This project aims to combine artificial intelligence theories and methods such as deep learning, machine learning, and data mining to study a new type of fault diagnosis and relay protection method for power systems. Taking the 500 kVA intelligent substation in Shenzhen. Then, due to the particularity of historical statistical data, a weight calculation method combining analytical hierarchy process (AHP) and entropy weight method is adopted to eliminate subjective factors in the weight calculation process. To prevent overfitting, this article can use a strictly separated set of training and testing samples to train the model. It is reshaping traditional grid architecture and making way for more flexible, efficient and. The text begins by covering computer-aided modeling and simulation of digital relays and focuses on the design of various relay characteristics as well as their hardware implementation.
[PDF Version]