Analysis of the EMHD nanofluid flow for geothermal pipelines using physics-driven deep learning
摘要
In recent years data-driven machine learning techniques attract the attention of researchers in analyzing many complex systems. This study introduces a novel unsupervised deep neural network approach to predict the temperature and velocity behaviour of electro-magneto-hydrodynamics hybrid nanofluid flow for geothermal pipelines application.The exceptional flow and thermal characteristics of hybrid nanofluids making them ideal for use in geothermal energy extraction applications. The dynamics of hybrid nanofluid flow through a pipe are examined using a third-grade sodium alginate model, which has a lot of potential for geothermal applications. The copper oxide (CuO) and zinc oxide (ZnO) nanoparticles make up the nanofluid. It is also investigated how the flow dynamics are affected by electric and magnetic fields. The energy equation takes into account the effects of Joule heating and viscous dissipation as the fully developed incompressible fluid passes through the pipe. Consequently, an unsupervised deep neural network (DNN) method is used to predict the dynamics of nonlinear differential equations (DEs). The accuracy of the deep neural network ranges from