Indian railways have become the latest in the town to have vouched for the importance of artificial intelligence. In a latest development, Indian railways announced to make an effective use of AI to predict signal failures. They aim to reduce the possibilities of signals failing and have undertaken remote condition monitoring of the system, which is a new approach for the largest transporter in the country.
The signalling system plays a vital role in the overall functioning of the railways and they completely depend on the its accurate working and the real time information fetched by these signals for their smooth running. In fact, failure of these signals is one of the major reasons for train accidents and delays. According to a parliamentary standing committee report presented in 2016, a total of 239 accidents happened on the network of Indian Railways between 2003 and 2015, 80% of it due to derailment.
This could be largely attributed to the current methods involved in the railways where they follow a manual maintenance system and depend on find-and-mix methods rather than predict-and-prevent approach. To bring more efficiency in their overall running, railways are introducing remote condition monitoring using non-intrusive sensors for continuous online monitoring of signals, track circuits, axle counters and their sub-systems of interlocking, power supply systems including the voltage and current levels, relays, timers.
This system would keep collecting the inputs on a predetermined interval and send it to a central location. This would ensure keeping a check on any flaws in the signaling system, and detect it on real time basis, allowing a speedy rectification of the error and avoiding possible delays and mishap.
The system allows data to be transferred through a wireless medium and with the help of artificial intelligence, predictive and prescriptive big data analytics, this data will be utilised to generate desired inputs. This would enable prediction of signalling asset failures, automated self-correction and informed decisions on intervention strategies.
The railways have plans to implement this in two sections of Western Railways and South Western Railway at Ahmedabad-Vadodara and Bengaluru-Mysuru route. Depending on the success of these trails, they will be gradually extended to other sections.
Try deep learning using MATLAB