基于Petri网和多种群遗传算法的海洋核动力平台电力系统网络重构

Network reconstruction of offshore nuclear power platform power system based on Petri net and multi-population genetic algorithm

针对现有海洋核动力平台电力系统网络重构方法中功率流分析复杂,且重构模型求解算法难以稳定收敛到全局最优解的问题,提出一种基于Petri网和多种群遗传算法的核动力平台电力系统网络重构方法。综合故障后负荷恢复量、开关操作代价和发电机运行效率指标建立系统网络重构目标函数;基于Petri网对系统进行拓扑建模,将储能装置等效为负值"负荷",通过动态更新机制确定功率流分布;采用多种群遗传算法求解网络重构问题,获取满足系统约束的最佳开关状态组合方案。典型核动力平台电力系统算例表明,所提方法能有效防止算法局部收敛以及减少迭代次数,快速提供完备的系统重构方案。

Aiming at the problem that the existing power flow analysis in network reconstruction method for offshore nuclear power platform power system is complex and the solving algorithm of the reconstruction model is hardly stably converged to the global optimal solution,a network reconstruction method based on Petri net and multi-population genetic algorithm is proposed for nuclear power platform power system.The indexes of load recovery amount after fault,switching operation cost,and generator operation efficiency are integrated to build the objective function of network reconstruction.The topological model of the system is built based on Petri net,in which,the energy storage device is equivalent to a negative load,and the power flow distribution is determined by dynamic update mechanism.Multi-population genetic algorithm is adopted to solve the network reconstruction problem,and the optimal switching state combination scheme satisfying system constraints is obtained.The example of a typical nuclear power platform power system shows that the proposed method can effectively prevent local convergence of the algorithm and reduce the iteration times,which provides complete reconstruction scheme quickly.