Open AccessData Preprocessing and Quality Evaluation for Building the Power Grid Supervision Knowledge Graph
Author(s) -
Zhang Xinjie,
Wang Jian,
Lingxu Guo,
Xin Wang,
Yao Wang,
Xu Lei
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/2005/1/012007
Subject(s) - computer science , preprocessor , data pre processing , data mining , data redundancy , graph , data quality , grid , redundancy (engineering) , reliability engineering , artificial intelligence , theoretical computer science , database , engineering , metric (unit) , operations management , geometry , mathematics , operating system
In the process of constructing the power grid supervision knowledge graph, it is necessary to sort out and integrate structured and unstructured multi-source data for knowledge extraction and reasoning. In order to solve the quality problems of multi-source data redundancy and errors, this paper proposes a multi-source data quality evaluation system to achieve multi-dimensional quality evaluation of power grid supervision data such as maintenance, defect, measurement, alarm and oil chromatography. Data preprocessing methods are firstly adopted for text and numerical data separately to complete data noise reduction, data filtering, data filling, etc. According to the actual calculation example, the practicability and effectiveness of the data preprocessing method and data quality evaluation system are verified. Finally, the data value of the power grid supervision knowledge graph is significantly improved, which is helpful to comprehensively improve the equipment state perception ability.