Task:

The behavior of real systems can be modeled in different ways. In addition to the "classical" methods, which describe a process using physical phenomenological approaches, among others, data-based models have increasingly gained importance in recent years. With the help of various machine learning algorithms (shallow learning/deep learning), very good results can be achieved in a short time given a sufficient data basis. However, one challenge here is that as the complexity of the processes increases, so does the depth of data required, and thus the measurement effort required to achieve this. In addition, the validity of the data-based models is linked to the prevailing conditions during the measurements (temperature, pressure,...), whereas physical-phenomenological models can respond within a certain range and still provide a meaningful result.

The aim of this literature study is to elaborate the current state of research regarding a combination of physical-phenomenological and data-based models. Of particular interest are so-called grey-box models and PINN (physics-informed neural networks), primarily in the field of drive technology. The work is particularly suitable as a study, project or bachelor thesis; if interested, a concrete application (modeling) in the context of a master/diploma thesis is also possible after successful completion.

Ansprechpartner:

Task:

Simulations are playing an increasingly important role in the development of combustion engines. In particular, 1D simulation has established itself as a powerful tool for identifying optimization potential in the area of charge change and combustion parameters. By applying this simulation method, targeted insights can be gained that lead to improvements in engine efficiency and performance.

In the context of this work, the focus is on transferring an existing 3D model of a variable valve train of an NFZ dual-fuel combustion engine into the 1D simulation environment GT-Suite and subsequently validating it.

The task includes familiarization with the functionality of modern variable valve trains (VVT) and the use of the simulation software GT-SUITE. The task consists of transferring an existing 3D model of a variable valve train of a commercial vehicle dual-fuel combustion engine into the 1D simulation environment GT-Suite and subsequently validating it. The actual modeling, simulation and final validation of the variable valve train is performed using GT-Power.

During the course of the work, various investigations are performed on the model to evaluate the functionality and performance of the variable valve train and to identify possible optimization opportunities.

The main objective of this work is to gain a sound understanding of variable valve train simulation and the application of the GT suite. The gained knowledge and simulation results shall make a valuable contribution to the further improvement of the NFZ dual-fuel combustion engine. In this context, a structured approach to modeling and evaluation of the simulation results is of particular importance in order to achieve robust results.

contact person:

Task:

The simulation of flow processes in combustion engines using computational fluid dynamics (CFD) has established itself as a powerful method. Both one-dimensional (1D) and three-dimensional (3D) simulations are used. As part of the search for more environmentally friendly propulsion solutions, dual-fuel engines are becoming increasingly important. This technology allows the combustion of two different fuels in one engine and promises a more efficient combustion and reduction of emissions.

This student research project will comprehensively investigate the current state of research in 1D and 3D simulation of dual-fuel engines. The focus will be on both one-dimensional (1D) and three-dimensional (3D) simulation of flow processes in dual-fuel engines.

The literature review will include relevant scientific publications, technical articles and research papers from national and international sources. Special attention will be given to the introduction of dual-fuel technology and the use of different fuel combinations in internal combustion engines, the importance and application of 1D simulation for dual-fuel engine analysis, and the modeling methods and approaches specifically for dual-fuel engines in 3D CFD simulation. In addition, the accuracy of 1D and 3D simulation models will be compared with respect to experimental data, and current research trends and future developments in 1D and 3D simulation of dual-fuel engines will be highlighted.

The results of the literature review will be summarized in a comprehensive written paper. The paper will provide an overview of the current state of research, highlight important findings as well as open questions in the field of 1D and 3D simulation of dual-fuel engines

contact person: