考虑地理信息的多电源配网规划

Multi-source distribution network planning with consideration of geographic information

针对城市新建供电区的配网网架规划问题,提出了一种考虑街道地理环境约束的多电源配网规划新模型。其中,在计算投资费用时,明确区分了电气走廊的土建与电气费用。在进行电网布线时,允许负荷任意选择电源和线路走廊,允许任意产生新的电气分支节点。模型在投资费用的计算和线路走廊的选择方面更加符合实际,但也因为新电气节点(支路和线路)的产生而更加复杂。采用遗传算法来求解上述规划问题。其中,设计了多个电网信息矩阵来记录网架的地理与电气信息,使遗传个体的电网信息查询、记录与修改更加方便。采用了边重组交叉、支路最短路径替换和末端负荷转移策略来实施遗传操作,使配网规划问题的有效求解得到了保证。以一个实际的城市电网规划系统为例,仿真分析了论文模型和算法的有效性。

To solve the planning of distribution networks in new urban service area, a new model for muhi-source planning of distribution networks was proposed with consideration of street geographic environmental constraints. In the objective function, the civil and electrical costs of the network investment are divided in calculation. In the constraint conditions, any power source and any power cable corridor besides the street can be selected randomly for any supply demand, and the new electric branch node on the feeder can also be produced randomly. The planning problem was solved by GA in this paper. In GA, the network information matrices were designed to stand for the geographical and electrical information of the network. Strategies such as edge recombination crossover, shortest path substitution and load node transferring were adopted to implement genetic operations. The effectiveness of the proposed model and algorithm was illustrated by an actual urban medium-voltage distribution network planning.