Residential Area Spatial Load Forecasting Method Based on Big Data Mining Technology

Loading...
Thumbnail Image
Paper number
1053
Working Group Number
Conference name
CIRED 2019
Conference date
3-6 June 2019
Conference location
Madrid, Spain
Peer-reviewed
Yes
Short title
Convener
Authors
Zhang, Xujun, Huazhong University of Science and Technology, China
Li, Yan, Huazhong University of Science and Technology, China
Liu, Yiming, Huazhong University of Science and Technology, China
Guo, Xusheng, Huazhong University of Science and Technology, China
Cai, Zhifei, State Grid Xuchang Power Supply Company, China
Ke, Song, State Grid Xuchang Power Supply Company, China
Abstract
Spatial load forecasting is the precondition and foundation of power system planning. With the requirement of lean management of power system, the inaccurate spatial load forecasting makes the actual maximum load of the residential area not match the reported capacity. Some areas have become light load and the remaining intervals are seriously less, while other areas have overloaded, causing waste to the city's public grid resources, failing to achieve reasonable optimal allocation of grid resources. This paper proposes a spatial load forecasting method based on big data mining technology. Firstly, big data collection and multi-source data fusion were carried out, and sample collections were constructed based on python and the electric power enterprise. Then K-means clustering algorithm is used to cluster the load samples. The historical load characteristics of the residential area are obtained according to the classified historical data of the residential area after clustering. The logistic function is used to predict the release characteristic curve of residential area load. The least squares fitting is used to determine the values of unknown coefficients k, a and b, and the model of load development law is established. Finally, the influence of factors such as geographical location, floor area ratio and occupancy rate on load development is analyzed by rough set information entropy theory. The proposed method is used to predict the spatial load of urban residential area in a certain area and the results show that the proposed method can improve the accuracy of load forecasting.
Table of content
Keywords
Publisher
AIM
Date
2019-06-03
Permanent link to this record
https://cired-repository.org/handle/20.500.12455/262
http://dx.doi.org/10.34890/490
ISSN
2032-9644
ISBN
978-2-9602415-0-1