Back to Search Start Over

A multi-agent system based fault diagnosis for active distribution systems

Authors :
Fangyuan Tian
Yusheng Xue
Fushuan Wen
Md. Abdus Salam
Xiaolei Wang
Source :
2016 IEEE Innovative Smart Grid Technologies - Asia (ISGT-Asia).
Publication Year :
2016
Publisher :
IEEE, 2016.

Abstract

The penetration of distributed generation (DG) in an active distribution system could affect the fault current level and characteristics of the original/traditional system. Thus, the protection system originally configured for a traditional distribution system should be updated to meet the operational requirement of an active distribution system. In developing a fault diagnosis method for an active distribution system, these factors should be taken into account to attain accurate fault diagnosis result. Given this background, a multi-agent system (MAS) based fault diagnosis method is presented for active distribution systems. A methodological framework is first proposed, and the communication mechanism in the MAS architecture addressed. Next, the fault diagnosis agent is designed with an existing analytical model for fault diagnosis embedded, and a correlation factor employed to represent the coupling relationship between related bus agents. The impacts of the information loss and information distortion on the accuracy of fault diagnosis results are also analyzed separately. A MAS based fault diagnosis strategy is then developed with a communication and negotiation mechanism utilized for attaining the global optimal fault diagnosis result. Test cases carried out on a modified 32-node active distribution system demonstrate the presented model and method for fault diagnosis.

Details

Database :
OpenAIRE
Journal :
2016 IEEE Innovative Smart Grid Technologies - Asia (ISGT-Asia)
Accession number :
edsair.doi...........b154dff7acacccf7377702620349afb1