The goal of flowsheet digitization is to extract the flowsheet topologies from the flowsheet images and save them in a graph format.
The digitization of chemical process flowsheets involves multiple object detection algorithms and a pathway exploration algorithm. In the object detection step, machine learning models identify the position and type of unit operation on the flowsheet. In the pathway exploration step, the connectivity of the unit operations is explored. Moreover, text and tables are identified and digitized.
We develop a large dataset of labeled chemical flowsheets and P&IDs. We further develop state-of-the-art machine learning models that identify unit operations and their connectivity. The algorithms are trained on a variety of flowsheets from various sources.