基于密度峰值快速搜索聚类的多场景分布式电源规划

Multi-scene distributed generation planning based on clustering by fast search and find of density peak

针对间歇性分布式电源出力的随机性、负荷需求的不确定性以及分布式电源与负荷之间存在相关性的问题,采用拉丁超立方抽样并结合Spearman秩相关系数的Cholesky分解,得到分布式电源具有相关性的出力与负荷需求样本。通过密度峰值快速搜索聚类算法对相关性样本进行有效削减得到典型场景,以分布式电源投资运行费用和配电网向上级电网购电费用最小为优化目标,建立分布式电源多目标规划模型。最后通过二阶锥松弛将规划模型转化为混合整数二阶锥规划问题,并调用Cplex求解器对规划模型求解。IEEE 33节点算例结果验证了所提模型的合理性。

Aiming at the problems of the randomness of intermittent distributed generation output,the uncertainty of load demand,and the correlation between distributed generation and load,Latin hypercube sampling and Cholesky decomposition combined with Spearman rank correlation coefficient were used to obtain an output and demand load relevant sample of distributed generation. The relevant sample was reduced by the algorithm of clustering by fast search and find of density peak( CFSFDP) to obtain typical scenes. The multi-objective distributed generation planning model was established with the optimization goal to minimize the investment and operation cost of the distributed generation and the power purchase cost of the superior grid. The programming model was transformed into a mixed-integer second-order cone programming( SOCP) problem through secondorder cone relaxation,and the Cplex solver was called to solve the programming model. The results of IEEE 33bus distribution network were applied to verify the rationality of the proposed model and method.