Research On User Electricity Consumption Behavior Analysis Based On Machine Learnin

With the development of traditional power grid towards smart grid,it not only intensifies the informatization of power grid,but also leads to the continuous increase of user power consumption data.Therefore,in the current context,selecting the appropriate machine learning algorithm to deeply mine the daily power consumption data of power users can recognize the user’s power consumption behavior patterns hidden in the massive power consumption data,and can accurately grasp the user’s power consumption law.This helps the power selling companies to understand the types of power customers,provide different power services for different types of power customers,and meet the intelligent power needs of users.Therefore,it is of great practical significance to study an effective clustering algorithm for analyzing users’ power consumption behavior.Starting from the power consumption data on the user side,this paper uses an appropriate clustering algorithm to analyze the power consumption behavior of power users.In order to achieve the purpose of research,this paper has done the following work.In view of the problem that k-means algorithm needs to manually set the number of clusters in advance in the process of power load data clustering,and the improper selection of the number of clusters is easy to lead to the local optimization of the clustering results,this paper introduces the GSA elbow criterion to determine the number of clusters,and verifies the effectiveness of the GSA elbow criterion through simulation experiments.Aiming at the problem that K-means clustering algorithm is easy to be affected by the selection of initial clustering center,this paper proposes a k-means algorithm based on chaotic sailfish optimizer.Firstly,we improved two aspects of the sailfish optimizer:(1)introducing Tent chaotic sequence,using the ergodic,randomness and regularity characteristics of Tent chaotic sequence to initialize the sailfish and sardine population,enrich the diversity of population,and enhance the global search ability of the algorithm.(2)Gaussian mutation operator is introduced to strengthen the local search ability of the algorithm;For some individuals who fall into local optimization,first use these individuals to generate tent chaotic sequence,and then use the generated tent chaotic sequence to disturb these individuals to help the algorithm keep searching after jumping out of local optimization.Then the chaotic sailfish optimizer is introduced into the k-means algorithm to search the optimal initial cluster center,so that the initial cluster center is closer to the actual cluster center.Finally,the k-means algorithm based on chaotic sailfish optimizer is simulated and compared with the original k-means algorithm.The example results show that the improved k-means algorithm is better in clustering results.Finally,the k-means algorithm based on chaotic sailfish optimizer is used to analyze the power consumption behavior of users.The results show that the proposed method can effectively identify users with different power consumption modes.