{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,3]],"date-time":"2026-09-03T12:13:43Z","timestamp":1788437623442,"version":"build-2803163510"},"reference-count":35,"publisher":"American Chemical Society (ACS)","issue":"34","license":[{"start":{"date-parts":[[2026,8,18]],"date-time":"2026-08-18T00:00:00Z","timestamp":1787011200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/100019305","name":"Manipal Academy of Higher Education","doi-asserted-by":"publisher","award":["0000860"],"award-info":[{"award-number":["0000860"]}],"id":[{"id":"10.13039\/100019305","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2026,9,1]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>Continuous Stirred Tank Reactors (CSTRs) are widely used in chemical process industries, where precise temperature control is crucial for product quality, operational effectiveness, and safety. However, the accuracy of strict first-principle models is limited by the extremely nonlinear thermal behavior brought on by Arrhenius-based kinetics, measurement noise, transport delay, and feed fluctuations. In order to overcome this, the current study uses multivariate pilot-plant time-series data to create a high-fidelity nonlinear autoregressive model with exogenous inputs (NARX). Twin Delayed Deep Deterministic Policy Gradient (TD3) and Soft Actor-Critic (SAC) are used in the NARX model as a reinforcement learning environment for closed-loop temperature regulation. Feed temperatures, feed flow rates, and hot-oil-bath actuation are among the various manipulated-variable combinations that are assessed. The suggested NARX\u2013RL framework effectively tracks set points, according to the results. Both SAC and TD3 exhibit acceptable closed-loop performance, according to a comparative analysis; however, TD3 consistently outperforms SAC in tracking indices, convergence speed, and control stability across the majority of manipulated-variable combinations. Overall, the proposed NARX-based reinforcement learning technique provides a feasible option for flexible, data-driven, and practically implementable temperature regulation in lab-scale CSTR systems.<\/jats:p>","DOI":"10.1021\/acsomega.6c04451","type":"journal-article","created":{"date-parts":[[2026,8,18]],"date-time":"2026-08-18T19:02:50Z","timestamp":1787079770000},"page":"51335-51359","source":"Crossref","is-referenced-by-count":0,"title":["Reinforcement Learning-Based\nTemperature Control of\nLab-Scale Continuous Stirred Tank Reactor"],"prefix":"10.1021","volume":"11","author":[{"given":"Jim","family":"George","sequence":"first","affiliation":[{"name":"Muthoot Institute of Technology and Science , , , ,","place":["Ernakulam, Kerala, India, 682038"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Anagha","family":"Ravikumar","sequence":"additional","affiliation":[{"name":"Muthoot Institute of Technology and Science , , , ,","place":["Ernakulam, Kerala, India, 682038"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Erin","family":"Joshy","sequence":"additional","affiliation":[{"name":"Muthoot Institute of Technology and Science , , , ,","place":["Ernakulam, Kerala, India, 682038"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Alin","family":"Cheriyan","sequence":"additional","affiliation":[{"name":"Muthoot Institute of Technology and Science , , , ,","place":["Ernakulam, Kerala, India, 682038"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1829-7362","authenticated-orcid":true,"given":"Thirunavukkarasu","family":"Indiran","sequence":"additional","affiliation":[{"name":"Manipal Institute of Technology, Manipal Academy of Higher Education , , ,","place":["Manipal, Karnataka, India, 576104"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"316","published-online":{"date-parts":[[2026,8,18]]},"reference":[{"key":"2026090307201915900_cit1","doi-asserted-by":"publisher","first-page":"106886","DOI":"10.1016\/j.compchemeng.2020.106886","article-title":"A review On reinforcement learning: Introduction and\napplications in industrial process control","volume":"139","author":"Nian","year":"2020","journal-title":"Comput. Chem. Eng."