{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,23]],"date-time":"2026-10-23T10:41:56Z","timestamp":1792752116338,"version":"4.5.0"},"reference-count":32,"publisher":"American Chemical Society (ACS)","issue":"33","license":[{"start":{"date-parts":[[2026,8,13]],"date-time":"2026-08-13T00:00:00Z","timestamp":1786579200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100002347","name":"Bundesministerium f?r Forschung, Technologie und Raumfahrt","doi-asserted-by":"publisher","award":["FKZ 03SFK3E1-3"],"award-info":[{"award-number":["FKZ 03SFK3E1-3"]}],"id":[{"id":"10.13039\/501100002347","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2026,8,25]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>In this work, a nonlinear model predictive control framework using a data-driven prediction model is used to control an air separation unit (ASU). The framework is integrated into an industrial automation platform, providing real-time control irrespective of the base layer control system vendor. The approach is a scalable solution that requires minimal effort in model training, leveraging an automated training framework. It is a powerful alternative to physically modeled nonlinear model predictive control (NMPC) and has already been validated by controlling several in-production-critical Linde ASUs. Additionally, a multimodel approach is introduced that combines several data-driven prediction models to only exploit the main dependencies of the considered process variables. Furthermore, the approach enables fast implementation at the plant as step tests are not required, which are usually time-consuming and interfere with the normal operation of the plant. Based on two applications, the control performance is presented for different load change scenarios, ensuring compliance with all product specifications. Furthermore, different operating modes, including different products, different equipment, and a different overall control task, can be captured by a single nonlinear model.<\/jats:p>","DOI":"10.1021\/acsomega.6c06082","type":"journal-article","created":{"date-parts":[[2026,8,13]],"date-time":"2026-08-13T20:02:03Z","timestamp":1786651323000},"page":"50281-50294","source":"Crossref","is-referenced-by-count":0,"title":["Data-Driven Nonlinear\nModel Predictive Control for\nAir Separation Units"],"prefix":"10.1021","volume":"11","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-1576-9253","authenticated-orcid":true,"given":"Valentin","family":"Krespach","sequence":"first","affiliation":[{"name":"Technical University of Munich , , Boltzmannstr. 15 , ,","place":["Garching, Germany, 85748"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Nicolas","family":"Blum","sequence":"additional","affiliation":[{"name":"Technical University of Munich , , Boltzmannstr. 15 , ,","place":["Garching, Germany, 85748"]},{"name":"Linde Engineering , , Dr.-Carl-von-Linde-Str. 6-14 , ,","place":["Pullach, Germany, 82049"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Martin","family":"Pottmann","sequence":"additional","affiliation":[{"name":"Linde Engineering , , Dr.-Carl-von-Linde-Str. 6-14 , ,","place":["Pullach, Germany, 82049"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Shreya","family":"Bhatia","sequence":"additional","affiliation":[{"name":"Linde Engineering , , Dr.-Carl-von-Linde-Str. 6-14 , ,","place":["Pullach, Germany, 82049"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Felix","family":"Kramer","sequence":"additional","affiliation":[{"name":"Linde Engineering , , Dr.-Carl-von-Linde-Str. 6-14 , ,","place":["Pullach, Germany, 82049"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lena","family":"Tetampel","sequence":"additional","affiliation":[{"name":"Linde Engineering , , Dr.-Carl-von-Linde-Str. 6-14 , ,","place":["Pullach, Germany, 82049"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Parth","family":"Shah","sequence":"additional","affiliation":[{"name":"Linde Engineering , , Dr.-Carl-von-Linde-Str. 6-14 , ,","place":["Pullach, Germany, 82049"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Gerhard","family":"Zapp","sequence":"additional","affiliation":[{"name":"Linde Engineering , , Dr.-Carl-von-Linde-Str. 6-14 , ,","place":["Pullach, Germany, 82049"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Sebastian","family":"Rehfeldt","sequence":"additional","affiliation":[{"name":"Technical University of Munich , , Boltzmannstr. 15 , ,","place":["Garching, Germany, 85748"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Harald","family":"Klein","sequence":"additional","affiliation":[{"name":"Technical University of Munich , , Boltzmannstr. 