Integrating FMI and ML/AI models on the open-sourcedigital twin framework OpenTwins

Abstract

The realm of digital twins is experiencing rapid growth and presents a wealth ofopportunities for Industry 4.0. In conjunction with traditional simulation meth-ods, digital twins offer a diverse range of possibilities. However, many existingtoolsinthedomainofopen-sourcedigitaltwinsconcentrateonspecificusecasesand do not provide a versatile framework. In contrast, the open-source digitaltwin framework, OpenTwins, aims to provide a versatile framework that can beappliedtoawiderangeofdigitaltwinapplications.Inthisarticle,weintroduceare-definition of the original OpenTwins platform that enables the managementof custom simulation services and the management of FMI simulation services,whichisoneofthemostwidelyusedsimulationstandardsintheindustryanditscoexistence with machine learning models, which enables the definition of thenext-gendigitaltwins.Thankstothisintegration,digitaltwinsthatreflectrealitybetter can be developed, through hybrid models, where simulation data can feedthe scarcity of machine learning data and so forth. As part of this project, a sim-ulation model developed through the hydraulic software Epanet was validatedin OpenTwins, in addition to an FMI simulation service. The hydraulic modelwas implementedandtestedinanagricultural usecaseincollaboration withtheUniversity of Córdoba, Spain. A machine learning model has been developed toassess the behavior of an FMI simulation through machine learning.