Modeling Integrated with Mechanism and Data-Driven for Hydrogen-Based Mineral Phase Transformation Process of Refractory Iron Ores

结合机理和数据驱动的建模方法用于难熔铁矿石氢基矿物相变过程研究

阅读:1

Abstract

This study proposes an integrated modeling approach that combines mechanistic and data-driven methods for the hydrogen-based mineral phase transformation (HMPT) process of refractory iron ores. First, a dynamic mechanistic model covering multistage preheaters, the heating furnace, and the restorer is developed, describing the heat transfer, mass transfer, and reaction kinetics in the redox asynchronous compartment system. Second, a parameter identification method combining fuzzy serial-parallel stochastic configuration networks (F-SPSCN) with a grid search is introduced to accurately estimate key unmeasurable parameters. Finally, under typical industrial operating conditions, simulation results show that the system achieves a final Fe(3)O(4) conversion rate of 99.01%, with the temperatures of the heating furnace and the restorer stabilizing at 848 and 863 °C, respectively, and temperature fluctuations of the preheaters remaining within ±2.5 °C. The control variable experiment and disturbance variable experiment further confirm that this mechanistic model is capable of effectively capturing the synergistic effects of multiple parameter changes.

特别声明

1、本页面内容包含部分的内容是基于公开信息的合理引用;引用内容仅为补充信息,不代表本站立场。

2、若认为本页面引用内容涉及侵权,请及时与本站联系,我们将第一时间处理。

3、其他媒体/个人如需使用本页面原创内容,需注明“来源:[生知库]”并获得授权;使用引用内容的,需自行联系原作者获得许可。

4、投稿及合作请联系:info@biocloudy.com。