摘要：In this paper, an ordered-specimen cluster analysis based network optimization is proposed. The holo-distributed automotive electrical system features that all electrical devices are the network nodes and connected with two separate networks, namely the digital signal network and the power supply network. The digital signal network features a set of hierarchy local subnets. The proposed optimization method for networking the widely scattered and utterly disordered body electrical devices, aims to a connection with the simplest harness. Through electrical device reduced-dimension ordering, the optimization problem is converted to an ordered-specimen with limited-group-capacity partition problem with the restrain of network schedulable condition. Problem’s solver algorithm is also proposed based on global search in shrank space. Finally, the method is used during the development of distributed body electrical system on prototype urban bus, and proved to be feasible.
关键词：Holo-distributed control system;automotive body electrical system;ordered-specimen cluster analysis;networking optimization;global search
