中文摘要：

近年来随着大量非线性负载的投入使用，船舶电网谐波问题日益严重，直接影响到船舶电气设备的正常工作和使用寿命。论文结合船舶电力系统动态负荷多等特点采用节点电势法进行基波潮流计算，忽略传输线路压降，对非线性负荷进行迭代求解，因而收敛速度快，精度高。在迭代求解过程中，采用改进的自适应遗传算法。根据适应度值的集中程度自适应地变化整个种群的交叉概率和变异概率，增加了种群多样性，扩大了搜索空间。应用改进的自适应遗传算法对一个12节点船舶电力系统进行滤波器优化配置，优化结果表明了算法的有效性。

英文摘要：

With a large number of non-linear load put into use, harmonics problem of the ship power system is becoming more and more serious, and it affects the normal work and useful life of the ship＇s electrical equipment. According to ship power system＇s characteristics such as installing with much dynamic load, node voltage method is used for the fundamental power flow calculation. Ignoring the transmission line＇s voltage and using the iterative method for the calculation of non-linear load, the fundamental power flow calculation can get fast convergence and high-precision. Improved adaptive genetic algorithm is used in the process of interation. The crossover probability and mutation probability are automatically changed according to the concentration of fitness value. The method increases the population diversity and expands the search space, so it can get a better result. The improved adaptive genetic algorithm is used to a 12-bus ship power system forfilters＇optimal allocation, and the results show the effectiveness of the algorithm.