<p dir="ltr">For turbocharged engine simulation analysis, manufacturer-provided data are typically used to predict the mass flow and efficiency of the turbine. To create a turbine map, physical tests are performed in labs at various turbine speeds and expansion ratios. These tests can be very expensive and time-consuming. Current testing methods can have limitations that result in errors in the turbine map. As such, only a modest set of data can be generated, all of which have to be interpolated and extrapolated to create a smooth surface that can then be used for simulation analysis.</p><p><br></p><p dir="ltr">The current method used by the manufacturer is a physics-informed polynomial regression model that depends on the Blade Speed Ratio (BSR ) in the polynomial function to model the efficiency and MFP. This method is memory-consuming and provides a lower-than-desired accuracy. This model is decades old and must be updated with new state-of-the-art Machine Learning models to be more competitive. Currently, CTT is facing up to +/-2% error in most turbine maps for efficiency and MFP and the aim is to decrease the error to 0.5% while interpolating the data points in the available region. The current model also extrapolates data to regions where experimental data cannot be measured. Physical tests cannot validate this extrapolation and can only be evaluated using CFD analysis.</p><p><br></p><p dir="ltr">The thesis focuses on investigating different AI techniques to increase the accuracy of the model for interpolation and evaluating the models for extrapolation. The data was made available by CTT. The available data consisted of various turbine parameters including ER, turbine speeds, efficiency, and MFP which were considered significant in turbine modeling. The AI models developed contained the above 4 parameters where ER and turbine speeds are predictors and, efficiency and MFP are the response. Multiple supervised ML models such as SVM, GPR, LMANN, BRANN, and GBPNN were developed and evaluated. From the above 5 ML models, BRANN performed the best achieving an error of 0.5% across multiple turbines for efficiency and MFP. The same model was used to demonstrate extrapolation, where the model gave unreliable predictions. Additional data points were inputted in the training data set at the far end of the testing regions which greatly increased the overall look of the map.</p><p><br></p><p dir="ltr">An additional contribution presented here is to completely predict an expansion ratio line and evaluate with CTT test data points where the model performed with an accuracy of over 95%. Since physical testing in a lab is expensive and time-consuming, another goal of the project was to reduce the number of data points provided for ANN model training. Furthermore, strategically reducing the data points is of utmost importance as some data points play a major role in the training of ANN and can greatly affect the model's overall accuracy. Up to 50% of the data points were removed for training inputs and it was found that BRANN was able to predict a satisfactory turbine map while reducing 20% of the overall data points at various regions.</p>

In den vorliegenden Untersuchungen wurde das Ziel verfolgt die Nutzbarkeit des Milchinhaltsstoffes Laktose als praxistaugliche Managementhilfe zu prüfen. Die Primärdaten stammen aus drei israelischen Hochleistungsherden, über mehrere Laktationen erhoben. Der Parameter Laktosegehalt wurde in der Datenaufbereitung dahingehend geprüft, ob dieser zur Gesundheits- und Leistungsvorhersage ausreicht oder welche zusätzlichen Merkmale für die Verwendung in einem Prognose-Modell von Bedeutung sein könnten. Als leistungs- bzw. gesundheitsrelevante Ereignisse (Events) wurden Brunst, Diarrhoe, Endometritis, Fieber, Infektionen, Klauenerkrankungen, Mastitis, Stress, Stoffwechselstörungen sowie Verletzungen zugeordnet. Die Bewertung der Nützlichkeit einzelner Merkmale für die Prädiktion erfolgte anhand der Erkennungsraten. Zwei- und dreistufige Entscheidungsbäume wurden entwickelt, um diese Events zu identifizieren. Ein einzelnes Merkmal ist oft nicht ausreichend, weshalb verschiedene Kombinationen von Variablen analysiert wurden. Die wichtigste Erkenntnis der vorliegenden Arbeit besteht darin, dass der Abfall der Laktosekonzentration und Laktosemenge immer ein kritisches Ereignis darstellt. Das Hauptziel eines Gesundheitsmonitorings im Milchkuhbestand sollte deshalb darin bestehen, frühzeitig eine Stoffwechselüberlastung "sichtbar" oder "erkennbar" zu machen. Unabhängig davon, welche Erkrankung sich anbahnt, muss das Herdenmanagement darauf hinwirken, die Glukoseversorgungssituation des Einzeltieres zu verbessern. Aus der Analyse für die einzelnen Herden und Laktationen kann grundlegend abgeleitet werden, dass die Ergebnisse der Milchkontrolldaten, die im Zuge der datengestützten Herdenüberwachung erhoben wurden, sich verwenden lassen, um den Leistungs- und Gesundheitsstatus der Kühe im Laktationsverlauf einzuschätzen und zu prognostizieren. Die Verwendung von Informationen zum Laktosegehalt des Gemelks verbesserten in jedem Fall die Erkennungsraten. / The aim of the current studies was to investigate whether the milk ingredient lactose can be used as a practical support management. The primary data comes from three Israeli high-performance herds, collected over several lactations. In the data preparation, the parameter "lactose content" was examined to see whether it is sufficient for a health and performance prediction or whether additional features may be of importance for usage in a forecasting model. Oestrus, diarrhea, endometritis, fever, infections, hoof diseases, mastitis, stress, metabolic disorders, and injuries have been assigned to the performance- and/or health-affecting events. The usefulness of individual features for the prediction was evaluated on the basis of the recognition rates. Thus two- and three-level decision trees have been developed to identify these events. As one single feature is often insufficient, different combinations of variables were analyzed. The most important finding of this study is that the drop in the lactose concentration and lactose quantity always represents a critical event. The main objective of a health monitoring in the dairy herd should therefore be to make a metabolic overload "visible" or "recognisable" at an early stage. Whichever disease begins to take shape, the herd management must work on improving the glucose supply situation of the individual animal. In conclusion from the analysis of the individual herds and lactations it can be inferred that the results of the milk control data collected in the course of the data-based herd monitoring can be used in order to assess and to predict the performance and health status of the cows in the course of lactation. The use of information on the lactose content of the milk improved in any case the recognition rates.