Research on mixing efficiency model of static mixer based on machine learning and flow field simulation

To address the low efficiency associated with traditional optimization design methods, machine learning was introduced into the research on structural optimization of static mixers in this paper. Flow field data of the static mixer was acquired using the Computational Fluid Dynamics (CFD) method. A quantitative evaluation mechanism was developed, taking into account the physical properties of mixing efficiency. Subsequently, the mixing efficiency was accurately assessed using the simulation results of the flow field. Correlation data samples between mixing efficiency and structural parameters were collected. A neural network was employed to establish a correlation between the key structural parameters of the mixer and the overall mixing efficiency.