15 , ,","place":["Garching, Germany, 85748"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"316","published-online":{"date-parts":[[2026,8,13]]},"reference":[{"key":"2026102306021959800_cit1","doi-asserted-by":"publisher","first-page":"295","DOI":"10.1016\/j.ifacol.2018.11.028","article-title":"Economic Nonlinear\nModel Predictive Control for Flexible Operation of Air Separation\nUnits","volume":"51","author":"Caspari","year":"2018","journal-title":"IFAC-PapersOnLine"},{"key":"2026102306021959800_cit2","doi-asserted-by":"publisher","first-page":"1937","DOI":"10.1002\/cite.202100094","article-title":"Investigation of a\nModel-Based Deep Reinforcement Learning Controller\nApplied to an Air Separation Unit in a Production Environment","volume":"93","author":"Blum","year":"2021","journal-title":"Chem. Ing. Tech."},{"key":"2026102306021959800_cit3","doi-asserted-by":"publisher","first-page":"1702","DOI":"10.1109\/TAC.2020.3000182","article-title":"Data-Driven\nModel Predictive Control With Stability and Robustness Guarantees","volume":"66","author":"Berberich","year":"2021","journal-title":"IEEE Trans. Autom. Control"},{"key":"2026102306021959800_cit4","doi-asserted-by":"publisher","first-page":"2091","DOI":"10.3390\/pr11072091","article-title":"Machine Learning-Based\nPrediction of Controlled Variables of APC\nSystems Using Time-Series Data in the Petrochemical Industry","volume":"11","author":"Lee","year":"2023","journal-title":"Processes"},{"key":"2026102306021959800_cit5","doi-asserted-by":"publisher","DOI":"10.1016\/j.conengprac.2020.104331","article-title":"Scalable reinforcement\nlearning for plant-wide control of vinyl acetate monomer process","volume":"97","author":"Zhu","year":"2020","journal-title":"Control Eng. Pract."},{"key":"2026102306021959800_cit6","doi-asserted-by":"publisher","first-page":"159","DOI":"10.1016\/0098-1354(95)00260-X","article-title":"Operating regime\nbased process modeling and identification","volume":"21","author":"Johansen","year":"1997","journal-title":"Comput.\nChem. Eng."},{"key":"2026102306021959800_cit7","doi-asserted-by":"publisher","DOI":"10.1002\/aic.16729","article-title":"Machine\nlearning-based predictive control of nonlinear processes. Part I:\nTheory","volume":"65","author":"Wu","year":"2019","journal-title":"AIChE J."},{"key":"2026102306021959800_cit8","doi-asserted-by":"publisher","DOI":"10.1002\/aic.16980","article-title":"Operable adaptive\nsparse identification of systems: Application to chemical processes","volume":"66","author":"Bhadriraju","year":"2020","journal-title":"AIChE J."},{"key":"2026102306021959800_cit9","doi-asserted-by":"publisher","DOI":"10.1016\/j.conengprac.2021.104956","article-title":"Hybrid Koopman\nmodel predictive control of nonlinear systems using multiple EDMD\nmodels: An application to a batch pulp digester with feed fluctuation","volume":"118","author":"Son","year":"2022","journal-title":"Control Eng. Pract."},{"key":"2026102306021959800_cit10","doi-asserted-by":"publisher","DOI":"10.1016\/j.compchemeng.2024.108789","article-title":"Machine\nlearning & conventional approaches to process control & optimization:\nIndustrial applications & perspectives","volume":"189","author":"Raven","year":"2024","journal-title":"Comput. Chem. Eng."},{"key":"2026102306021959800_cit11","doi-asserted-by":"publisher","first-page":"359","DOI":"10.1515\/revce-2024-0055","article-title":"A tutorial review of machine\nlearning-based\nmodel predictive control methods","volume":"41","author":"Wu","year":"2025","journal-title":"Rev. Chem.\nEng."},{"key":"2026102306021959800_cit12","doi-asserted-by":"publisher","DOI":"10.1016\/j.compchemeng.2022.107956","article-title":"A tutorial\nreview of neural network\nmodeling approaches for model predictive control","volume":"165","author":"Ren","year":"2022","journal-title":"Comput. Chem. Eng."},{"key":"2026102306021959800_cit13","doi-asserted-by":"publisher","first-page":"455","DOI":"10.1002\/aic.690420215","article-title":"Predictive control\nof quality in batch polymerization using hybrid ANN models","volume":"42","author":"Tsen","year":"1996","journal-title":"AIChE J."},{"key":"2026102306021959800_cit14","doi-asserted-by":"publisher","first-page":"2547","DOI":"10.1021\/acs.iecr.0c05208","article-title":"Model Predictive Control Embedding\na Parallel Hybrid Modeling Strategy","volume":"60","author":"Ghosh","year":"2021","journal-title":"Ind. Eng.\nChem. Res."},{"key":"2026102306021959800_cit15","doi-asserted-by":"publisher","DOI":"10.1002\/aic.18012","article-title":"Deep hybrid model-based predictive control with guarantees\non domain of applicability","volume":"69","author":"Bangi","year":"2023","journal-title":"AIChE J."},{"key":"2026102306021959800_cit16","doi-asserted-by":"publisher","first-page":"1153","DOI":"10.1016\/S0098-1354(03)00043-7","article-title":"Hybrid\nmodel-based approach to batch-to-batch\ncontrol of particle size distribution in emulsion polymerization","volume":"27","author":"Doyle","year":"2003","journal-title":"Comput. Chem. Eng."},{"key":"2026102306021959800_cit17","volume-title":"12th International Conference on Distillation\n& Absorption","author":"Krespach","year":"2022"},{"key":"2026102306021959800_cit18","doi-asserted-by":"publisher","DOI":"10.1002\/aic.17692","article-title":"Improving the load\nflexibility of industrial air separation units using a pressure-driven\ndigital twin","volume":"68","author":"Kender","year":"2022","journal-title":"AIChE J."