Digital transformation of coal-water suspension production: algorithms and innovations
Keywords:
Coal-Water Fuel (CWF), Artificial Intelligence (AI), Vision Transformer (ViT), Deep Reinforcement Learning (DRL), Generative Adversarial Networks (GAN), Quantum Optimization, Colloidal Stability, Digital TwinAbstract
The integration of Artificial Intelligence (AI) into Coal-Water Fuel (CWF) production is revolutionizing the control of key parameters, such as coal concentration (0–30%), ensuring CWF stability and energy efficiency. Traditional CWF synthesis methods suffer from insufficient accuracy and prolonged adaptation times to raw material changes, limiting their industrial applicability. This article presents an advanced algorithm combining transformer neural networks, quantum optimization, and hybrid physicochemical models, specifically designed for the precise control of coal particle concentration and dispersity. We detail a multi-level AI control architecture, including a Vision Transformer (ViT) for particle morphology analysis, Deep Reinforcement Learning (DRL) for accurate blending, and Generative Adversarial Networks (GANs) for granulometry optimization. The implementation of this system allows for a 29% reduction in energy consumption and a 31% reduction in CWF production cost, demonstrating its high relevance for industrial deployment and the transition toward a carbon-neutral energy sector.
References
Chai, Y., &Li, Q. (2022). Artificial intelligence in optimization and control of coal-water slurry: A review. Fuel Processing Technology, 229, 107172.
Cheng J., Li Y., Zhou J., Liu J., CenK. (2010)Maximum concentrations of solids in coal-water slurries predicted using neural network models Fuel Processing Technology, 91(12), pp. 1832-1838.
Duan, Y., et al. (2023). Vision Transformer for micro-morphological analysis of coal particles and its application in grinding control. Powder Technology, 418, 118258.
Gao, W., & Zhang, H. (2021). Deep Reinforcement Learning for real-time rheology control of high-concentration coal-water slurries. Energy & Fuels, 35(12), 10103-10115.
Verwey, E. J. W., &Overbeek, J. T. G. (1948). Theory of the Stability of Lyophobic Colloids. Elsevier.
Yang, Z., et al. (2024). Generative Adversarial Networks for optimizing particle size distribution and enhancing the stability of coal-water slurry. Chemical Engineering Journal, 479, 147570.
Kim, J., & Lee, S. (2023). Digital Twin implementation for predictive maintenance and quality control in coal-water fuel production. Applied Energy, 345, 121175.
Wang, L., & Liu, G. (2022). Quantum optimization algorithms for complex colloidal systems: A perspective on DLVO calculations. Journal of Colloid and Interface Science, 607, 1870-1880.
