Comprehensive Condition Assessment Model of Metal Oxide Surge Arresters Based on Fusion Cloud Theory and Improved Evidence Theory

For the complex fault mechanism of metal oxide surge arresters(MOAs), and it is difficult to effectively assess the potential hidden troubles by using only a few reference physical quantities which are isolated from each other, therefore, this paper proposes an operating condition assessment model of MOAs that integrates multi-source heterogeneous data. The model firstly selects the heterogeneous datas that can reflect different fault types from online monitoring, live testing, outage maintenance and on-site inspection information database, as typical index quantities. On the basis of determining the weights of indexes by using combination weights, cloud theory is used to achieve data extraction and make membership judgements, for objectively describing ambiguity and randomness; meanwhile, improved evidence theory is used to reduce conflict between evidences and fuse uncertain information, so as to quantitatively describe the operation condition which is fuzzy and qualitative. The examples verify the validity of the model and provide a theoretical reference for the actual condition maintenance of arresters.