对于配用电光通信网中无源光网络多层级、多节点网络规划问题, 首次提出采用基本遗传算法实现多层级的循环继承的层叠遗传算法。其采用外层分光网络继承内层网络基因用于本层网络的适应度函数选择, 该算法解决了多层级无源光(PON)网络各层级分光优化的问题, 并且通过仿真证明了其可行性, 实现了四层级网络星形结构拓扑优化。优化后的网络获得了最优的星形网络拓扑和通信建设成本的降低。

According to the planning problem of multi-layer level and multi node passive optical network (PON) in distribution power supply optical communication network, multi-layer level cycle inheritance stacked genetic algorithm based on basic genetic algorithm is proposed for the first time. The outer layer distribution optical network is used to inherit inner layer network gene to choose adaptation function in the layer network. The layer level optical distribution optimization problem of multi-layer level PON is resolved by the algorithm. Simulation results show that the algorithm is feasible and the topology optimization of four-layer level network with star structure is realized. The optimized network obtains the best star structure network topology and the communication construction cost is reduced.