基于帕累托最优的配电网多目标规划

Multi-objective Planning of Distribution Network Based on Pareto Optimality

提出了一种能够反映配电网结构合理程度的可靠性指标——网络风险指数,并建立了以配电网投资费用、网络损耗和网络风险指数为目标函数,综合考虑经济性和可靠性的配电网规划模型。设计了一种基于节点关联矩阵的网络修复算法,能够对随机生成的网络中的孤岛、孤链和闭环进行检测和修复,使网络满足配电网辐射状要求。对强度帕累托进化算法进行了改进,给出了新的适应度函数,以提高算法的搜索速度和搜索能力,并利用逼近理想解排序法(TOPSIS)对帕累托前沿中的个体进行排序,筛选出最优方案。最后,对一个54节点配电网进行规划,结果表明网络风险指数与配电网的停电损失具有极强的线性相关性,其可以作为表征配电网可靠性的指标;与传统强度帕累托进化算法相比,所提改进算法具有较强的搜索能力和较快的搜索速度。

A reliability indicator called the network risk index capable of reflecting the rationality of the distribution network structure is proposed.By taking the investment cost,network loss and network risk index as the objective functions,a distribution network planning model is developed with an overall consideration of the economy and reliability of the distribution network.A network repair algorithm based on bus incidence matrix is designed to find and repair the randomly generated isolated busses,isolated chains and closed loops for the requirement of grid radiation.New fitness functions are proposed to improve the search speed and capability of the strength Pareto evolutionary algorithm (SPEA).To sort the individuals in the Pareto frontier,the technique for order preference by similarity to an ideal solution(TOPSIS)is used and the optimal solution is screened out.Finally,the model is used in a 54-bus network planning.The result shows that there is a strong linear correlation between the network risk index and the outage loss,thus the network risk index can be used as an indicator of distribution network reliability.Compared with the traditional SPEA,the improved algorithm has a better search ability and a faster search speed.