基于电力大数据的电力用户用电特征识别模型研究

Research on the Recognition Model of Power User Power Consumption Characteristics Based on Power Big Data

采用目前方法对电力用户用电特征进行识别时,存在识别准确率低、F1分数低和识别结果易受用电数据分帧长度影响的问题。为此提出基于电力大数据的电力用户用电特征识别模型,利用电力数据采集系统采集用户用电数据,并调节用电数据负荷曲线、数据标准化和数据降维,再利用K-means聚类算法提取预处理后优化用电数据的特征,将用电特征带入支持向量机中,根据分类结果实现电力用户用电特征的识别。实验结果表明,所提方法识别准确率高、F1分数高、识别结果不受用电数据分帧长度的影响。

When using the current method to identify the power consumption characteristics of power users,there are some problems,such as low recognition accuracy,low F1 score and easy to be affected by the frame length of power consumption data.Therefore,a power user power consumption feature recognition model based on power big data is proposed.The power data acquisition sys-tem is used to collect the user power consumption data,adjust the power consumption data load curve,data standardization and data dimensionality reduction,and then the K-means clustering algorithm is used to extract the features of the optimized power consumption data after preprocessing,so as to bring the power consumption features into the support vector machine.According to the classification results,the power consumption characteristics of power users are recognized.The experimental results show that the proposed method has high recognition accuracy and high F1 score,and the recognition results are not affected by the frame length of power data.