},{"issue":"1","key":"2026090307201915900_cit2","doi-asserted-by":"publisher","first-page":"484","DOI":"10.1109\/TII.2020.2972383","article-title":"Adaptive Critic Designs for Optimal Event-Driven Control of a CSTR\nSystem","volume":"17","author":"Yang","year":"2021","journal-title":"IEEE Trans. Ind. Inform."},{"key":"2026090307201915900_cit3","doi-asserted-by":"publisher","first-page":"94944","DOI":"10.1109\/ACCESS.2025.3574730","article-title":"Tracking Control\nof CSTRs Based on Improved OU Noise\nand the TD3 Algorithm","volume":"13","author":"Shi","year":"2025","journal-title":"In IEEE Access"},{"key":"2026090307201915900_cit4","doi-asserted-by":"publisher","first-page":"16632","DOI":"10.1109\/TASE.2025.3579230","article-title":"Dynamic Decentralized\nEvent-Triggered\nTracking Control of Continuous Stirred Tank Reactor Systems","volume":"22","author":"Yang","year":"2025","journal-title":"IEEE Trans. Autom. Sci. Eng."},{"key":"2026090307201915900_cit5","doi-asserted-by":"publisher","first-page":"107496","DOI":"10.1016\/j.compchemeng.2021.107496","article-title":"A novel\nimplicit hybrid machine learning\nmodel and its application for reinforcement learning","volume":"155","author":"Machalek","year":"2021","journal-title":"Comput. Chem. Eng"},{"issue":"9","key":"2026090307201915900_cit6","doi-asserted-by":"publisher","first-page":"4966","DOI":"10.1021\/acs.iecr.4c03233","article-title":"Calvin Tsay, and\nEhecatl Antonio del Ri\u0301o-Chanona Industrial Engineering Chemistry\nResearch","volume":"64","author":"Bloor","year":"2025","journal-title":"Ind. Eng. Chem. Res."},{"issue":"1","key":"2026090307201915900_cit7","doi-asserted-by":"publisher","first-page":"265","DOI":"10.30492\/ijcce.2024.2029225.6615","article-title":"Reinforcement Learning\nBased Adaptive PID Controller for a Continuous Stirred Tank Heater\nProcess","volume":"44","author":"Veerasamy","year":"2025","journal-title":"Iran. J. Chem. Chem. Eng."},{"key":"2026090307201915900_cit8","doi-asserted-by":"publisher","first-page":"135351","DOI":"10.1016\/j.cej.2022.135351","article-title":"A new CSTR method for scale inhibitor\nevaluation","volume":"437","author":"Dai","year":"2022","journal-title":"Chem. Eng. J."},{"key":"2026090307201915900_cit9","doi-asserted-by":"publisher","first-page":"40","DOI":"10.1016\/j.matpr.2022.11.037","article-title":"Experimental\nanalysis of Cascade CSTRs with step and pulse inputs","volume":"78","author":"Khatoon","year":"2023","journal-title":"Mater. Today: Proc."},{"issue":"8","key":"2026090307201915900_cit10","doi-asserted-by":"publisher","first-page":"4516","DOI":"10.1109\/TII.2018.2884214","article-title":"Neuro-Optimal Tracking\nControl for Continuous Stirred Tank Reactor With Input Constraints","volume":"15","author":"Zhou","year":"2019","journal-title":"IEEE Trans. Ind. Inform."},{"key":"2026090307201915900_cit11","doi-asserted-by":"publisher","first-page":"558","DOI":"10.1016\/j.psep.2024.11.059","article-title":"PPO-based resilient\ncontrol framework for safer operation of exothermic CSTR","volume":"193","author":"Yu","year":"2025","journal-title":"Process Saf. Environ. Prot."},{"issue":"1","key":"2026090307201915900_cit12","doi-asserted-by":"publisher","first-page":"2069","DOI":"10.1109\/TR.2024.3407090","article-title":"Reinforcement Learning-Boosted\nEvent-Triggered Reliability Control for Uncertain CSTR System With\nAsymmetric Constraints","volume":"74","author":"Liu","year":"2025","journal-title":"IEEE Trans. Reliab."},{"key":"2026090307201915900_cit13","doi-asserted-by":"publisher","first-page":"162611","DOI":"10.1109\/ACCESS.2025.3609753","article-title":"Robust Adaptive Neural\nNetwork Control for Nonlinear CSTRs With Prescribed\nControl Performance and Event-Triggered Inputs","volume":"13","author":"Liu","year":"2025","journal-title":"In IEEE Access."},{"issue":"2","key":"2026090307201915900_cit14","doi-asserted-by":"publisher","DOI":"10.1002\/aic.70141","article-title":"Neural network-based offset-free\nmodel predictive control\nfor nonlinear systems","volume":"72","author":"Hassanpour","year":"2026","journal-title":"AIChE J."