融入多源量测数据的配电网分布式区间状态估计

A Distributed Interval State Estimation Framework of Distribution Networks Based on Multi-source Measurements

新能源出力间歇性使得大规模配电网状态估计需要同时考虑不确定因素与模型计算复杂度,且多类型量测数据的兼容性问题也会影响状态估计结果。首先,定义适用于三相不平衡配电网的等效电气距离概念,结合社区发现算法将配电网合理划分为若干子区域;其次,根据子区域内光伏发电系统伪量测与实时量测数据构成本地多源量测数据的区间表达形式,利用区间量测变换技术将电压、功率等多源量测数据统一转换为系统注入电流数据,并建立考虑量测数据与线路参数双重不确定性的配电子区域区间状态估计模型;最后,采用改进区间优化方法对本地区间状态估计模型进行求解,并完成相邻子区域间边界状态信息的交互,从而输出配电网全局区间状态结果。通过算例仿真与结果对比分析,结果验证所提配电网分布式区间状态估计方法在估计结果精度与效率方面相比现有方法而言都具备一定的优势,且能够有效跟踪系统多重不确定变量对状态估计结果的影响。

Due to stochastic power injections of the renewable energy,the processing of state estimation in the large-scale distribution network needs to consider more uncertainties and complexities,and the compatibility of multiple types of measurements would also affect the state estimation results.The equivalent electrical distance was firstly defined and the unbalanced distribution network was divided into several subareas based on the community discovery algorithm.Then,the interval number of local multi-source measurements consisting of pseudo-measurements and real-time measurements were obtained.A local interval state estimation model for the local distribution network considering the bi-level uncertainty of measurements and line parameters was established,by using interval measurement transformation technology to uniformly convert multi-source interval measurements into system injection current data.Finally,a modified interval optimization method was used to effectively solve the local interval linear state estimation model,and the interaction of boundary state between adjacent sub-areas was completed to output the global interval state results of the distribution network.Case simulations and comparisons have illustrated that the proposed method possesses better performance in estimation accuracy and computational efficiency than these traditional ones,and it can track the influence of the multiple uncertain variables on the state estimation results.