Train Alarm Data Linking via Geographic Binning and Peer Voting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In the rail industry, there is a critical need to immediately notify train drivers of an over temperature alarm in wheel bearings, as it precedes catastrophic failure, and existing systems struggle to effectively link alarm data from physically disassociated wireless sensors to a moving train.
Innovation Solution
A system utilizing an in-cab monitor with GPS, cellular modem, and a 'Publish and Subscribe' Message Broker to validate and notify the driver of wheel bearing temperature alarms, employing geographic binning and peer voting to accurately associate alarms with the correct train, allowing for timely action.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If wireless sensors are physically disassociated from the train control system, then sensor deployment flexibility is improved, but the ability to link alarm data to the correct train deteriorates
Solution Approach 1:
The patent introduces a Message Broker as an intermediary component that receives alarm data from wireless sensors and matches them to trains using GPS location information. The Message Broker acts as a mediator between the physically disassociated sensors and the train control system, enabling accurate association without physical connection. The broker compares sensor GPS coordinates with train GPS coordinates to determine which train should receive the alarm notification.
2Measurement precision
If GPS location data is continuously transmitted for every alarm, then alarm location accuracy is improved, but data bandwidth consumption deteriorates
Solution Approach 1:
The patent segments the service area into geographic bins and assigns each bin a unique identifier. Instead of transmitting complete GPS coordinates for every alarm, the system determines which bin the alarm occurred in and transmits only the bin identifier. This segmentation approach maintains location accuracy by providing sufficient geographic information while dramatically reducing data bandwidth consumption by replacing precise coordinate data with compact bin codes.
3Reliability
If the system subscribes to all possible location topics, then alarm detection completeness is improved, but processing overhead deteriorates
Solution Approach 1:
The patent implements local quality by having each train's Monitor/Relay subscribe only to the specific location bin topics corresponding to its current geographic position, rather than subscribing to all possible topics. The system dynamically determines which bins the train is currently in and subscribes only to those topics, reducing processing overhead while maintaining complete alarm detection for the train's operational area.
4Measurement precision
If peer voting mechanism is implemented to verify alarm ownership, then alarm association accuracy is improved, but system complexity deteriorates
Solution Approach 1:
The patent implements a peer voting mechanism where multiple Monitor/Relay systems that receive the same alarm message can vote to determine which train the alarm belongs to. The alarm is associated with the train that receives the majority of votes. This feedback-based consensus approach improves alarm association accuracy by cross-validating location data from multiple sources, while the added complexity is managed through automated voting algorithms rather than manual intervention.
Data Source
AI summary
A system for linking alarm data from physically disassociated wireless sensors to a train in motion. The sensor secured to a wheel axle-box of a rail carriage of a train to detect an over temperature alarm to enable a driver to stop the train. If the alarm condition is confirmed as belonging to the train, a Monitor/Relay immediately notes that there is a problem and warns an operator.