Three-day and Twenty-day Compressive Strength Predictions of Cement Clinker Based on PSO-ELM

• Published in: 2024 39th Youth Academic Annual Conference of Chinese Association of Automation (YAC) pp. 834 - 838
• Main Authors: Qi, Guangwei, Yu, Hongliang, Lu, Shizeng, Zhang, Rongfeng, Dong, Huijun, Zhang, Chongshan
• Format: Conference Proceeding
• Language: English
• Published: IEEE 07.06.2024
• Subjects: 28-day compressive strength; 3-day compressive strength; Accuracy; Automation; cement clinker; Cement industry; Predictive models; Process control; Production; PSO-ELM; Statistical analysis
• ISSN: 2837-8601
• DOI: 10.1109/YAC63405.2024.10598439

Cement clinker has a lot of control indexes to pay attention to in the normal production process, among which, cement clinker strength is one of the important indexes. Cement clinker strength is usually divided into two aspects: flexural strength and compressive strength, of which 3-day and 28-day compressive strength are the most concerned indicators for cement plants. At present, the traditional methods of cement clinker strength prediction are mostly statistical analysis, multiple regression, etc. These traditional methods of predicting cement clinker strength have the problems of low precision, large error, and insufficiently comprehensive consideration of variables. In order to solve these problems, this paper proposes a PSO-ELM-based prediction model that can predict the 3-day and 28-day compressive strength of cement clinker more accurately, thus improving the accuracy of the prediction. At the same time, the model enables the operator to know the strength of the clinker of this batch in advance, which shortens the time of clinker quality determination, and plays a role in the improvement of cement clinker quality.