High-quality power equipment is the basis for ensuring the safe operation of the power grid and improving the reliability of power supply. In actual operation, although some electrical equipment can continue to be used, abnormal operation or hidden dangers will affect the safety of people, equipment and power grid, reliable and economic operation of power grid and equipment, equipment output or life span, and power quality. Therefore, research on multi-dimensional analysis of the same batch of substation equipment, and identification of suspected family defects and frequent defects is of great significance for improving the company’s equipment management and defect management. The maturity and promotion of unstructured text data mining, graph computing technology, and semantic analysis have provided a wider dimensional space for the analysis of power equipment defect data. Aiming at the shortcomings of traditional defect data analysis, this article summarizes and analyzes the defects and hidden dangers found in the equipment, traces the distribution status of the same equipment through the physical “ID”, carries out multi-dimensional analysis of the same batch of equipment, and investigates the hidden dangers of equipment family defects. A big data analysis algorithm to build automatic identification models of suspected familial defects and frequent defects in main substation equipment such as transformers, disconnectors, circuit breakers, and use graph computing technology to quickly integrate analysis capabilities in multi-source heterogeneous data fusion with hidden relationship discovery capabilities, identify equipment defects with potential impact relationships. From historical data, discover which manufacturers or purchased batches of equipment are more likely to have the same defect, and which defect types have a higher frequency, improve the recognition model of family defects and frequent defects, and improve the accuracy and comprehensiveness of defect recognition.

Data of power transmission and transformation in smart grid demonstrates the characteristics of big data. First of all, the article listed types of the data, introduced the source of the data and explained how to gain the data; then analyzed the characteristics of the data in aspects of volume, velocity, variety and value which were well-known in field of big data.

No abstract