Shape optimization of cut-off in a multi-blade fan/scroll system using neural network

In order to improve efficiency of a system with three-dimensional flow characteristics, this paper presents a new method that overcomes the computational difficulties associated with three-dimensional effects by using two-dimensional CFD and a neural network. The method was applied to the shape optimization of cut-off in a multi-blade fan/scroll system. As for the entrance conditions of two-dimensional CFD analysis, the experimental values at the positions apart from the inactive zone were used. The distributions of velocity and pressure obtained by two-dimensional CFD analysis were compared with those of three-dimensional CFD analysis and experimental results. It was found that the distributions of velocity and pressure have qualitative similarity. The results of two-dimensional CFD analysis were used for learning as target values of a neural network. The optimal angle and radius of cut-off were determined as 71° and 0.092 times the outer diameter of impeller, respectively. It was concretized in a previous report that the optimal angle and radius of cut-off are approximately 72° and 0.08 times the outer diameter of impeller, respectively.