Method of Analyzing Transformer DC Magnetic Bias Based on Big Data Cleaning and Dimensionality Reduction

2019 2nd International Conference of Green Buildings and Environmental Management

生态环境科学

Yuan, Pei^1 ; Wu, Xiaowen^1 ; Wan, Tao^1 ; Ye, Huisheng^1 ; Liu, Yun^1

State Grid Hunan Electric Power Company Limted Research Institute, Changsha, Hunan

410007, China^1

关键词: Dc magnetic bias;    Dimensionality reduction;    Key influencing factors;    Magnetic analysis;    Magnetic treatment;    Multi dimensional;    Prediction model;    Sequence control;   

Others  :  https://iopscience.iop.org/article/10.1088/1755-1315/310/3/032026/pdf
DOI  :  10.1088/1755-1315/310/3/032026

学科分类：环境科学（综合）

来源: IOP

【 摘 要 】

Transformer DC magnetic bias is difficult to accurately analyze and predict because it is caused by complex factors and has a wide influence. This paper combines big data with DC bias magnetic analysis method, making a breakthrough in the DC bias prediction model that only considers a single factor. From the perspective of multi-system coupling, this paper analyzes the modeling of DC magnetic bias based on multi-dimensional big data, and extracts the key influencing factors of transformer DC bias through multi-system big data cleaning and dimensionality reduction technology, thus providing an effective basis for "one-button sequence control" bias magnetic treatment.