Condition monitoring of components in internal combustion engines is an essential tool for increasing engine durability and avoiding critical engine operation. If lubrication at the crankshaft main bearings is insufficient, metal-to-metal contacts become likely and thus wear can occur. Bearing temperature measurements with thermocouples serve as a reliable, fast responding, individual bearing-oriented method that is comparatively simple to apply. In combination with a corresponding reference model, such measurements could serve to monitor the bearing condition. Based on experimental data from an MAN D2676 LF51 heavy-duty diesel engine, the derivation of a data-driven model for the crankshaft main bearing temperatures under steady-state engine operation is discussed. A total of 313 temperature measurements per bearing are available for this task. Readily accessible engine operating data that represent the corresponding engine operating points serve as model inputs. Different machine learning methods are thoroughly tested in terms of their prediction error with the help of a repeated nested cross-validation. The methods include different linear regression approaches (i.e., with and without lasso regularization), gradient boosting regression and support vector regression. As the results show, support vector regression is best suited for the problem. In the final evaluation on unseen test data, this method yields a prediction error of less than 0.4 °C (root mean squared error). Considering the temperature range from approximately 76 °C to 112 °C, the results demonstrate that it is possible to reliably predict the bearing temperatures with the chosen approach. Therefore, the combination of a data-driven bearing temperature model and thermocouple-based temperature measurements forms a powerful tool for monitoring the condition of sliding bearings in internal combustion engines.

The fatigue properties and cyclic deformation behaviour of materials are usually determined using cylindrical specimens (e.g., Ø7.5 mm). Since the transferability to very small dimensions has not been comprehensively considered so far, this study investigates the transferability of specimen data from high-strength steel X5CrNiCuNb16-4 to real structures characterised by a wall thickness of 100 µm. Regarding fatigue, extensive calculations demonstrate that for notched specimens and thin structures, both the material-mechanical support factor concept according to the FKM guideline and the point method of the Theory of Critical Distance (TCD) make correct predictions of the local fatigue strength, with a maximum deviation from experimental values of less than 5%. However, the study points out that the TCD is only conditionally applicable for thin-walled structures, as the material parameter $a_0$ must be significantly smaller than the wall thickness. Regarding the deformation behaviour, the material reveals special flow characteristics in the first hysteresis. Nevertheless, a combined hardening approach is suitable for modelling. The validation of the model by a plastic deformation of the structure seems plausible, although geometric influences prove to be dominant. In conclusion, even 100 µm thin structures can be evaluated using conventional specimen tests and established assessment or modelling methods.

Hydrogen is a promising option to decarbonize energy conversion in internal combustion engines (ICE). A key limiting factor of high-load hydrogen ICE operation is the occurrence of combustion anomalies, including pre-ignition, backfiring into the intake manifold and knocking combustion. The existing literature indicates that there currently exists no generally accepted conceptual model that is able to explain the occurrence of all combustion anomalies for hydrogen-fueled internal combustion engines. This publication is intended to help fill this research gap based on experimental investigations on a single-cylinder research engine (SCE) with hydrogen port fuel injection and spark ignition in the 3 dm ³ displacement per cylinder class and related 3D-CFD simulation work. To enable in-depth combustion analysis and particularly to identify the origin and root cause of pre-ignition, the SCE was instrumented with AVL List GmbH’s “Visiolution” system, an advanced optical instrumentation that relies on fiberoptic sensors in the combustion chamber. Different types of pre-ignition including backfire, early and late pre-ignition could be provoked during engine operation. Backfire and early pre-ignition were found to originate from the piston-liner edge at the “bottom” of the combustion chamber and tend to occur in specific areas at the piston circumference where comparatively rich local mixture prevails. Mixture homogeneity may be key to reduce the risk of backfire and early pre-ignition to occur. The locations of origin suggest that various ignition sources emerging from the piston top land crevice such as hot residual gases or reactive species (e.g. due to reverse blowby) are possible. Late pre-ignition events (i.e. shortly before ignition timing) occurred in large numbers with a random spatial distribution in the combustion chamber and appear to be mostly related to lubricant oil droplet combustion. Future research must study the impact of mixture homogeneity on pre-ignition and strive to better understand the root causes for the underlying processes that provide the required ignition energy.