},{"key":"2026090307201915900_cit15","doi-asserted-by":"publisher","first-page":"108643","DOI":"10.1016\/j.compchemeng.2024.108643","article-title":"Challenges\nin data-based reactor\nmodeling: A critical analysis of purely data-driven and hybrid models\nfor a CSTR case study","volume":"184","author":"Peterson","year":"2024","journal-title":"Comput. Chem. Eng."},{"issue":"15","key":"2026090307201915900_cit16","doi-asserted-by":"publisher","first-page":"420","DOI":"10.1016\/j.ifacol.2024.08.565","volume":"58","author":"Banerjee","year":"2024","journal-title":"FAC-Pap."},{"key":"2026090307201915900_cit17","doi-asserted-by":"publisher","first-page":"103","DOI":"10.1016\/j.jprocont.2022.12.004","article-title":"A multiple neural network and reinforcement\nlearning-based\nstrategy for process control","volume":"121","author":"Dutta","year":"2023","journal-title":"J. Process Control."},{"issue":"8","key":"2026090307201915900_cit18","doi-asserted-by":"publisher","first-page":"3643","DOI":"10.1109\/TNNLS.2020.3015869","article-title":"Deep Learning-Based\nModel Predictive Control for Continuous Stirred-Tank Reactor System","volume":"32","author":"Wang","year":"2021","journal-title":"IEEE Trans. Neural Netw. Learn. Syst."},{"key":"2026090307201915900_cit19","doi-asserted-by":"publisher","first-page":"29755","DOI":"10.1109\/ACCESS.2020.2972621","article-title":"A Neural Network-Based Adaptive Backstepping\nControl Law With Covariance Resetting for Asymptotic Output Tracking\nof a CSTR Plant","volume":"8","author":"Alshammari","year":"2020","journal-title":"In IEEE Access."},{"key":"2026090307201915900_cit20","doi-asserted-by":"publisher","first-page":"170268","DOI":"10.1109\/ACCESS.2025.3615715","article-title":"Iterative\nLearning Observer-Based Predefined-Time Sliding Mode Control for Continuous\nStirred Tank Reactor With Input and Output Constraints","volume":"13","author":"Zhang","year":"2025","journal-title":"In IEEE Access"},{"key":"2026090307201915900_cit21","doi-asserted-by":"publisher","first-page":"103609","DOI":"10.1016\/j.jprocont.2025.103609","article-title":"Prescribed performance\nbased direct data-driven model\npredictive control for continuous stirred tank reactor","volume":"157","author":"Zhou","year":"2026","journal-title":"J. Process Control."},{"key":"2026090307201915900_cit22","doi-asserted-by":"publisher","first-page":"200053","DOI":"10.1016\/j.sasc.2023.200053","article-title":"Optimal\nmetaheuristic state-dependent parameter proportional-integral-plus\ncontrol: Alternative to gain-scheduled controller for control of a\nnonlinear continuous stirred tank reactor","volume":"5","author":"Talaei","year":"2023","journal-title":"Syst.\nSoft Comput."},{"key":"2026090307201915900_cit23","doi-asserted-by":"publisher","first-page":"117164","DOI":"10.1016\/j.ces.2021.117164","article-title":"Heikki Haario,Determination\nof kinetic\nconstants by using transient temperature data from continuous stirred\ntank reactors","volume":"248","author":"Salmi","year":"2022","journal-title":"Chem. Eng. Sci."},{"key":"2026090307201915900_cit24","doi-asserted-by":"publisher","first-page":"119087","DOI":"10.1016\/j.ces.2023.119087","article-title":"Optimal control of switched nonlinear systems with\napplication to\nchemical processes","volume":"281","author":"Fan","year":"2023","journal-title":"Chem. Eng. Sci."},{"key":"2026090307201915900_cit25","doi-asserted-by":"publisher","first-page":"299","DOI":"10.1016\/j.sajce.2024.09.005","article-title":"System identification\nof a nonlinear\ncontinuously stirred tank reactor using fractional neural network","volume":"50","author":"Kumar","year":"2024","journal-title":"S. Afr. J. Chem. Eng."},{"key":"2026090307201915900_cit26","doi-asserted-by":"publisher","first-page":"101361","DOI":"10.1016\/j.jestch.2023.101361","article-title":"A hybrid sliding\nmode control based on a nonlinear PID surface for nonlinear chemical\nprocesses","volume":"40","author":"Va\u0301squez","year":"2023","journal-title":"Eng. Sci. Technol. Int. J."