Benefits of Using a Real-Time Engine Model During Engine ECU Development 2003-01-1049
In the world of automotive control and electronics, technology is the driving force behind every decrease in product cycle time. Cost and time to market are crucial factors to ensure continued success in a competitive world where products must be tailored precisely to the customers' needs. To achieve these goals, co-operation between suppliers and vehicle manufacturers need to reach unprecedented levels through the use of common tools.
This paper aims to show the competitive edge that may be derived from using current state-of-the-art control development tools and methodologies. The tools shift the emphasis away from expensive on-vehicle work to lever the power of simulation and engine emulation in the development of a prototype EMS (Engine Management System) controller. Focus will be placed on the ECU (Electronic Control Unit) development process, new control algorithms and the benefits derived from using real time capable engine models, Software-In-the-Loop (SiL), Rapid Control Prototyping (RCP) and Hardware-In-the-Loop (HiL) based development methodology.
We are discussing the development of engine ECU functions looking in detail at functions like idle speed control, Air Fuel ratio control during transients, driveability, take off and EOBD.
The controller was first developed in Simulink® and tested by using the SiL method with an engine model, the controller was then put on a development ECU by using RP and the prototype ECU was tested on the HiL simulator before testing it in the real vehicle.
One of the key issues of the described methodology is to shift as much of the development as possible off-vehicle. The vehicle is used purely for the important final calibration task where subjective feel is engineered into the vehicle behaviour and final system integrity checked for real-world robustness.
For the development process it was very important to use the same engine model for SiL and HiL, reducing the effort of re-parameterisation and maintenance of several models.
We will point out the key features of the model used as well as results obtained with the described methodology and the model. The results will also be illustrated by several plots.