A NSGA-II-Based Layout Method for Cable Bundles With Branches Using Machine Learning

There are a large number of cables in complex electronic systems, and these cables are usually arranged in multiple cable bundles with branches. In the layout of the cable bundles, the total length and weight of all cables must be minimized, and the cables must be arranged in bundles as much as possible; in addition, the electromagnetic compatibility (EMC) performance of the system must be as good as possible. In response to this problem, this study proposes a layout method suitable for cable bundles with branches in complex electronic systems: first, an automatic layout method for cable bundles with branches is proposed to obtain the initial layout of the cable bundles; then, four different machine learning methods are used to establish prediction models with four outputs and eight input variables, and a fast weighted linear fusion method based on prior information and genetic algorithm is proposed to obtain a better performance fusion model; finally, a multi-objective optimization algorithm based on machine learning and non-dominated sorting genetic algorithm-II (NSGA-II) is proposed and used to optimize the layout parameters. The validation results show the effectiveness of the proposed method.