Transient multi-field catalytic mechanism of derived CO in deep energy extraction and optimization of synergistic conversion efficiency

Abstract

In deep coal fluidized energy extraction, complex geological conditions and nonlinear fluctuations heighten risks of gas explosions and CO diffusion. This study proposes a Transient Multifield-Catalytic Synergistic (TMFCS) technology utilizing micro-nano particles, integrating particle catalysis experiments with a coupled thermal-flow-solid-reaction model to establish rapid CO conversion. The dynamic diffusion mechanisms of internally excited particle clouds and their impacts on component conversion efficiency are systematically analyzed for in-situ carbon conversion and explosion-induced smoke-gas diffusion control. Experimental results demonstrate that under catalytic particle sizes of 50–75 μm and delayed release times of 12–22 ms, the TMFCS achieves over 70 % in-situ CO conversion. To enhance prediction accuracy in complex conditions, a hybrid model combining Convolutional Neural Network (CNN), Bidirectional Long Short-Term Memory Network (BiLSTM), and Attention mechanisms is developed. Integrated with the Grey Wolf Optimizer (GWO) algorithm, optimal catalytic particle parameters and spraying processes are inversely identified, elevating CO conversion rates beyond 79 %. The TMFCS framework enables rapid CO consumption during deep mining, offering theoretical and technical foundations for carbon resource recycling and safety control. This approach addresses challenges in waste carbon utilization and explosion prevention, advancing sustainable and secure energy extraction practices.