Preprint Article Version 1 Preserved in Portico This version is not peer-reviewed

Data Driven AI Models within a User-Defined Optimization Objective Function in Cement Production

Othonas Manis
,
Michalis Skoumperdis
* ,
Christos Kioroglou
,
Dimitrios Tzilopoulos
,
Miltos Ouzounis
,
Michalis Loufakis
,
Nikos Tsalikidis
,
Nikolaos Kolokas
ORCID logo ,
Panagiotis Georgakis
,
Ilias Panagoulis
,
Tsolkas Alexandros
,
Dimosthenis Ioannidis
ORCID logo ,
Dimitrios Tzovaras
,
Mile Stankovski
Version 1 : Received: 18 December 2023 / Approved: 18 December 2023 / Online: 18 December 2023 (11:51:00 CET)

A peer-reviewed article of this Preprint also exists.

Manis, O.; Skoumperdis, M.; Kioroglou, C.; Tzilopoulos, D.; Ouzounis, M.; Loufakis, M.; Tsalikidis, N.; Kolokas, N.; Georgakis, P.; Panagoulias, I.; Tsolkas, A.; Ioannidis, D.; Tzovaras, D.; Stankovski, M. Data-Driven AI Models within a User-Defined Optimization Objective Function in Cement Production. Sensors 2024, 24, 1225. Manis, O.; Skoumperdis, M.; Kioroglou, C.; Tzilopoulos, D.; Ouzounis, M.; Loufakis, M.; Tsalikidis, N.; Kolokas, N.; Georgakis, P.; Panagoulias, I.; Tsolkas, A.; Ioannidis, D.; Tzovaras, D.; Stankovski, M. Data-Driven AI Models within a User-Defined Optimization Objective Function in Cement Production. Sensors 2024, 24, 1225.

Abstract

The energy-intensive sector of the cement industry needs to find technologically advanced methods to produce cement with as little energy as possible, without polluting the environment and without sacrificing the quality (fineness) of the cement. In addition, the stress on the machines changes the behavior of the data. By combining the k-means method, two (2) feature selection methods, nine (9) machine learning methods, and the differential evolution (DE) method, a dynamic system is created where the user determines the key performance indicators (KPIs) each time. The user has the potential to control the distribution of the data, the input of machine learning models, and the ability to switch the system to the suggested values without damage. The selection of the best-performing models is based on the normalized root mean squared error (NRMSE). The DE method optimizes the dynamic objective function. The optimal values of the manipulated variables are evaluated compared to the value of the objective function from the actual values of the variables. In all experiments, the recommendations are acceptable. The final results show an improvement in the operation of both the cement mill and kiln compared to the current operating condition.

Keywords

Optimization; Feature selection; Machine Learning; Clustering; Differential Evolution; Cement Mill; Cement Kiln; Key performance indicator

Subject

Computer Science and Mathematics, Artificial Intelligence and Machine Learning

Copyright: This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Download PDF

Comments (0)

We encourage comments and feedback from a broad range of readers. See criteria for comments and our Diversity statement.

Leave a public comment
Send a private comment to the author(s)
* All users must log in before leaving a comment
Views 0
Downloads 0
Comments 0
Metrics 0
Get PDF Cite Share 0
Bookmark BibSonomy Mendeley Reddit Delicious
×
Alerts
Notify me about updates to this article or when a peer-reviewed version is published.
We use cookies on our website to ensure you get the best experience.
Read more about our cookies here.