中文摘要：

将太阳能光伏阵列中的每组太阳电池、逆变器及升压变构成一个连接单元，影响太阳能光伏阵列集电网络成本的因素包括：连接单元的数量、位置及电压等级、电气设备负载、拓扑接线形式、电缆截面、长度以及电缆沟的布置等。这些变量相互关联又相互制约，使得集电网络优化模型具有多离散变量和强非线性的特点。通过建立光伏阵列集电网络的成本模型，对其拓扑连线进行分层优化，采用聚类算法分区、扇面分割算法分串，采用Prime算法以及多旅行商问题求解不同拓扑的最优结构，最终得到光伏阵列中各离散点的最优路由方案，为光伏阵列集电网络的拓扑优化提供有益参考。

英文摘要：

Each group of solar cells, inverter and step-up transformer can be composed a connection unit, the cost factors of solar PV array collecting network include connection unit number, location and different voltage class, the load of electrical equipment, the topological connection, cable section, length and the layout of cable trench, etc. These variables relate to each other and restrict each other, which makes the collection of electric network optimization model with discrete variables and strong nonlinear characteristics. Through establish the cost of PV array collecting network model to layered optimize the model, using clustering algorithm partitions, sectorial segmentation algorithm split, Prime algorithm and multiple traveling salesman problem to solve the different topology optimal connections, finally get the optimal routing scheme of each discrete point in the PV arrays. The results of this work will contribute to the planning and optimization of the electrical collection network for solar PV array.