An innovative approach for the simulation of the interaction between the software to be implemented in the Electronic Control Unit (ECU) of the vehicle and the engine is described. The aim was to perform a complete simulation of the engine coupled with the ECU, in order to support multi-disciplinary development and to enable engineers to verify and validate control models in early development stages, reducing costs by performing fewer live engine tests. Also, the simultaneous simulation of the models allows to study their interaction, thus allowing an early exploration of the possible design choices over multiple disciplines.
A first prototype of the coupling has been implemented, with an emphasis on realizing a common notion of time and a proper treatment of data exchange between the control model and the engine model. The approach followed for the development of the coupling technique has been described to show how the different environments can share and manage the memory area allocated for the storage of the control variables.
The on-board pressure drop measurement system of a diesel engine, integrated in a feedback loop with the control algorithm of the ECU, has been simulated by a 1D mathematical DPF model (representing a virtual probe) coupled with different control models (written in Simulink), which modifies the engine control variables while the simulation is running. The 1D DPF model, developed by the authors, has been previously validated and calibrated by a comparison with the experimental data available from literature [3, 4], in order to correctly simulate the signal of the virtual probes.