Machine-Learning-Based Soft Sensors for Energy Efficient Operation of Crude Distillation Units

Rožanec, Jože M.; Trajkova, Elena; Onat, Melike K.; Sarantinoudis, Nikolaos; Arampatzis, George; Fortuna, Blaž; Mladenić, Dunja

doi:10.1109/icecet55527.2022.9872983

Cited by 3 publications

(2 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…temperature, pressure and reboiler flow. More details on the used machine learning models can be found in [43], [44].…”

Section: Simulation and Resultsmentioning

confidence: 99%

Model-Based Simulation Framework for Digital Twins in the Process Industry

Sarantinoudis,

Tsinarakis,

Dedousis

et al. 2023

IEEE Access

Self Cite

View full text Add to dashboard Cite

A process modelling and simulation theoretical framework of general use for the study of continuous process industrial systems is introduced. The proposed process modelling methodology is based on Material Flow Networks and is implemented on a Process Simulation Modelling Tool developed for this purpose. The tool introduced can also serve the requirements arising for online use of the models as digital shadows of the physical systems, in the context of digital twinning the process industry. The implemented models in conjunction with tools from other scientific fields can be used for monitoring, root cause analysis, performance optimization, limitation and recovery of the behaviour of systems. An application example of the proposed methodology is provided and useful conclusions arise. Finally, extensions of the proposed method and potential challenges are discussed.

show abstract

“…temperature, pressure and reboiler flow. More details on the used machine learning models can be found in [43], [44].…”

Section: Simulation and Resultsmentioning

confidence: 99%

Model-Based Simulation Framework for Digital Twins in the Process Industry

Sarantinoudis,

Tsinarakis,

Dedousis

et al. 2023

IEEE Access

Self Cite

View full text Add to dashboard Cite

show abstract

“…A detailed pre-processing strategy ensures the robustness and integrity of the developed models. Moreover, the comprehensive evaluation using customized metrics underscores the models’ validity and applicability compared to prior research [ 51 , 52 , 53 ]. This study fills a significant gap in the literature regarding the deployment of soft sensors in oil refineries, as most previous studies were focused on other industrial applications.…”

Section: Discussionmentioning

confidence: 99%

Soft Sensing of LPG Processes Using Deep Learning

Sifakis,

Sarantinoudis,

Tsinarakis

et al. 2023

Sensors

Self Cite

View full text Add to dashboard Cite

This study investigates the integration of soft sensors and deep learning in the oil-refinery industry to improve monitoring efficiency and predictive accuracy in complex industrial processes, particularly de-ethanization and debutanization. Soft sensor models were developed to estimate critical variables such as the C2 and C5 contents in liquefied petroleum gas (LPG) after distillation and the energy consumption of distillation columns. The refinery’s LPG purification process relies on periodic sampling and laboratory analysis to maintain product specifications. The models were tested using data from actual refinery operations, addressing challenges such as scalability and handling dirty data. Two deep learning models, an artificial neural network (ANN) soft sensor model and an ensemble random forest regressor (RFR) model, were developed. This study emphasizes model interpretability and the potential for real-time updating or online learning. The study also proposes a comprehensive, iterative solution for predicting and optimizing component concentrations within a dual-column distillation system, highlighting its high applicability and potential for replication in similar industrial scenarios.

show abstract