AUTHOR=Luo DianSheng , Xi RuiYao , Che LiXuan , He HongYing 

TITLE=Health condition assessment of transformers based on cross message passing graph neural networks

JOURNAL=Frontiers in Energy Research

VOLUME=Volume 10 - 2022

YEAR=2022

URL=https://www.frontiersin.org/journals/energy-research/articles/10.3389/fenrg.2022.973736

DOI=10.3389/fenrg.2022.973736

ISSN=2296-598X

ABSTRACT=A method is presented to assess the health condition of transformer based on cross message passing graph neural network (CMPGNN) in this paper. Aiming at high accuracy of transformer health condition assessment, multiple indicators and their strong correlation are taken into accounted. Indicators which impact the assessment are divided into four comprehensive state categories, and each category have several indicators. Firstly, the correlations between indicators in the one state category are extracted by health index method and criteria importance though intercriteria correlation (CRITIC) method and the health index of comprehensive indicator is obtained. Then, a relationship graph of comprehensive indicators is established and the correlations between indicators of different comprehensive state categories are extracted. Finally, CMPGNN is constructed to achieve the health assessment. The experimental results show that, the method presented in this paper highly improves the accuracy of transformer health condition assessment.