Abstract

insulator defect discs model download link：https://drive.google.com/drive/folders/1dIHCax9tdh_BO_W3zgKnMFAaE99Y20US?usp=drive_link

The images data can be downloaded after registration on this website.

Data, algorithms, and computing power are the three fundamental elements driving the development of artificial intelligence technology. A comprehensive set of defect samples in power distribution equipment is a prerequisite for enhancing the accuracy of intelligent diagnostic algorithms.

In response to the scarcity of defect samples during the inspection of power distribution equipment, this dataset employs simulation modeling to generate a vast amount of defect sample data under different conditions, including diverse backgrounds, lighting, and shooting angles, within a virtual three-dimensional space. This approach nurtures the training of intelligent recognition algorithms, thereby enhancing their transferability and accuracy.

Released by the Digital Equipment Laboratory at Shanghai Jiao Tong University, this database aims to generate tens of millions of simulated defect data, encompassing various defect samples in different scenarios for transmission, substation, and distribution equipment (initially focusing on insulator defect data). Some data from different scenarios are freely accessible for academic research upon registration and login. Commercial users can access all data after approval upon submitting their enterprise registration certificate.