基于PCA-SOM的低压配网运行态势评估方法

A Method to Evaluate the Operation Situation of Low Voltage Distribution Network Based on Principal Component Analysis and Self-organizing Neural Network

基于低压配网自动化水平较低且现有运行态势评估方法未能涵盖配电网智能化的需求,提出了一种基于PCA-SOM的低压配网运行态势评估方法,首先利用改进的主成分分析提取低压配网运行态势评估指标并建立评估指标体系;其次对已构建指标体系中相关的配网运行数据进行基于自组织神经网络的异常数据辨识和清理;再次对各层次评估指标建模并计算权重;然后计算各层次指标值及评估得分,找出配网运行薄弱环节;最后对广西某地区低压配网进行运行态势评估.结果表明,该方法能够科学合理地建立配网运行态势评估指标体系,以及有效减少异常数据对配网运行态势评估的影响,可为低压配网运行态势监测和运行优化提供参考.

Based on the fact that the automation level of low-voltage distribution network is low and the existing operation situation assessment methods can not cover the demand of distribution network intelligence,a low-voltage distribution network operation situation assessment method based on PCA-SOM is proposed in this paper.Firstly,the improved principal component analysis is used to extract the low-voltage distribution network operation situation assessment index and establish the evaluation index system.Secondly,the relevant distribution network operation data in the constructed index system are identified and cleaned up based on self-organizing neural network.Thirdly,the evaluation index of each level is modeled and the weight is calculated.Then the index value and evaluation score of each level are calculated and the weak links in the operation of the distribution network are found out.Finally,the operation situation assessment of the low-voltage distribution network in a certain area of Guangxi Province is performed.The results show that the distribution network operation situation assessment index system can be established through this method scientifically and reasonably,and then the influence of abnormal data on the distribution network operation situation assessment can be reduced effectively.It can provide a reference for low-voltage distribution network operation situation monitoring and operation optimization.