几种遗传算法在舰船电力系统网络重构中应用及比较

Comparative of Network Reconfiguration of the Shipboard Power System Based on Some Genetic Algrithms

舰船电力系统(SPS)的网络重构是恢复系统故障、提高舰船生命力的重要途径之一。在满足各种运行约束条件下,快速地恢复因战斗受损的负荷供电是一典型的非线性整数规划问题。该文给出了大型舰船电力系统的网络拓扑结构,建立了综合考虑线路容量限制、节点电压约束及供电负荷优先性的重构数学模型,提出了全局遗传算法和免疫遗传算法来解决舰船电力系统网络重构问题。通过算例,对这几种遗传算法进行了比较,结果表明免疫遗传算法具有较高的收敛速度和精度,能够迅速求得全局最优解,避免了不成熟收敛,较好地实现了舰船电力系统的多目标故障恢复。

The electrical network reconfiguration of shipboard power system（SPS） is one of the most important methods to restore power supply and improve survivability. With all kinds of operation constraints, fast restoration of supply under batde damage is a typical nonlinear integer programming problem. The network topology of the large shipboard power system is given. And the mathematical models of the reconfiguration considering the capacity of the generators nodes voltage constraints and load priorities are also given which can grantee the maximun load power supply. Global GA and immune GA are provided. From the resuits of comparative analysis between immune GA and other GAs, immune GA can accelerate convergence, get global optimization quickly, avoid premature convergence. And the SPS multi-object service restoration is well realized.