Equivalence Judgment of Equation Groups Representing Process Dynamics

Abstract

Abstract Digital twins are expected to play a key role in digital transformation of the process industry. Although process informatics, i.e., process data analytics, has attracted a lot of attention, physical models are essential to realizing the digital twins. Building a physical model of a complex industrial process is toil. We aim to free the engineers from physical model building by developing an automated physical model builder, AutoPMoB. AutoPMoB performs five tasks: 1) retrieving documents regarding a target process from literature databases, 2) converting the format of each document to HTML format, 3) extracting information necessary for building a physical model from the documents, 4) judging the equivalence of the information extracted from different documents, and 5) reorganizing the information to output a desired physical model. In this study, we propose a method of judging the equivalence of two equation groups to accomplish task 4. The proposed method first converts two equation groups in mathematical markup language (MathML) format into a format that can be manipulated in a computer algebra system (CAS). Then, the variables not shared between the groups are eliminated using the CAS. The method judges whether the two equation groups are equivalent. The results of several case studies demonstrated that the proposed method accurately judged the equivalence, including physical models of a continuous stirred-tank reactor. We also developed a web application that can easily judge the equivalence of MathML-formatted equation groups. This application is expected to reduce the effort required to find out the different models contained in multiple documents, and become an important part of AutoPMoB.