Automotive Research Lab at Penn State Gives Students Practical Hardware-in-the-Loop Experience

Penn State's HIL network links fuel cells, batteries, ultracapacitors, electric drive motors, combustion engines, chassis dynamometers, and other vehicle subsystems across the campus.

"To support such a wide variety of equipment and data exchange interfaces across different labs, we needed tools that were quickly reconfigurable and flexible," Dr. Brennan explains. "Our students needed to interface with this equipment without interfering with research projects going on in the labs, and they needed to conduct offline and online simulations."

In addition, Brennan and Anstrom needed analysis and simulation tools that would enable students from a wide variety of disciplines to learn advanced vehicle powertrain analysis concepts quickly.

"C source code does not work well as a teaching aid," says Dr. Anstrom. "We wanted the students to be able to visualize systems. We needed tools that could help us quickly explain a system in the classroom and that students could use on their own to explore concepts in more detail."

The Penn State instructors use MathWorks products in all lectures and labs in Advanced Vehicle Hardware-in-the-Loop Methods. "MathWorks tools enable students to do modeling, analysis, and controller design on a single platform," says Anstrom.

Each of the first four labs in the course focuses on the dynamics of one specific powertrain component: the battery pack, ultracapacitor, DC motor, or engine. In these labs, students use powertrain system models in Argonne National Laboratory’s Powertrain System Analysis Toolkit© (PSAT). Based on Simulink, PSAT enables automotive engineers to analyze and simulate advanced powertrain technologies.

Using the PSAT model as a starting point, the students replace the software representation of a single component with real hardware. They use Simulink Coder™ to generate code from the model and Simulink Real-Time to run real-time simulations, control the hardware, and collect simulation results. In the battery lab, for example, students use Simulink Real-Time to control and gather performance measurements from a power-processing system capable of producing up to 150 kilowatts.

In each lab, students use MATLAB^® and Simulink^® to analyze system behavior, compare simulated and measured results, and determine how the results might be used for advanced vehicle powertrain optimization.

In the final lab, students work with a full electric vehicle on a chassis dynamometer. They emulate the terminal response of a fuel cell using a power-processing machine driven by a Simulink model and Simulink Real-Time.

Brennan and Anstrom used Stateflow^® to develop sophisticated control algorithms that manage startup and shutdown of the full electric vehicle to ensure safe operation. The controller was deployed on a standalone PC using Simulink Real-Time.

They are developing a scaled-down version of a hybrid electric vehicle that students can use in the classroom to create their own control strategy and real-time controller.

• Labs set up in minutes. "We began designing the hardware and implementing the course material within a few weeks of acquiring MathWorks products," says Brennan. "We were often able to set up a lab in minutes that otherwise would have taken days."
• Career opportunities enhanced. "One of our students mentioned the HIL class during an interview for an internship," says Anstrom. "The interviewer was really excited, and they talked about it for the rest of the interview. Thanks to his experience with MathWorks software and HIL, the student was offered an opportunity to work in the company’s controls group."
• Research and teaching integrated. "The MathWorks environment is extremely fast and very visual, which helps in classroom discussion," Brennan says. "It is an excellent collaborative environment for research that carries over into the classroom."