Based on the predictive maintenance (PdM) method

the operation and maintenance cost can be further reduced while ensuring the stability of the transmission line. The core of the predictive maintenance method of transmission lines is to predict the spatial and temporal distribution of future high random fault risk of transmission lines in complex data environments such as bias and heterogeneity. Therefore

a fuzzy correlation pattern identification for imbalanced elements (FCPIie) prediction method is proposed

combining transmission line operational data with external environmental state information. Firstly

condition correlation pattern identification (CCPI) and probabilistic fuzzy inference systems (PFIS) are designed and integrated within a parallel learning architecture. The multi-type heterogeneous input features are evaluated separately

with common and rare distribution factors identified. Next

to address potential data bias

a fuzzy conditional correlation pattern identification (FCCPI) model is established for in-depth qualitative screening of both common and rare factors. High risk (HR) and rare high risk (RHR) factors are then extracted to clarify the underlying fault mechanisms

enabling the prediction of future fault risk distribution. Finally

the proposed method is validated through an example simulation

demonstrating its feasibility in practical applications.