A novel intelligent train condition monitoring system coupling laser beam into image processing algorithm

摘要

Wheels, brake pads and pantograph carbon strips of modern electric trains are three safety critical elements, which are always subject to wear and tear during normal running and daily operation. Inspection process of the wheel and brake pad wear is extremely labour intensive and time-consuming, due to their large quantity in each train. Moreover, before the manual inspection can be carried out from the underframe of the train, it is necessary to shunt the train concerned onto the maintenance pit. Thus additional time is required for shunting logistics due to the limited number of pits in depot. On the other hand, to check the wear on carbon strips, manual gauging is the common practice, which requires access to the train roof. Thus it needs a high-level platform and electrical isolation of the overhead power line prior to the examination. It is again no doubt to be a time-consuming process and causes degradation on the efficiency of the regular maintenance. Therefore, to enhance the productivity, an automatic and integrated Vehicle Monitoring System (VMS) utilising laser assisted image processing technology, which comprises three monitoring systems, namely wheel profile, brake pad and pantograph wear monitoring systems, is introduced in the Kowloon-Canton Railway Corporation (KCRC) Tai Wai Depot to automatically monitor the condition of wheels, brake pads and pantograph carbon strips of the Ma On Shan 4-car Electric Multiple Units (EMUs) and measure their respective wear pattern, when the train passes the train washing plant daily. The data obtained not only provides a useful database for condition monitoring 8 maintenance planning of wheels, brake pads and pantographs, it is also able to alarm any premature failure for necessary corrective actions.