Explore the vast range of titles available.

The tightening of emission standards and homologation rules lead car manufacturers to rely on simulation testing in early development phases. Coupling an engine to a testbench controlled by a real-time simulation environment allows flexible, reliable, and reproducible testing for consumption and emission studies. However, interest in this method referred to as engine-in-the-loop (EiL) is relatively recent and few details can be found regarding the simulation environment. Following previous work, this study details a driver model based on the PI structure and augmented with preview and anti-windup. The focus is set on a conventional powertrain with a manual transmission for which the driver must also manage the clutch pedal during gearshift and take-off phases. Extended analysis of vehicle tests allows defining the driver’s behavior during these phases for different profiles. The driver model is then tested in the EiL environment and the impact of the gearshift profile on fuel consumption and pollutant emissions can be assessed. Besides the slight increase in fuel consumption, results show that increasing the gearshift duration degrades the regulation of the richness by the ECU, thus increasing CO engine-out emissions as well as decreasing NOx emissions. Finally, results suggest that a longer gearshift also affects the catalyst efficiency, which results in higher NOx tailpipe emissions.

Predicting end-of-process variables in internal combustion engines, such as brake-specific fuel consumption or pollutant emissions, is crucial for engine design decisions. However, errors in common multi-layer-perceptron-based artificial neural network models often match the magnitude of the expected fuel consumption improvements, potentially leading to incorrect decisions. This study introduces a hybrid model where artificial neural networks replace engine block elements, while the 1D gas circuit and turbocharger models are retained. To enhance metamodel accuracy, two modifications are proposed: incorporating a pulsating mass flow rate in the exhaust line to capture pulsating effects missing in mean-value engine models and using a hierarchical arrangement of several multi-layer perceptrons instead of a parallel arrangement. The pulsating mass flow rate approach improves the accuracy of all tested configurations by replicating pulsating effects from a detailed 1D engine model. Meanwhile, the hierarchical arrangement refines predictions of end-of-process variables, such as fuel consumption, by increasing the total layers, with a minimal trade-off in the accuracy of other variables. These findings are validated using a metamodel derived from a calibrated 1D engine model in GT-Suite. The proposed methods are expected to enhance the accuracy of data-driven artificial neural network approaches, contributing to more reliable engine design optimization.

The objective of this study is to understand how each variable impacts the optimal configuration of a marine diesel engine equipped with an electric hybrid air-charging system that allows energy assistance and recovery. The aim is to minimize CO2 emissions by reducing fuel consumption. The hybrid system offers flexibility in adjusting parameters from both the engine and air-charging system. It is compared with the baseline engine, which uses a free-floating turbocharger. The results show a significant improvement at low engine loads, where the baseline engine struggles to provide sufficient air. While turbine speed has little influence, compressor power reduces fuel consumption at low loads. However, at mid loads, resizing the turbomachine is necessary for further improvements. At high loads, full optimization of all variables is required to reduce fuel consumption. The electric hybrid system is particularly effective in tugboat-like conditions, where low loads dominate, but less impactful for ro-pax ferries. Despite the potential of the hybrid system, a fully optimized turbocharger could provide greater benefits due to reduced losses. Future studies could explore combining the adaptability of the hybrid system with a highly efficient turbocharger to reduce emissions across all load conditions.

Automated monitoring systems are now the standard on most large vessels; however, few are equipped with diagnostic systems. This paper presents new developments in the area of fault diagnosis based on intelligent software agents. The research objective was to design an agent capable of continuous real-time machine learning by using an artificial neural network known as the cerebellar model articulation controller (CMAC). An engine simulator that can model both normal and faulty engine operations was used to develop the learning system controller in a flexible and cost-efficient manner. This paper provides a description of the selected CMAC, a brief overview of the real-time engine simulator and its integration with the learning system as well as a few results.