[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-123143-en":3,"doc-seo-123143-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":4,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},123143,7971461740886,"Theodore","https://ap-avatar.wpscdn.com/davatar_3d24733baf745e90a7e4bdd5f77d97b2",8,"Research & Report","Machine learning multi-step-ahead modelling with uncertainty assessment - 2024","This study presents a multi-step-ahead identification strategy for robust machine learning (ML) models, contrasting standard single-step prediction with multi-step forecasting needs in model predictive control and optimization frameworks. The work investigates how a multi-step-ahead approach reduces prediction uncertainty relative to traditional single-step methods, and it benchmarks multiple model architectures, including recursive neural networks. Analysis applies the models to a polymerization reactor benchmark. Multi-step recursive models significantly cut uncertainty, especially when feedback is incorporated, supporting improved control and optimization performance.","[Available online at www.sciencedirect.com](Available online at www.sciencedirect.com)  \nScienceDirect  \nIFAC PapersOnLine 58-14 (2024) 25–30  \nMachine learning multi-step-ahead modelling with uncertainty assessment  \nErbet Almeida Costa ∗ Carine Menezes Rebello ∗  \nVin´ıcius Viena Santana ∗ Idelfonso B. R. Nogueira ∗∗ Department of Chemical Engineering, Norwegian University of Science and Technology, Gløshaugen, Trondeim, Norway (e-mail:  \n[erbet.a.costa@ntnu.no](erbet.a.costa@ntnu.no), [carine.m.rebello@ntnu.no](carine.m.rebello@ntnu.no), [vinicius.v.santana@ntnu.no](vinicius.v.santana@ntnu.no), idelfonso.b.d.r.nogueira@ntnu.no).  \nAbstract: This study presents a strategy for multi-step-ahead identification of robust machine learning (ML) . Hence, we focus on the disparity between standard single-step prediction models and the requirement for multi-step forecasting, which is crucial for Model Predictive and Optimization schemes. This work explores how the proposed muti-step-ahead strategy can diminish the prediction uncertainty compared to the traditional single-step approach. The paper evaluates the multi-step identification with uncertainty assessment in different model architectures, including those based on recursive neural networks. A key aspect of the analysis is the application of these models to a polymerization reactor, a standard benchmark in algorithm evaluation. The results reveal that multi-step recursive models significantly reduce prediction uncertainty compared to single-step models, particularly when feedback mechanisms are involved. This study highlights the advantages of multi-step models and their potential benefits for control and optimization schemes.  \nCopyright © 2024 The Authors. This is an open access article under the CC BY-NC-ND license ([https://creativecommons.org/licenses/by-nc-nd/4.0/](https://creativecommons.org/licenses/by-nc-nd/4.0/))  \nKeywords: Machine Learning Assisted Modeling; Estimation and Robust Estimation; Model Predictive Control.  \n1. INTRODUCTION  \nDynamic modeling of chemical processes is usually related to control and optimization schemes. Hence, it is an important topic for the proper and efficient operation of these processes. On the other hand, using rigorous models for such scenarios is computationally expensive, raising the opportunity to apply surrogate models.  \nHowever, applying surrogate models in such schemes can lead to several limitations, as these models usually have good accuracy for predicting but not in simulation scenarios, where multi-step-ahead predictions are necessary. Due to these limitations, it is usual to find works in the literature that employ machine learning-based models in control and optimization schemes using a single-step approach, like Shin et al. (2020); Entezari et al. (2023); Renet al. (2022); Norouzi et al. (2023); Afram et al. (2017); Wang et al. (2022) .  \nFurthermore, the efficacy of these models in dynamic prediction — in terms of accuracy, robustness, and efficiency — is intrinsically linked to the model’s prediction uncertainty. In environments with stringent operational constraints, high uncertainty can compromise the prediction feasibility and, therefore, the model application. Conversely, using a model with excessive uncertainty may result in overly conservative actions in more lenient settings, hindering the system’s performance and objective attainment (Costa et al., 2024, 2023) .  \n⋆ This paper has been sponsored by the Norwegian Research Council.  \nAddressing this gap, Park et al. (2023) introduces an innovative network architecture that facilitates multi-step forecasting, enabling the model to predict over a horizon matching the Model predictive control (MPC) prediction horizon in a single iteration. This approach, merging anonlinear autoregressive exogenous model (NARX) data architecture with recurrent networks, presents numerous benefits for dynamic prediction.  \nOur study proposed a systematic strategy for multi-stepahead ","cbCaipn3qo7GHdu2","https://ap.wps.com/l/cbCaipn3qo7GHdu2","pdf",1218712,1,6,"English","en",105,"# Introduction\n# Methodology\n## Prediction window and simulation setup\n# Case study and results\n# Conclusions and implications","[{\"question\":\"Why is multi-step-ahead forecasting important in model predictive control and optimization?\",\"answer\":\"Multi-step forecasting matches the prediction horizon required by MPC and optimization schemes, whereas single-step models often only provide good accuracy for direct prediction rather than simulation scenarios.\"},{\"question\":\"How does the proposed strategy quantify prediction uncertainty?\",\"answer\":\"The study uses Bayesian inference and addresses the inference challenge with Markov chain Monte Carlo simulations.\"},{\"question\":\"What do the results show for recursive multi-step models versus single-step models?\",\"answer\":\"Multi-step recursive models significantly reduce prediction uncertainty compared with single-step models, with the largest improvements when feedback mechanisms are involved.\"}]","Machine learning multi-step-ahead modelling with uncertainty assessment - 2024 | PDF",1785814867,15,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"machine-learning-multi-step-ahead-modelling-with-uncertainty-assessment-2024","",{"@graph":36,"@context":85},[37,54,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/machine-learning-multi-step-ahead-modelling-with-uncertainty-assessment-2024/123143/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-04",true,{"@type":65,"interactionType":66,"userInteractionCount":4},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"Why is multi-step-ahead forecasting important in model predictive control and optimization?","Question",{"text":75,"@type":76},"Multi-step forecasting matches the prediction horizon required by MPC and optimization schemes, whereas single-step models often only provide good accuracy for direct prediction rather than simulation scenarios.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does the proposed strategy quantify prediction uncertainty?",{"text":80,"@type":76},"The study uses Bayesian inference and addresses the inference challenge with Markov chain Monte Carlo simulations.",{"name":82,"@type":73,"acceptedAnswer":83},"What do the results show for recursive multi-step models versus single-step models?",{"text":84,"@type":76},"Multi-step recursive models significantly reduce prediction uncertainty compared with single-step models, with the largest improvements when feedback mechanisms are involved.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,114,119,122,127,130,134],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":21,"doc_module":4,"doc_module_name":46,"category_name":111,"show_sort_weight":112,"slug":113},"Technology",50,"technology",{"id":115,"doc_module":4,"doc_module_name":46,"category_name":116,"show_sort_weight":117,"slug":118},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":120,"slug":121},30,"research-report",{"id":123,"doc_module":4,"doc_module_name":46,"category_name":124,"show_sort_weight":125,"slug":126},9,"Religion & Spirituality",20,"religion-spirituality",{"id":125,"doc_module":4,"doc_module_name":46,"category_name":128,"show_sort_weight":125,"slug":129},"World Cup","world-cup",{"id":131,"doc_module":4,"doc_module_name":46,"category_name":132,"show_sort_weight":131,"slug":133},10,"Lifestyle","lifestyle",{"id":135,"doc_module":4,"doc_module_name":46,"category_name":136,"show_sort_weight":106,"slug":137},19,"General","general"]