Virtual Testing and Design Of Experiments for Functional Validation of Cam Phaser Control 2004-01-0785

In order to improve the emission and performance capabilities of modern engines several devices are introduced in the system. One of such devices is the Continuously Variable Cam Phaser (CVCP), which, by shifting the relative angle between the crank and cam shafts, allows to modify the opening and closing times of the inlet and outlet valves with respect to the base engine configuration, thus giving more design freedom. The control of the CVCP is performed by the Engine Control Unit, which both sets the desired values for the CVCP shift angle and commands the CVCP actuator to control it at the correct position.

The new FIRE engine 1.4l 16v also includes a CVCP. The aim of the activity described in the paper was to perform a functional validation of the CVCP position control algorithms, through a sensitivity analysis of the CVCP control performance to a series of environmental and production variables.

In order to reduce at a minimum the costs of this analysis the choice to use virtual testing as an alternative to experimental testing on the engine test bench, was taken. By this choice it was possible to analyze in simulation many more system configurations, than those available for a real experimental set-up. The building bricks for the analysis were: the definition of the test cases, operating conditions, and performance indicators for the analysis; functional simulation models of the CVCP, its actuators and sensors, developed with the device supplier, INA, of the ECU software for the control of the CVCP, and of the main environmental parameters affecting the CVCP control performance; a toolbox for the Design of Experiments, by which, once selected the design parameters subject to production, mounting or operating variability, defining the set of virtual tests to perform; some tools for the visualization and analysis of the results of the DOE.

The sensitivity analysis results showed that the CVCP control system was robust to parameters variability within the tolerance ranges, being able to satisfy the prescribed performance targets.