicles, so the simulations accurately depict the way all parts of the structure and components would react in a crash.
“The understanding we gain by using these tests allows us to make more informed decisions regarding the final designs,” Manzor said.
Rendered directly from digital design files, these virtual vehicle models measure many terabytes in size and can be excruciatingly detailed down to the last nut and bolt found on a physical vehicle. These parts are further broken down into finite elements, which allow engineers to precisely model and evaluate how a physical part will behave once manufactured.
Vehicle crash models may consist of 6 million to 7 million elements, and take advanced supercomputers several hours to put through a simulated test. The result? Engineers can see how a vehicle will perform in a collision, often from angles difficult to replicate with a physical test.
“In a physical crash test, high-speed cameras allow us to see how a vehicle performs in a collision from certain angles – say, from within the passenger compartment,” said Sajid Syed, safety team lead for the Trax program. “But the virtual testing allows us to see much more detail. Not only can we look through the vehicle as if its outer skin was transparent, but we can also view how a single part behaves.”
Said Ken Bonello, senior manager of safety computer-aided engineering (CAE) integration: “A crash event may take only 100 milliseconds, but we’re able to step through millisecond by millisecond, and see the sequence of events that might be unfolding beneath the surface. We can look at the forces applied to the vehicle’s structure, determine what part might be affected next, and design the vehicle so it channels energy in a way that best protects the occupants.”
Because Trax engineers are located in South Korea and the U.S., digital tests can be performed around the globe at any time, and the international engineering team responsible for Trax’s safety performance can collectively test and refine vehicle designs to elevate occupant safety in a crash.
Virtual vehicle models are growing increasingly detailed and more realistic. Not only are the crash simulations constantly being evaluated and checked against real-world test results, but the vehicle models themselves are constantly evolving.
“Safety is a big part of our simulation efforts, but it’s also a tool used by engineers evaluating noise/vibration, durability, aerodynamics, fuel economy, and other important qualities,” said Bonello. “All these groups are using the same models; the same set of math data. It’s a very integrated and coordinated effort.”
Source: GM