Data-driven modelling of complex reactive flows

复杂反应流在许多化学、物理和生物过程中占主导地位，应该在线优化操作效率和稳定性，但由于缺乏可靠的模型技术，它受到阻碍。该项目通过将连续/离散尺度流体-颗粒动力学与系统/控制理论相结合，在实验室/工厂实验的支持下，开发了下一代数据驱动建模方法，解决了这一挑战。在在线数据的驱动下，通用方法为可靠地描述反应堆内部状态和在线预测运行异常开辟了一条强有力的途径。研究结果可以帮助一系列行业向智能制造和设计转型，这对澳大利亚的技术未来至关重要。

Complex reactive flow is dominant in many chemicals, physical and biological processes and should be optimised online for operational efficiency and stability, yet it is hindered by the lack of reliable model techniques. The project tackles this challenge by developing a next-generation data-driven modelling approach via integrating continuum/discrete-scale fluid-particle dynamics with system/control theories, supported by lab/plant experiments. Driven by online data, the generic approach can open up a powerful way to reliably describe the inner state of reactors and online predict operation anomalies. The outcomes can help transform a range of industries to smart manufacturing and design, which is vital to Australia's technological future.