Search Results

Prospective combustion engine applications require the highest possible energy conversion efficiencies for environmental and economic sustainability. For conventional Spark-Ignition (SI) engines, the quasi-hemispherical flame propagation combustion method can only be significantly optimized in combination with high excess air dilution or increased combustion speed. However, with increasing excess air dilution, this is difficult due to decreasing flame speeds and flammability limits. Pre-Chamber (PC) initiated jet ignition combustion systems significantly shift the flammability and flame stability limits towards higher dilution areas due to high levels of introduced turbulence and a significantly increased flame area in early combustion stages, leading to considerably increased combustion speeds and high efficiencies. By now, vehicle implementations of PC-initiated combustion systems remain niche applications, especially in combination with lean mixtures.

Future combustion engine applications require highest possible energy conversion efficiencies to reduce their environmental impact and be economically competitive. So far, spark-ignition (SI) engine combustion development mostly consisted of optimizing the hemispherical flame propagation combustion method. Thereby, a significant efficiency increase is only achievable in combination with high excess air dilution or increased combustion speed. However, with increasing excess air dilution, this is difficult due to decreasing flame speeds and flammability limits. Simultaneously, researchers have been investigating homogeneous charge compression ignition (HCCI) that achieves higher efficiencies due to its rapid volume reaction combustion and also enables high excess air dilution. However, the combustion is complex to control as it is initiated by auto-ignition (AI) processes. In-cylinder conditions reliably need to be reproduced to prevent damaging pre-ignitions.

State-of-the-art spark-ignition engines mainly rely on the quasi-hemispherical flame propagation combustion method. Despite significant development efforts to obtain high energy conversion efficiencies while avoiding knock phenomena, achieved indicated efficiencies remain around 35 - 40 %. Further optimizations are enabled by significant excess air dilution or increased combustion speed. However, flammability limits and decreasing flame speeds with increasing air dilution prevent substantial improvements. Pre-Chamber (PC) initiated jet ignition combustion systems improve flame stability and shift flammability limits towards higher dilution levels due to increased turbulence and a larger flame area in the early Main-Chamber (MC) combustion stages. Simultaneously, the much-increased combustion speed reduces knock tendency, allowing the implementation of an innovative combustion method: PC-initiated jet ignition coupled with Spark-Assisted Compression Ignition (SACI).

Whether as an optimization problem or a development tool, neural networks help engineers to work more efficiently. This paper’s central aspect is to add metadata to the core files of the project simulation data. To understand the project and its simulation models, a pre-processing methodology and convolutional neural network architecture are presented. With the added labels, it is possible to access the content of the model files of an engine performance simulation tool without examining them. At first, a pre-processing approach and its design are introduced to extract and filter the desired data from the XML data structure. Then, the data is split into sequences and paired with labels. Expert knowledge is used to label the models. These labels are further paired with the extracted sequences.

High-efficient simulations are mandatory to manage the ever-increasing complexity of automotive powertrain system and reduce development time and costs. Integrating AI methods into the development process provides an ideal solution thanks to massive increase in computational power. Based on an 1D physical engine model of a turbo-charged direct injection gasoline engine with variable valve timing (VVT), a high-performance hybrid simulation model has been developed for increasing computing performance. The newly developed model is made of a physics-based low-pressure part including intake and exhaust peripheries and a neural-network-based high-pressure part for combustion chamber calculations. For the training and validation of the combustion chamber neural networks, a data set with 10.5 million operating points was generated in a short time thanks to the parallelizable combustion chamber simulations in stand-alone mode.

Due to increasingly strict emission regulations, the demand for internal combustion engine performance has enhanced. Combustion stability is one of the main research focuses due to its impacts on the emission level. Moreover, the combustion instability becomes more significant under the lean combustion concept, which is an essential direction of internal combustion engine development. The combustion instability is represented as the cycle-to-cycle variation. This paper presents a quasi-dimensional model system for predicting the cycle-to-cycle variation in 0D/1D simulation. The modeling is based on the cause-and-effect chain of cycle-to-cycle variation of spark ignition engines, which is established through the flow field analysis of large eddy simulation results [1]. In the model system, varying parameters are turbulent kinetic energy, the distribution of air-to-fuel equivalence ratio, and the in-cylinder velocity field.