14 September 2021 in Chemical Society Reviews
This study highlights new opportunities for optimal reaction route selection from large chemical databases brought about by the rapid digitalisation of chemical data.
30 March 2023 in Nature Methods
12 May 2021 in Optimization and Engineering
Data-driven models are becoming increasingly popular in engineering, on their own or in combination with mechanistic models. Commonly, the trained models are subsequently used in model-based optimization of design and/or operation of processes. Thus, it is critical to ensure that data-driven models are not evaluated outside their validity domain during process optimization. We propose a method to learn this validity domain and encode it as constraints in process optimization. We first perform a topological data analysis using persistent homology identifying potential holes or separated clusters in the training data. In case clusters or holes are identified, we train a one-class classifier, i.e., a one-class support vector machine, on the training data domain and encode it as constraints in the subsequent process optimization. Otherwise, we construct the convex hull of the data and encode it as constraints. We finally perform deterministic global process optimization with the data-driven models subject to their respective validity constraints. To ensure computational tractability, we develop a reduced-space formulation for trained one-class support vector machines and show that our formulation outperforms common full-space formulations by a factor of over 3000, making it a viable tool for engineering applications. The method is ready-to-use and available open-source as part of our MeLOn toolbox (https://git.rwth-aachen.de/avt.svt/public/MeLOn).
13 February 2018 in Macromolecules
The structure and vibrational density of states (VDOS) of polymer glasses are investigated using numerical simulations based on the classical Kremer–Grest bead–spring model. We focus on the roles of chain length and bending stiffness, the latter being set by imposing three-body angular potentials along chain backbones. Upon increasing the chain length and bending stiffness, structural reorganization leads to volumetric expansion of the material and buildup of internal stresses. The VDOS has two dominant bands: a low-frequency one corresponding to inter- and intrachain nonbonding interactions and a high-frequency one corresponding principally to vibrations of bonded beads that constitute skeletal chain backbones. Upon increasing the steepness of the angular potential, vibrational modes associated with chain bending gradually move from the low-frequency to the high-frequency band. This redistribution of modes is reflected in a reduction of the so-called Boson peak upon increasing chain stiffness. Remarkably, the finer structure and the peaks of the high-frequency band, and their variations with stiffness, can, for short chains, be explained using an analytical solution derived for a model triatomic molecule. For longer chains, the qualitative evolution of the VDOS with chain stiffness is similar, although the distinct peaks observed for short chains become increasingly smoothed out. Our findings can be used to guide a systematic approach to interpretation of Brillouin and Raman scattering spectra of glassy polymers in future work, with applications in polymer processing diagnostics.
22 December 2022 in AIChE Journal
Fuels with high‐knock resistance enable modern spark‐ignition engines to achieve high efficiency and thus low CO 2 emissions. Identification of molecules with desired autoignition properties indicated by a high research octane number and a high octane sensitivity is therefore of great practical relevance and can be supported by computer‐aided molecular design (CAMD). Recent developments in the field of graph machine learning (graph‐ML) provide novel, promising tools for CAMD. We propose a modular graph‐ML CAMD framework that integrates generative graph‐ML models with graph neural networks and optimization, enabling the design of molecules with desired ignition properties in a continuous molecular space. In particular, we explore the potential of Bayesian optimization and genetic algorithms in combination with generative graph‐ML models. The graph‐ML CAMD framework successfully identifies well‐established high‐octane components. It also suggests new candidates, one of which we experimentally investigate and use to illustrate the need for further autoignition training data.
29 August 2024 in Molecular Systems Design & Engineering
Self-supervised learning for polymer property prediction in scarce data domains.