A low resistance circular diverter tee based on an improved random forest model.

Abstract

Local components are prevalent in building transmission and distribution systems, and their resistance can significantly increase a system's operating energy consumption. This paper takes a tee as an example and proposes a novel resistance reduction method for building transmission and distribution systems that utilizes an improved random forest model. Unlike existing studies on local component resistance reduction that rely on trial-and-error empirical methods, this study introduces a posterior optimization approach that can obtain a global optimal solution within a given range. The optimal tee shape is first predicted and then validated through experiments and numerical simulations to verify the resistance reduction effect. The results show that under different working conditions, the optimized tee achieves a resistance reduction rate of 28-66% in the main line and 16-93% in the branch line. Previous research on the resistance reduction mainly focused on rectangular components that can be reduced to two dimensions. This study proposes an a posteriori resistance reduction method for circular components, providing a reference for resistance reduction in building transmission and distribution systems.