The Construction Techniques of Artificial Intelligence Hierarchical Dataset in Power Industry

Artificial intelligence datasets now are common in power electric companies for their own research work. Many Artificial Intelligence research are data-driven and require large amount of data. A single company's dataset only contains quite limited amount of data. In this paper, a hierarchical dataset is introduced, with larger and higher level dataset to unify lower level dataset to provide more valuable data than a single one. The basic ideas in designs of different existing datasets are analyzed and a hierarchical structure that integrate existing dataset features is proposed. In order to condense the data in higher level dataset, the data collection standard is designed based on data category priority and annotation. By separating the data synchronization API and data management API, both physically and logically data collection can be realized. We discussed the unification of annotation rules of the existing different datasets. Based on the relationship between the annotation data and the annotation rules, we put forward the method to get unified annotation rules. At last we introduced the implementation of hierarchical dataset, which considers the critical points of data storage method and data synchronization.