A Normal Cloud Model for Transformer Insulation Condition Prognosis with Optimal Weights
Lei Yang · 2018 Condition Monitoring and Diagnosis (CMD) · 2018
Prognosis of transformer insulation condition is a task with randomness and uncertainty coexist. In this paper, a normal cloud model is proposed for handle such issue. meanwhile, an optimal weighting method is adopted to deal with the subjective in evaluating the importance of each variable. The proposed approach can realize a qualitative-quantitative scale transformation with utilizing the composite scale of binary semantics method and the variable weighting of each variable. A normal cloud model with reasonably assigned properties is established to characterize the fuzziness and randomness of the transformer's insulation condition. With this model available, the condition of the entire transformer and its fault type are determined via a process named cloud aggregation. Case study with field collected data demonstrated the feasibility of the proposed approach. Results demonstrate that the proposed normal cloud model can be a potential choice for online condition prognosis of power transformer.