},{"key":"2026102306021959800_cit19","volume-title":"Industrial\nGases Processing","author":"Ha\u0308ring","year":"2008"},{"key":"2026102306021959800_cit20","doi-asserted-by":"crossref","DOI":"10.1002\/9781119414674","volume-title":"Distillation: Principles\nand Practice","author":"Stichlmair","year":"2021","edition":"2"},{"key":"2026102306021959800_cit21","doi-asserted-by":"publisher","first-page":"255","DOI":"10.1016\/B978-0-12-386876-3.00006-5","volume-title":"Distillation: Operation\nand Applications","author":"Moll","year":"2014"},{"key":"2026102306021959800_cit22","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-662-10553-5","volume-title":"Tieftemperaturtechnik: Erzeugung\nsehr tiefer Temperaturen, Gasverflu\u0308ssigung und Zerlegung von\nGasgemischen","author":"Hausen","year":"1985","edition":"2"},{"key":"2026102306021959800_cit23","doi-asserted-by":"publisher","first-page":"171","DOI":"10.1016\/j.jprocont.2019.10.008","article-title":"Economic nonlinear\nmodel predictive control using hybrid mechanistic data-driven models\nfor optimal operation in real-time electricity markets: In-silico\napplication to air separation processes","volume":"84","author":"Scha\u0308fer","year":"2019","journal-title":"J.\nProcess Control"},{"key":"2026102306021959800_cit24","doi-asserted-by":"publisher","first-page":"1156","DOI":"10.1016\/j.jprocont.2014.03.010","article-title":"A tutorial\nreview of economic model\npredictive control methods","volume":"24","author":"Ellis","year":"2014","journal-title":"J. Process Control"},{"key":"2026102306021959800_cit25","doi-asserted-by":"publisher","first-page":"108368","DOI":"10.1016\/j.compchemeng.2023.108368","article-title":"Nonlinear\nhybrid control model for accurate steady-state predictions","volume":"178","author":"Krespach","year":"2023","journal-title":"Comput. Chem. Eng."},{"key":"2026102306021959800_cit26","unstructured":"Kingma, D. P.; Ba, J. L.\n          Adam: A Method\nfor Stochastic Optimization. 2015."},{"key":"2026102306021959800_cit27","doi-asserted-by":"publisher","first-page":"108953","DOI":"10.1016\/j.compchemeng.2024.108953","article-title":"Improving\nextrapolation capabilities of a data-driven prediction model for control\nof an air separation unit","volume":"194","author":"Krespach","year":"2025","journal-title":"Comput. Chem. Eng."},{"key":"2026102306021959800_cit28","doi-asserted-by":"publisher","DOI":"10.5281\/zenodo.579985","article-title":"TensorFlow: Large-scale machine learning on\nheterogeneous systems","author":"Abadi","year":"2015"},{"key":"2026102306021959800_cit29","unstructured":"Chollet, F.\n          \n          Keras. 2015."},{"key":"2026102306021959800_cit30","doi-asserted-by":"publisher","first-page":"108250","DOI":"10.1016\/j.compchemeng.2023.108250","article-title":"Reduced order\nmodeling of a pressure column of an air separation unit using the\nDynamic Edmister Method","volume":"174","author":"Kender","year":"2023","journal-title":"Comput. Chem. Eng."},{"key":"2026102306021959800_cit31","doi-asserted-by":"publisher","first-page":"98","DOI":"10.1016\/j.cherd.2019.04.031","article-title":"Pressure-driven\ndynamic simulation of start up and shutdown procedures\nof distillation columns in air separation units","volume":"147","author":"Kender","year":"2019","journal-title":"Chem. Eng. Res. Des."},{"key":"2026102306021959800_cit32","doi-asserted-by":"publisher","first-page":"107349","DOI":"10.1016\/j.compchemeng.2021.107349","article-title":"Development of a digital twin for\na flexible air separation unit\nusing a pressure-driven simulation approach","volume":"151","author":"Kender","year":"2021","journal-title":"Comput. Chem. Eng."}],"container-title":["ACS Omega"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/pubs.acs.org\/doi\/pdf\/10.1021\/acsomega.6c06082","content-type":"application\/pdf","content-version":"vor","intended-application":"unspecified"},{"URL":"https:\/\/pubs.acs.org\/acsodf\/article-pdf\/11\/33\/50281\/66535318\/acsomega.6c06082.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/pubs.acs.org\/acsodf\/article-pdf\/11\/33\/50281\/66535318\/acsomega.6c06082.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,10,23]],"date-time":"2026-10-23T10:09:08Z","timestamp":1792750148000},"score":1,"resource":{"primary":{"URL":"https:\/\/pubs.acs.org\/acsodf\/article\/11\/33\/50281\/5254804\/Data-Driven-Nonlinear-Model-Predictive-Control-for"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,13]]},"references-count":32,"journal-issue":{"issue":"33","published-online":{"date-parts":[[2026,8,13]]},"published-print":{"date-parts":[[2026,8,25]]}},"URL":"https:\/\/doi.org\/10.1021\/acsomega.6c06082","relation":{},"ISSN":["2470-1343"],"issn-type":[{"value":"2470-1343","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,13]]}}}