Institutional Repository of Coll Elect & Informat Engn
Power load data preprocessing based on PFCM algorithm and neural network | |
Hao, Xiaohong1; Zhang, Chunyan2; Li, Hang2; Wang, Weizhou3; Liu, Fuchao4 | |
2019-05-01 | |
会议名称 | 8th IEEE Joint International Information Technology and Artificial Intelligence Conference, ITAIC 2019 |
会议录名称 | Proceedings of 2019 IEEE 8th Joint International Information Technology and Artificial Intelligence Conference, ITAIC 2019 |
页码 | 1579-1583 |
会议日期 | May 24, 2019 - May 26, 2019 |
会议地点 | Chongqing, China |
出版者 | Institute of Electrical and Electronics Engineers Inc. |
摘要 | Load data of power system are easily affected by various random factors to generate the abnormal data, which affects the accuracy of load forecasting and the effectiveness of load dispatch, even damages the safe and stable operation of power grid. In response to these anomalous data, a method of outliers modification integrated by PFCM clustering algorithm and RBF neural network is proposed in this paper. Firstly, the FPCM algorithm adaptively determines the number of clusters is used to cluster load curve according to the similarity and identify the abnormal data; Secondly, Secondly, the RBF network which optimized by genetic algorithm optimization is trained by classification of normal data where abnormal data belong to; Finally, the trained neural network is used to correct the abnormal data. The experiment indicates: the method can obtain the clustering result of the ideal load curve, and correct outliers more accurately at same. Compared with the traditional correction method, the average relative error of the method is reduced by 1.08% and 1.51% respectively. © 2019 IEEE. |
关键词 | Classification (of information) Curve fitting Electric load dispatching Electric power plant loads Electric power transmission networks Genetic algorithms Radial basis function networks Statistics Average relative error Clustering results Genetic-algorithm optimizations Number of clusters Outliers Power load RBF Neural Network Trained neural networks |
DOI | 10.1109/ITAIC.2019.8785443 |
收录类别 | EI |
语种 | 英语 |
EI入藏号 | 20193507359859 |
EI主题词 | Clustering algorithms |
来源库 | Compendex |
分类代码 | 706.1.1 Electric Power Transmission - 903.1 Information Sources and Analysis - 921.6 Numerical Methods - 922.2 Mathematical Statistics |
引用统计 | |
文献类型 | 会议论文 |
条目标识符 | https://ir.lut.edu.cn/handle/2XXMBERH/118127 |
专题 | 电气工程与信息工程学院 |
作者单位 | 1.College of Electrical and Information Engineering, Lanzhou University of Technology, China; 2.College of Computer and Communication, Lanzhou University of Technology, China; 3.Power Control Center of Gansu Electric Power Company, China; 4.Power Grid Technology Center of Gansu Electric Power Research Institute, China |
第一作者单位 | 兰州理工大学 |
推荐引用方式 GB/T 7714 | Hao, Xiaohong,Zhang, Chunyan,Li, Hang,et al. Power load data preprocessing based on PFCM algorithm and neural network[C]:Institute of Electrical and Electronics Engineers Inc.,2019:1579-1583. |
条目包含的文件 | 条目无相关文件。 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论