Mechatronic powertrain virtual sensor: system-level model and sensor selection

Virtual sensors provide measurements for variables and parameters which are difficult to measure, based on a small number of (preferably pragmatic) measurements in combination with a numerical model of the observed system. This work presents a global virtual sensor to extract the full state of a mechatronic drivetrain based on a system-level model and a Kalman filter. Sensor selection is performed based on an observability analysis and the Kalman filter settings are based on a Monte-Carlo simulation. The virtual sensor is validated experimentally on a mechatronic powertrain test setup.

ISBN:978-989-99424-4-8

Publication year:2017

Accessibility:Open