摘要
针对电力负荷数据存在不良数据的问题,本文提出一种基于爬山-蚁群-FCM模糊聚类算法的不良负荷数据辨识及修正方法,将启发式算法与元启发式算法结合,克服搜索算法容易陷入局部最优的缺陷。以爬山算法为蚁群算法提供初始解,由蚁群算法为FCM模糊聚类算法提供聚类数目和聚类中心,克服传统聚类算法选取聚类数据和聚类中心的偶然性。根据FCM模糊聚类结果计算特征曲线和可行域上下限,辨识不良负荷数据,并对不良负荷数据采用插值法进行数据修正。算例结果表明,本文方法可以有效实现不良负荷数据辨识功能,在模型的准确性上相比单一搜索算法更优,具有较高的准确性和鲁棒性,有助于提升电网数据质量,为负荷数据辨识乃至其他电力数据辨识等领域提供了一种研究思路。
Aiming at the problem of bad data in power load data, this paper proposes a method for identifying and correcting bad load data based on the hill clinbing-ant colony-FCM fuzzy clustering algorithm, which combines heuristic algorithms and meta-heuristic algorithms to overcome the tendency of search algorithms to fall into defects of local optima. The hill climbing algorithm is used to provide the initial solution for the ant colony algorithm, and the ant colony algorithm provides the number of clusters and cluster centers for the FCM fuzzy clustering algorithm, which overcomes the contingency of the traditional clustering algorithm in selecting cluster data and cluster centers. According to the FCM fuzzy clustering results, the characteristic curve and the upper and lower limits of the feasible region are calculated, the bad load data is identified, and the bad load data is corrected by interpolation. The results of the calculation examples show that the proposed method can effectively realize the identification function of bad load data. It is better than a single search algorithm in the accuracy of the model, has higher accuracy and robustness, and helps to improve the quality of power grid data. it provides a research idea for load data identification and other power data identification fields.
作者
甘迪
金岩磊
葛立青
郭鑫溢
GAN Di;JIN Yanlei;GE Liqing;GUO Xinyi(NR Electric Co.,Ltd.,Nanjing 211102,Jiangsu,China)
出处
《电力大数据》
2022年第1期1-8,共8页
Power Systems and Big Data
关键词
数据辨识
不良负荷
蚁群算法
爬山算法
聚类
data identification
bad load
ant colony algorithm
hill climbing algorithm
clustering
