No abstract

The KNN algorithm is one of the most famous algorithms in machine learning and data mining. It does not preprocess the data before classification, which leads to longer time and more errors. To solve the problems, this paper first proposes a PK-means++ algorithm, which can better ensure the stability of a random experiment. Then, based on it and spherical region division, an improved KNNPK+ is proposed. The algorithm can select the center of the spherical region appropriately and then construct an initial classifier for the training set to improve the accuracy and time of classification.

As the development of intelligent power system, much importance and requirement have been attached to the load data analysis. In view of the current rough classification of load data, propose a weighted fuzzy clustering algorithm to detail the load classification dividing, which adds a weight distribution process to balance the different influence of various factors. In addition, Two group of experiments are set to verify the efficiency of this method. The experiment results show that the algorithm is effective to accurately cluster the load data and supportive to the fine analysis of load data.

KEYWORD:Orderly power utilization; Twice clustering algorithm; Load data analysis; Power load data clustering ABSTRACT: As the requirement of smart grid increasing, a twice clustering algorithm is proposed here to extract the characteristic of load data curve. In addition, the calculating algorithm of load performance is also provided to analyze the curves, including seasonal maximum load curve, seasonal safe load curve, economic production load curve, summer (winter) air conditioning load and production protected load curve, upon which the adjustable potential of user can be estimated, so that to set up the economic estimation for companies. Moreover, the above processing method is finally verified through the experiment design upon the data resources from an industry company in Jiangyin.