Automated model based engine calibration procedure using co-simulation

The final validation and sign-off of a production powertrain control module (PCM) calibration is a time-consuming and expensive task and requires a high degree of expertise. There are two main reasons for this; firstly, the validation test is an iterative process due to the fact that calibration changes may affect the true operating point of the engine at the desired test point. Secondly, modifications to the calibration require expert knowledge of the complete control strategy so as to improve the correlation to validation data without potentially negatively impacting the correlated mapping points. This paper describes the implementation of an optimisation routine on a virtual platform in order to both reduce the requirement for experimental testing during the validation procedure, and for development of the optimisation routine itself prior to execution on the engine dynamometer. It is shown that in simulation, the optimisation routine is capable of producing an acceptable calibration within just 5 iterations, reducing the 11-week process down to just a few days. It is also concluded that there are also a number of further improvements that could be made to further improve the efficiency of this process.