Neuroevolutive Control of Industrial Processes Through Mapping Elites

Simple item page
dc.contributor.authorLangarica Chavira, Saúl Alberto
dc.contributor.authorNuñez Retamal, Felipe Eduardo
dc.date.accessioned2022-05-18T14:04:50Z
dc.date.available2022-05-18T14:04:50Z
dc.date.issued2021
dc.description.abstractClassical model-based control techniques used in process control applications present a tradeoff between performance and computational load, especially when using complex nonlinear methods. Learning-based techniques that allow the controller to learn policies from data represent an appealing alternative with potential to reduce the computational burden of real-time optimization. This article presents an efficient learning-based neural controller, optimized using evolutionary algorithms, designed especially for maintaining diversity of individuals. The search of solutions is conducted in the parameter space of a population of deep neural networks, which are efficiently encoded with a novel compression algorithm. Evaluation against strong baselines demonstrates that the proposed controller achieves better performance in most of the chosen evaluation metrics. Results suggest that learning-based controllers are a promising option for next-generation process control in the context of Industry 4.0.
dc.fuente.origenIEEE
dc.identifier.doi10.1109/TII.2020.3019846
dc.identifier.eissn1941-0050
dc.identifier.issn1551-3203
dc.identifier.urihttps://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9178992
dc.identifier.urihttps://doi.org/10.1109/TII.2020.3019846
dc.identifier.urihttps://repositorio.uc.cl/handle/11534/64118
dc.information.autorucEscuela de ingeniería ; Langarica Chavira, Saúl Alberto ; S/I ; 222832
dc.information.autorucEscuela de ingeniería ; Nuñez Retamal, Felipe Eduardo ; S/I ; 131441
dc.issue.numero5
dc.language.isoen
dc.nota.accesoContenido parcial
dc.pagina.final3713
dc.pagina.inicio3703
dc.publisherIEEE
dc.revistaIEEE Transactions on Industrial Informatics
dc.rightsacceso restringido
dc.subjectProcess control
dc.subjectSociology
dc.subjectStatistics
dc.subjectOptimization
dc.subjectComputational modeling
dc.subjectApproximation algorithms
dc.subjectEvolutionary computation
dc.titleNeuroevolutive Control of Industrial Processes Through Mapping Eliteses_ES
dc.typeartículo
dc.volumen17
sipa.codpersvinculados222832
sipa.codpersvinculados131441
Files
Collections
Artículos de revistas