Design of Online Intelligent Fault Diagnosis Experiment for Single-phase Pulse Rectifiers
A data-driven intelligent diagnostic method was proposed for power device open-circuit faults and sensor faults in the single-phase pulse rectifier of the high-speed train power traction system.Firstly,a mathematical model of the single-phase pulse rectifier system was performed,and was used to analyse the topology of power device open-circuit faults and the mechanism of sensor faults.Secondly,an intelligent diagnostic model was designed based on the extreme learning machine algorithm and nonlinear autoregressive structure.Besides,the experiment for fault online intelligent diagnosis was divided into offline training and online verification phases.Finally,a rapid prototype control experiment platform was built based on the RT BOX semi-physical simulator to experimentally validate the fault models and diagnostic models.Experimental results indicate that the constructed intelligent diagnostic model for single-phase pulse rectifier faults can detect faults within 1/4 of the fundamental wave period and identify the specific fault type within 3/4 of the fundamental wave period.
万方数据
维普
