随着电力系统规模的日益增大，新能源的不断加入，系统中的知识总量呈爆炸式增长，电力系统运行需基于更高的数据质量实现，以便为系统提供全方位，全周期的数据共享。国内电力信息系统所使用的数据库一般为结构化数据库。而传统关系型数据库在处理大数据复杂关系问题过程中，一系列技术瓶颈日益凸显，传统数据库已经无法满足海量数据的处理建模与分析。本文提出了一种全自动化新能源大数据异常检测的技术方法，它利用知识图谱天然反应数据间现有关系的优势，基于图结构和图顶点的属性信息，对异常图模式进行形式化定义以直接挖掘电网拓扑结构中的异常数据。本文挖掘的异常数据在现实中具有语义信息，在异常数据检测问题上具有可行性和实用价值。通过挖掘富有语义信息的异常图模式，检测新能源大数据中的异常数据，以保证数据的可靠性和准确性，避免错误或无效数据影响电力系统精细化管理和电网安全运行。算例实验效果良好，表明所提出的辨识方法具有理论价值和实际应用价值。 With the increasing scale of the power system and the continuous addition of new energy, the total amount of knowledge in the system has exploded. The operation of the power system needs to be realized based on higher data quality in order to provide the system with all-round and full-cycle data shared. The database used by the domestic electric power information system is generally a structured database. In the process of traditional relational databases dealing with the complex relational problems of big data, a series of technical bottlenecks have become increasingly prominent, and traditional databases can no longer satisfy the processing modeling and analysis of massive data. This paper proposes a fully automated new energy big data anomaly detection technology method, which uses the advantages of the existing relationship between the natural response data of the knowledge graph, and based on the graph structure and the attribute information of the graph vertices, the abnormal graph mode is formalized to directly mine abnormal data in the grid topology. The abnormal data mined in this paper has semantic information in reality, and has feasibility and practical value in the detection of abnormal data. By mining abnormal graph patterns rich in semantic information, abnormal data in new energy big data is detected to ensure the reliability and accuracy of the data, and prevent errors or invalid data from affecting the refined management of the power system and the safe operation of the power grid. The experimental results of the calculation examples are good, indicating that the proposed identification method has theoretical value and practical application value.
展开▼
机译：随着电力系统规模的日益增大，新能源的不断加入，系统中的知识总量呈爆炸式增长，电力系统运行需基于更高的数据质量实现，以便为系统提供全方位，全周期的数据共享。国内电力信息系统所使用的数据库一般为结构化数据库。而传统关系型数据库在处理大数据复杂关系问题过程中，一系列技术瓶颈日益凸显，传统数据库已经无法满足海量数据的处理建模与分析。本文提出了一种全自动化新能源大数据异常检测的技术方法，它利用知识图谱天然反应数据间现有关系的优势，基于图结构和图顶点的属性信息，对异常图模式进行形式化定义以直接挖掘电网拓扑结构中的异常数据。本文挖掘的异常数据在现实中具有语义信息，在异常数据检测问题上具有可行性和实用价值。通过挖掘富有语义信息的异常图模式，检测新能源大数据中的异常数据，以保证数据的可靠性和准确性，避免错误或无效数据影响电力系统精细化管理和电网安全运行。算例实验效果良好，表明所提出的辨识方法具有理论价值和实际应用价值。 With the increasing scale of the power system and the continuous addition of new energy, the total amount of knowledge in the system has exploded. The operation of the power system needs to be realized based on higher data quality in order to provide the system with all-round and full-cycle data shared. The database used by the domestic electric power information system is generally a structured database. In the process of traditional relational databases dealing with the complex relational problems of big data, a series of technical bottlenecks have become increasingly prominent, and traditional databases can no longer satisfy the processing modeling and analysis of massive data. This paper proposes a fully automated new energy big data anomaly detection technology method, which uses the advantages of the existing relationship between the natural response data of the knowledge graph, and based on the graph structure and the attribute information of the graph vertices, the abnormal graph mode is formalized to directly mine abnormal data in the grid topology. The abnormal data mined in this paper has semantic information in reality, and has feasibility and practical value in the detection of abnormal data. By mining abnormal graph patterns rich in semantic information, abnormal data in new energy big data is detected to ensure the reliability and accuracy of the data, and prevent errors or invalid data from affecting the refined management of the power system and the safe operation of the power grid. The experimental results of the calculation examples are good, indicating that the proposed identification method has theoretical value and practical application value.
展开▼
