Effects of inner iteration times on the performance of IDEAL algorithm

An efficient segregated algorithm for non-linear incompressible fluid flow and heat transfer problems, called IDEAL (Inner Doubly-Iterative Efficient Algorithm for Linked-Equations) for short, was proposed in reference [9]. Subsequently, it was extended to the 3D staggered/collocated grid systems. IDEAL includes inner doubly-iterative processes for solving pressure equations at each iteration level, and it could adjust the inner iteration times to control the convergence rate and the stability of iteration process, which is greatly different from other segregated algorithms. The objective of this paper is to analyze the effects of inner iteration times on the performance of IDEAL by four incompressible fluid flow problems, two of which belong to open systems, and the others refer to closed systems. It is found that: (1) the robustness of IDEAL is enhanced greatly with the increase of inner iteration times; (2) at the same time step multiple, the outer iteration number decreases with the increase of inner iteration times and the computation time is not related to the inner iteration times; (3) at the optimal time step multiple, the large inner iteration times of 4&4 and 7&7 could reduce the outer iteration number by about 70% and the computation time by about 40% over the small inner iteration times of 1&1.