Search Results

For metrological traceability of pressure sensors, static calibration procedures are standard. If these sensors are used in dynamic systems, unexpected phenomena or deviations occur in the recorded signal characteristics. By setting up a dynamic pressure calibration facility, it is possible to investigate this dynamic behavior and learn about the interactions between sensor and investigated system. To be able to identify the disturbing influences and interactions occurring during calibration and in subsequent measurement use, it is necessary to increase the existing understanding of the system. In the context of the contribution, the calibration procedure used, its properties such as repeatability, reproducibility as well as the system interaction of the influencing variables are analyzed. Special attention is paid to the effects of varying gas content in the calibration medium, its influence on the system and on the observed phenomena occurring.

Knocking is still one of today’s major limitations regarding efficiency-increasing measures for SI combustion engines. Due to the complex stochastic nature of the phenomenon, not only prediction and consideration within the engine development is of relevance. A further challenge is control of the phenomenon during engine operation, with the aim of maximizing the efficiency while preventing engine damage and maintaining the driver comfort. Conventional knock control is characterized by detecting knock events and subsequently adjusting the spark timing depending on whether knock was detected. This paper proposes a new knock control concept based on the prediction and direct control of the knock frequency, compared to the conventional reactive and indirect control of the knock frequency. For the prediction of the knock frequency, a calculation approach based on three different parameters is utilized.

Predictive physical modeling is an established method used in the development process for automotive components and systems. While accurate predictions can be issued after tuning model parameters, long computation times are expected depending on the complexity of the model. As requirements for components and systems continuously increase, new optimization approaches are constantly being applied to solve multidimensional objectives and resulting conflicts optimally. Some of those approaches are deemed not feasible, as the computational times for required single predictions using conventional simulation models are too high. To address this issue it is proposed to use data-driven model such as neural networks. Previous efforts have failed due to sparse data sets and resulting poor predictive ability. This paper introduces an AI Toolchain used for data-driven modeling of combustion engine components. Two methods for generating scalable and fully variable datasets will be shown.

In order to comply with more and more stringent emission standards, like EU6 which will be mandatory starting in September 2014, GDI engines have to be further optimized particularly in regard of PN emissions. It is generally accepted that the deposition of liquid fuel wall films in the combustion chamber is a significant source of particulate formation in GDI engines. Particularly the wall surface temperature and the temperature drop due to the interaction with liquid fuel spray were identified as important parameters influencing the spray-wall interaction [1]. In order to quantify this temperature drop at combustion chamber surfaces, surface temperature measurements on the piston of a single-cylinder engine were conducted. Therefore, eight fast-response thermocouples were embedded 0.3 μm beneath the piston surface and the signals were transmitted from the moving piston to the data acquisition system via telemetry.