Application of artificial intelligence CFD based on neural network in vapor-water two-phase flow

A new numerical calculation method has been developed based on the nonlinear analysis characteristics of artificial neural network (ANN). This method is a novel algorithm for computational fluid dynamics (CFD) using the concept of the artificial intelligence. It is completely different from prior traditional numerical methods. It does not generate mesh in the calculation domain except for some random points, even does not solve the algebraic equations. This method has been applied to compute the vapor-water two-phase flows in a tube with uniform and nonuniform heating. Through the comparison of the calculated results between this method and theoretical analysis (the finite difference method), the validity of this meshfree method based on the ANN is confirmed.