KEYWORDS: Patents, Control systems, Power grids, Data modeling, Semantics, Batteries, Data storage, Mathematical optimization, Education and training, Databases
At present, scientific and technological personnel in electric power enterprises mainly rely on patent search websites to obtain cutting-edge electric patent information. But these websites are mainly based on string matching, which fails to capture the connection between patents, making the recommend result unsatisfying. In view of the above problems, we first constructed a grid patent knowledge map, where knowledge extraction was carried out for entities such as title, abstract and applicant in the patent text, and the entity and defined relational data were stored in the Neo4j graph database. Secondly, the Transe-SNS algorithm with optimized negative sampling was used for vectorization of the graph entity relationship. Experiments showed that Mean Rank and hit@10 improved by 27.5 and 2.2% respectively compared with the traditional TransE algorithm. Finally, the similarity between patents was calculated by combining the results of knowledge graph vector embedding of patent entities with the word vector embedding of patent title abstraction, and top- k patents in similar fields were recommended for users. The experiment proved that the proposed method is superior to the traditional text embedding method in recommending similar patent technology fields.
In the production work of the State Grid, the power equipment is often in a high and low temperature environment. The timely detection of the fault of the power equipment is an important part of the stable development of the State Grid. This paper proposes a method to detect the quality of power equipment based on the measurement data of power equipment at room temperature. Based on the LSTM model, the working data of power equipment is predicted in the form of serialized time nodes during the experiment. Then, the feature data is extracted and dimensionality reduction according to the operating environment, and finally the anomaly detection is carried out according to the Local Outlier Factor (LOF) algorithm. The experimental results show that the prediction effect of the model prediction diagnosis results is ideal, and the LSTM-LOF combined with the model can be used to assist engineers in the quality diagnosis of power equipment.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.