},{"issue":"9","key":"2026090307201915900_cit27","doi-asserted-by":"publisher","DOI":"10.1002\/aic.18906","article-title":"Model-based\nfault diagnosis in closed-loop\nsafety-critical chemical reactors: An experimental study","volume":"71","author":"Du","year":"2025","journal-title":"AIChE J."},{"key":"2026090307201915900_cit28","doi-asserted-by":"publisher","first-page":"84554","DOI":"10.1109\/ACCESS.2023.3303036","article-title":"Reactor Temperature Compound Control Strategy Based on Disturbance\nObserver","volume":"11","author":"Tang","year":"2023","journal-title":"In IEEE Access."},{"key":"2026090307201915900_cit29","doi-asserted-by":"publisher","first-page":"30864","DOI":"10.1021\/acsomega.5c03219","article-title":"Reinforcement\nlearning-based nonlinear model predictive controller for a jacketed\nreactor: A machine learning concept validation using Jetson Orin","volume":"10","author":"Selvamurugan","year":"2025","journal-title":"ACS Omega"},{"key":"2026090307201915900_cit30","doi-asserted-by":"publisher","first-page":"24106","DOI":"10.1021\/acs.iecr.5c02895","article-title":"Lyapunov-constrained\nhybrid modeling for stable parameter\nlearning in nonlinear systems","volume":"64","author":"Shah","year":"2025","journal-title":"Ind. Eng. Chem.\nRes."},{"key":"2026090307201915900_cit31","doi-asserted-by":"publisher","first-page":"97783","DOI":"10.1109\/ACCESS.2020.2997249","article-title":"Gain-Scheduled Mean-Field-Type Control for a Non-Linear\nContinuous Stirred Tank Reactor","volume":"8","author":"Barreiro-Gomez","year":"2020","journal-title":"In IEEE Access."},{"key":"2026090307201915900_cit32","doi-asserted-by":"publisher","first-page":"108272","DOI":"10.1016\/j.compchemeng.2023.108272","article-title":"Polynomial NARX-based\nnonlinear model predictive control\nof modular chemical systems","volume":"177","author":"Nikolakopoulou","year":"2023","journal-title":"Comput. Chem. Eng."},{"issue":"2","key":"2026090307201915900_cit33","doi-asserted-by":"publisher","first-page":"210","DOI":"10.1109\/TSM.2025.3539223","article-title":"\u2018Process Control\nin Semiconductor Manufacturing Based on Deep Distributional Soft Actor\u2013Critic\nReinforcement Learning","volume":"38","author":"Liu","year":"2025","journal-title":"IEEE T. Semicond. Manuf."},{"issue":"2","key":"2026090307201915900_cit34","doi-asserted-by":"publisher","first-page":"283","DOI":"10.1109\/JAS.2024.124227","article-title":"Reinforcement Learning\nin Process Industries: Review and Perspective","volume":"11","author":"Dogru","year":"2024","journal-title":"IEEE CAA J. Autom. Sin."},{"issue":"2","key":"2026090307201915900_cit35","doi-asserted-by":"publisher","first-page":"540","DOI":"10.35833\/MPCE.2024.000471","article-title":"Optimal Power Dispatch\nof Active Distribution Network and P2P Energy\nTrading Based on Soft Actor-Critic Algorithm Incorporating Distributed\nTrading Control","volume":"13","author":"Zhang","year":"2024","journal-title":"J. Mod. Power Syst. Clean Energy"}],"container-title":["ACS Omega"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/pubs.acs.org\/doi\/pdf\/10.1021\/acsomega.6c04451","content-type":"application\/pdf","content-version":"vor","intended-application":"unspecified"},{"URL":"https:\/\/pubs.acs.org\/acsodf\/article-pdf\/11\/34\/51335\/66854438\/acsomega.6c04451.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/pubs.acs.org\/acsodf\/article-pdf\/11\/34\/51335\/66854438\/acsomega.6c04451.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,3]],"date-time":"2026-09-03T11:20:33Z","timestamp":1788434433000},"score":1,"resource":{"primary":{"URL":"https:\/\/pubs.acs.org\/acsodf\/article\/11\/34\/51335\/5275750\/Reinforcement-Learning-Based-Temperature-Control"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,18]]},"references-count":35,"journal-issue":{"issue":"34","published-online":{"date-parts":[[2026,8,18]]},"published-print":{"date-parts":[[2026,9,1]]}},"URL":"https:\/\/doi.org\/10.1021\/acsomega.6c04451","relation":{},"ISSN":["2470-1343"],"issn-type":[{"value":"2470-1343","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,18]]}}}