Jennifer Werner

It is well known that the widely used activity coefficient models NRTL and UNIQUAC are not unique: a given pair of activity coefficients can be represented by different pairs of interaction parameters. This non-uniqueness poses challenges for model parameterization. In this work, we present an in-depth analysis of the nature of these multiplicities and identify the regions in parameter space where they occur. Based on these insights, we develop a strategy to eliminate multiplicities without restricting the range of activity coefficients the models can represent. Thus a one-to-one mapping between activity coefficients and parameters is established, resulting in unique NRTL and UNIQUAC models, being otherwise identical to their parent models. Finally, we discuss how the unique models can be applied to parameter fitting and illustrate the resulting advantages.

Accurate prediction of urban wind flow is essential for urban planning and environmental assessment. Classical computational fluid dynamics (CFD) methods are computationally expensive, while machine learning approaches often lack explainability and generalizability. To address the limitations of both approaches, we propose Diff-KNN, a hybrid method that combines Coarse-Scale CFD simulations with a K-Nearest Neighbors (KNN) model trained on the residuals between coarse- and fine-scale CFD results. Diff-KNN reduces velocity prediction errors by up to 83.5% compared to Pure-KNN and 56.6% compared to coarse CFD alone. Tested on the AIJE urban dataset, Diff-KNN effectively corrects flow inaccuracies near buildings and within narrow street canyons, where traditional methods struggle. This study demonstrates how residual learning can bridge physics-based and data-driven modeling for accurate and interpretable fine-scale urban wind prediction.

Thermodynamic activity coefficients play an essential role in the description of phase equilibria, which are fundamental to the simulation and optimization of all separation processes. Although it is known that multiplicities with respect to binary interaction parameters can occur, no systematic study of these multiplicities has been undertaken yet. In this contribution, their rigorous pointwise classification is presented, valid for the entire range of binary compositions. Furthermore, implications not only for the non‐random two‐lqiuid (NRTL)‐parameterization to capture temperature‐dependence are discussed, but also for extrapolation and regression tasks. The findings are illustrated for vapor‐liquid equilibria of different binary mixtures.

For the location optimization of several ambulance stations, we outline a multi-criteria integer program to maximize the reached population within different time bounds. This technique is based on a realistic driving time estimation for ambulances on a detailed road network. The results show that, depending on which time bound is more valued, ambulance stations are more likely to be located in urban areas or evenly distributed across the study region. We used this model for real-world studies in Southwest Germany, where several ambulance station locations were optimized simultaneously.