Real-Time Stability Performance Monitoring and Evaluation of Maglev Trains’ Levitation System: A Data-Driven Approach
Yunsong Xu, Zhiqiang Long, Zhengen Zhao, Mingda Zhai, Zhiqiang Wang
Abstract
The commercial operation of the maglev train puts critical demands on reliability and operation performance. Since maglev trains’ levitation system is open-loop unstable, its closed-loop stability under various operation conditions is a primary consideration. This paper thus addresses real-time stability performance monitoring and evaluation of maglev trains’ levitation system. First, a novel real-time performance indicator for stability performance monitoring is proposed. It utilizes only the online data of the levitation system, without knowing its accurate model. Then, performance evaluation is carried out by grading the monitored stability performance into different levels for further handling. Moreover, the proposed methods offer a way of controller evaluation for the levitation system before the train is put into commercial operation. It is noteworthy that, the proposed methods do not require the model of the levitation system to be known and run in real-time. The effectiveness and efficiency of the proposed methods are validated on a levitation system.