Research on Clustering and Pattern Mining Technology for Power Business Mapping Analysis of Power Grid

With the rapid development of modern power system and the increasing scale of power grid, the operation and management of power business are facing many challenges. Especially in the era of big data, how to extract valuable information from massive power grid data to optimize the operation and management of power business has become a research hotspot. In this paper, an improved K-means clustering algorithm-Weighted K-means (WK-means) and Weighted Sequential Pattern Mining (WSPM) is proposed to deeply analyze the mapping relationship of power services in power grid. WK-means algorithm improves the accuracy and robustness of clustering by introducing the weight of data points; WSPM method can identify frequent sequence patterns in power grid data more accurately by considering the weights of elements in the sequence. The experimental results show that WK-means algorithm can successfully cluster power grid data into different load periods, while WSPM method can effectively mine frequent sequence patterns of power load, equipment status and other characteristics. These research results provide valuable reference information for optimal dispatching, equipment maintenance and demand forecasting of power grid, and have important application prospects.