Using Flory-Huggins-Informed Human-in-the-Loop Bayesian Optimization to Map the Phase Diagram of Polymer Blends

This record contains the code, data, and hardware design files associated with the manuscript: "Using Flory-Huggins-Informed Human-in-the-Loop Bayesian Optimization to Map the Phase Diagram of Polymer Blends" The repository contains the implementation of a Flory-Huggins-informed human-in-the-loop Bayesian optimization workflow used to construct and analyze polymer blend phase diagrams. Included in this repository are: Jupyter notebooks implementing experimental control flow, validation simulations, and an example of instantiating a Gaussian process optimizer Python source files for Gaussian process surrogate modeling and optimization Excel spreadsheets with grayscale pixel intensity measurements extracted from sample images, as well as calibration data Example image files CAD (STEP) file for the imaging system used to collect sample images J.C.H. acknowledges support from the Vagelos Institute for Energy Science and Technology (VIEST) at the University of Pennsylvania.