000888840 001__ 888840 000888840 005__ 20240712112851.0 000888840 0247_ $$2arXiv$$aarXiv:2010.03405 000888840 0247_ $$2Handle$$a2128/26520 000888840 0247_ $$2altmetric$$aaltmetric:91965610 000888840 037__ $$aFZJ-2020-05254 000888840 1001_ $$0P:(DE-HGF)0$$aSchweidtmann, Artur M$$b0 000888840 245__ $$aObey validity limits of data-driven models 000888840 260__ $$c2020 000888840 3367_ $$0PUB:(DE-HGF)25$$2PUB:(DE-HGF)$$aPreprint$$bpreprint$$mpreprint$$s1607952690_18333 000888840 3367_ $$2ORCID$$aWORKING_PAPER 000888840 3367_ $$028$$2EndNote$$aElectronic Article 000888840 3367_ $$2DRIVER$$apreprint 000888840 3367_ $$2BibTeX$$aARTICLE 000888840 3367_ $$2DataCite$$aOutput Types/Working Paper 000888840 520__ $$aData-driven models are becoming increasingly popular in engineering, on their own or in combination with mechanistic models. Commonly, the trained models are subsequently used in model-based optimization of design and/or operation of processes. Thus, it is critical to ensure that data-driven models are not evaluated outside their validity domain during process optimization. We propose a method to learn this validity domain and encode it as constraints in process optimization. We first perform a topological data analysis using persistent homology identifying potential holes or separated clusters in the training data. In case clusters or holes are identified, we train a one-class classifier, i.e., a one-class support vector machine, on the training data domain and encode it as constraints in the subsequent process optimization. Otherwise, we construct the convex hull of the data and encode it as constraints. We finally perform deterministic global process optimization with the data-driven models subject to their respective validity constraints. To ensure computational tractability, we develop a reduced-space formulation for trained one-class support vector machines and show that our formulation outperforms common full-space formulations by a factor of over 3,000, making it a viable tool for engineering applications. The method is ready-to-use and available open-source as part of our MeLOn toolbox (https://git.rwth-aachen.de/avt.svt/public/MeLOn). 000888840 536__ $$0G:(DE-HGF)POF3-899$$a899 - ohne Topic (POF3-899)$$cPOF3-899$$fPOF III$$x0 000888840 588__ $$aDataset connected to arXivarXiv 000888840 7001_ $$0P:(DE-HGF)0$$aWeber, Jana M$$b1 000888840 7001_ $$0P:(DE-HGF)0$$aWende, Christian$$b2 000888840 7001_ $$0P:(DE-HGF)0$$aNetze, Linus$$b3 000888840 7001_ $$0P:(DE-Juel1)172025$$aMitsos, Alexander$$b4$$eCorresponding author$$ufzj 000888840 8564_ $$uhttps://arxiv.org/abs/2010.03405 000888840 8564_ $$uhttps://juser.fz-juelich.de/record/888840/files/2020_Schweidtmann_Obey_Validity_Limits.pdf$$yOpenAccess 000888840 909CO $$ooai:juser.fz-juelich.de:888840$$pdnbdelivery$$pdriver$$pVDB$$popen_access$$popenaire 000888840 9101_ $$0I:(DE-588b)36225-6$$6P:(DE-HGF)0$$aRWTH Aachen$$b0$$kRWTH 000888840 9101_ $$0I:(DE-588b)36225-6$$6P:(DE-HGF)0$$aRWTH Aachen$$b2$$kRWTH 000888840 9101_ $$0I:(DE-588b)36225-6$$6P:(DE-HGF)0$$aRWTH Aachen$$b3$$kRWTH 000888840 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)172025$$aForschungszentrum Jülich$$b4$$kFZJ 000888840 9101_ $$0I:(DE-588b)36225-6$$6P:(DE-Juel1)172025$$aRWTH Aachen$$b4$$kRWTH 000888840 9131_ $$0G:(DE-HGF)POF3-899$$1G:(DE-HGF)POF3-890$$2G:(DE-HGF)POF3-800$$3G:(DE-HGF)POF3$$4G:(DE-HGF)POF$$aDE-HGF$$bProgrammungebundene Forschung$$lohne Programm$$vohne Topic$$x0 000888840 9141_ $$y2020 000888840 915__ $$0StatID:(DE-HGF)0510$$2StatID$$aOpenAccess 000888840 920__ $$lyes 000888840 9201_ $$0I:(DE-Juel1)IEK-10-20170217$$kIEK-10$$lModellierung von Energiesystemen$$x0 000888840 9801_ $$aFullTexts 000888840 980__ $$apreprint 000888840 980__ $$aVDB 000888840 980__ $$aUNRESTRICTED 000888840 980__ $$aI:(DE-Juel1)IEK-10-20170217 000888840 981__ $$aI:(DE-Juel1)ICE-1-20170217