Increasing powertrain complexity and the growing number of vehicle variants are putting a strain on current calibration development processes. This is particularly challenging for vehicle drivability calibration, which is traditionally completed late in the development cycle, only after mature vehicle hardware is available. Model-based calibration enables a shift in development tasks from the real world to the virtual world, allowing for increased system robustness while reducing development costs and time.
A unique approach for drivability calibration was developed by incorporating drivability analysis software with online optimization software into a virtual engine test cell environment. Real-time, physics-based engine and vehicle simulation models were coupled with real engine controller hardware and software to execute automated drivability calibration within this environment.
This paper will present an assessment of the feasibility of carrying out automated drivability calibration within a virtual engine test environment. It will highlight the accuracy of the engine simulation model under steady-state and transient operating conditions as well as demonstrate the sensitivity of the model to calibration changes. This paper will also present the results from the automated calibration tests conducted in the virtual engine test cell.
Performing drivability analysis in a virtual test cell environment allows the development of drivability calibrations as soon as representative engine and vehicle models are available. This reduces the effort required during the vehicle calibration phase and compresses the overall timeline from concept to start of production.