Development of a Platform to Enable Real time, Non-Disruptive Testing and Early Fault Detection of Critical High Voltage Transformers and Switchgears in High Speed Rail

Start: July 1, 2017, End: June 30, 2020

Principal Investigators/Researcher

Yingtao Jiang and Ruixue Sun, University of Nevada, Las Vegas

Project Description

Key to the success of any high speed rail system is to keep the operation and maintenance cost and schedule in check, and one big element of any operation, maintenance and safety effort is to ensure that all the critical high voltage systems, traction transformers (e.g., 2 x 25kV Autotransformer Feed Type Traction Power Supply Systems in California High Speed Rail), sulfur hexafluoride Gas Insulated Switchgear (GIS), and sulfur hexafluoride switchgear (C-GIS) and alike for high-speed trains, are working properly. Of many effective detection methods and tools that are available, Partial Discharge (PD) detection holds its advantage for monitoring the state of health of high voltage transaction transformers and GIS.  In this study, we tend to use a group of sensors, instead of a single type of sensor, to reliably detect and locate PD in high voltage transformers and GIS in real time.

The goals of this research thus include (i) developing a platform and method to improve the quality and efficiency of testing critical high voltage hardware systems used in high speed rails; (ii) developing methods and hardware for localization and detection of partial discharge in high voltage transformers and switchgears  using UHF, ultrasonic, TEV and HF current sensors;  (iii) developing  a software package to automate the testing and analysis processes; and (iv) testing and validating  the developed platform (hardware and software) in controlled lab environment and in the field.

It is expected that the proposed PD detection platform and equipment can be eventually help reliably and continuously monitor and evaluate high speed rail system during its entire lifecycle, and any problem or potential hazard should be promptly detected at low cost.