Bayesian Network-based Terminal Fault Diagnosis Technology for Low-voltage Station Area in Online Loss Application
This study analyzed the line loss in low-voltage distribution areas using decision trees and Bayesian networks.Initially,key influencing factors were identified from historical data through a decision tree model,establishing their relationships with line loss.Then,Bayesian distributions refined model parameters and uncertainty,enhancing prediction accuracy and reliability.Results show this hybrid approach effectively pinpoints critical loss factors and offers precise forecasts under varying conditions,crucial for management and optimization.The findings offer novel insights into power system operation,maintenance,and economic dispatch,demonstrating practical value.
Transformer area line losselectricity stealing dataBayesian network
万方数据
