Optimization investigation on the distribution of both inlets and outlets for the shell of a novel separated structure steam generator

Abstract

The distribution of both inlets (downcomers for the steam generator) and outlets (risers for the steam generator) for the shell has an important influence on the overall thermal hydraulic performance of a novel separated structure steam generator (SS-SG). Nevertheless, there is no basis for reference in the design process so far. In this article the effect of both the inlets and outlets distribution on the vapor-liquid two-phase flow and heat transfer characteristics was studied and optimized with the help of experiment, numerical simulation and machine learning algorithm. The surrogate model was constructed by the backpropagation neural network based on the genetic algorithm (GA-BPNN). The average boiling heat transfer coefficient (HTC) was selected as the objective function to characterize the rationality of the flow field in the shell. To get the optimal distribution, the global optimization was performed by the genetic algorithm (GA). For the optimal structure obtained, the longitudinal flow was enhanced greatly compared to the original structure. This is beneficial to avoid the local flow stagnation and eliminate the enrichment of ion concentration for long-term operation. Meanwhile, the vapor accumulation at the top of the shell near the front tube-sheet has also been alleviated to some extent. It is beneficial to alleviate the thermal damage of the high heat flux due to heat transfer deterioration. Besides, the average pressure drop was only increased by 0.96%. The results of this investigation are of significance as a reference in the actual design of the novel SS-SG.