Search Results

Performance predictions of advanced turbocharged engines are becoming difficult because conventional engine models are built using performance map data of turbochargers with a proportional integral derivative (PID) controller. Improving prediction capabilities under transient test cycles or real driving conditions is a challenging task. This study applies a machine learning technique to predict turbocharger performances with high accuracy under steady-state and transient conditions. The manipulated signals of engine speed and torque created based on Compressed High-Intensity Radiated Pulse (Chirp signal) and Amplitude-modulated Pseudo-Random Binary Signal (APRBS) are used as inputs to the engine testbed. Data from the engine experiments are used as training data for the AI-based turbocharger model. High prediction accuracy of the AI turbocharger model is achieved with the co-efficient of determination in the model, and cross-validation results are higher than 0.8.

Emission regulations are becoming tighter, and Real Driving Emissions (RDE) is proposed as a testing cycle for evaluating modern engine emissions under a wide operation range. For this reason, engine manufacturers have been developing a method to effectively assess engine performances and emissions under a wide range of transient conditions. Transient engine performances can be evaluated efficiently by applying time-series data created by chirp signals. However, when the time-series data produced by the chirp signal are used directly, the engine hardware may damage, and emission performances deteriorate drastically. It is therefore essential to develop a method to avoid these undesirable operating conditions. This work aims to develop an algorithm to avoid such unrealistic operation conditions for engine performance evaluation. A virtual diesel engine (VDE) model is developed based on a four-cylinder engine using GT-POWER software.