Flowsheet generation through hierarchical reinforcement learning and graph neural networks

Process synthesis experiences a disruptive transformation accelerated by artificial intelligence. We propose a reinforcement learning algorithm for chemical process design based on a state-of-the-art actor-critic logic. Our proposed algorithm represents chemical processes as graphs and uses graph convolutional neural networks to learn from process graphs. In particular, the graph neural networks are implemented within the agent architecture to process the states and make decisions. We implement a hierarchical and hybrid decision-making process to generate flowsheets, where unit operations are placed iteratively as discrete decisions and corresponding design variables are selected as continuous decisions. We demonstrate the potential of our method to design economically viable flowsheets in an illustrative case study comprising equilibrium reactions, azeotropic separation, and recycles. The results show quick learning in discrete, continuous, and hybrid action spaces. The method is predestined to include large action-state spaces and an interface to process simulators in future research.     

Graph machine learning for design of high-octane fuels

Fuels with high-knock resistance enable modern spark-ignition engines to achieve high efficiency and thus low CO₂ 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.     

Monte Carlo simulation of terpolymerization: Optimizing the simulation and post-processing times

Monte Carlo simulations are a useful and easy way to understand a polymerization reaction process properly. However, achieving reliable results with Monte Carlo simulations can also lead to prohibitive computational times and a considerable amount of data to be processed afterward. The present study analyses the Monte Carlo simulation of a steady-state terpolymerization process to reduce the overall computational time of the simulation and the post-processing of its results. Different sorting algorithms (Bubble, Insertion, Selection, and Tim) and Python libraries (Joblib and Numba) were used. The chain composition distribution and the micro-structures resultant of different scenarios were assessed by processing the simulated mechanism results. The simulation time results indicate the Tim sorting algorithm as the best to use in the post-processing step and the Numba library as the best suited for both the simulation and the post-processing step.