Cluster Analysis of Power User Loads Based on KPCA and K-means++
- Resource Type
- Conference
- Authors
- Guan, Yan; Lu, XinYi; Gao, XiYing; Zhou, Hang; Liu, Ye; Tian, GuiYang
- Source
- 2022 3rd International Conference on Big Data, Artificial Intelligence and Internet of Things Engineering (ICBAIE) Big Data, Artificial Intelligence and Internet of Things Engineering (ICBAIE), 2022 3rd International Conference on. :229-233 Jul, 2022
- Subject
- Communication, Networking and Broadcast Technologies
Robotics and Control Systems
Signal Processing and Analysis
Power demand
Clustering methods
Clustering algorithms
Companies
Big Data
Power markets
Stability analysis
KPCA
K-means++
User clustering
- Language
In view of China's fiercely competitive power market, more power companies want to study the power consumption habits of power users to provide more personalized services. This paper proposes a power user load clustering method based on KPCA and K-means++. After the user load data is dimensionally reduced by KPCA, K-means++ is used for clustering processing. The experiment uses the 2018 power user load data in a region in northeastern China to simulate. The results show that the proposed method has improved effectiveness and stability compared with other traditional clustering methods.