| About this Abstract |
| Meeting |
2025 TMS Annual Meeting & Exhibition
|
| Symposium
|
Alumina and Bauxite
|
| Presentation Title |
Predicting Particle Size of Aluminum Hydroxide in Alumina Production using Machine Learning |
| Author(s) |
long duan, Yanfang Zhang, Shuai Shao, Qiaoyun Liu |
| On-Site Speaker (Planned) |
long duan |
| Abstract Scope |
The Bayer process for alumina production is complex and lengthy. Accurately and timely predicting process indicators and mining important parameters is one of the key to optimize this process. This study started with the precipitation process during alumina production, achieving accurate and timely prediction of particle size of aluminum hydroxide by using machine learning. The research process included collecting the equipment and material data, pre-processing data, constructing time-series dataset, selecting effective parameters, and establishing particle size prediction model. The determination coefficients of the prediction model was greater than 0.8. The results showed that based on machine learning and mass production process data to establishing prediction model can accurately and timely predict the particle size of aluminum hydroxide in the Bayer process. It provides a theoretical basis for predicting the process indicators of the entire alumina production process and lays the foundation for the intelligent of alumina production. |
| Proceedings Inclusion? |
Planned: Light Metals |
| Keywords |
Machine Learning, |