Title: A Prototype Development of Reliable Sensor Network Based Structural Health Monitoring System For Railway Bridges
Abstract:
The Indian Railways consists of about 1,27,000 bridges, of which 40% are over 100 years old. It is critical to have a system to monitor the structural health of these bridges for maintenance and safety of the public. Present systems used for such monitoring are mainly wired systems. These systems are generally bulky and requires expertise in manpower. Also it takes days to deploy these systems on the bridge.
In this thesis work we develop an automated wireless sensor network (WSN) based system, which makes use of sensor motes and MEMS accelerometers for railway bridge monitoring. We call this system as BriMon. The system is easily deployable and requires minimum maintenance. The system though primarily designed for long term monitoring of remotely located bridges, can be used for short term monitoring too. The system has been developed by making use of off the shelf hardware. We have adopted an application driven approach in developing the protocols and components of the system. The implemented solution shows how the design choices dictated solely by the application requirement are different from general solutions that exist in the sensor network domain.
The system incorporates a novel event detection mechanism that triggers data acquisition by the data acquiring nodes in response to an oncoming train. The data from the data acquiring nodes is then reliably gathered at the base nodes. We make use of train itself to automatically transfer the vibration data from the remotely located bridge. For this we employ a simple yet effective multi-channel data transfer mechanism to transfer the data from the base nodes onto sensor nodes located on the moving train on 802.15.4 radio. The data from these nodes is transferred to a data analysis centre whenever the train passes a nearby station which provides some network connectivity to the data analysis centre.
In this work we discuss, data acquisition system, data packing and compression, transport protocol for reliable data transfer, debugging tool, and data analysis tool which we have designed, developed and evaluated. The data acquisition system is DMA based and it acquires the vibration data with high fidelity and precision. We do ten point averaging of acquired data along with compaction to compress the data acquired by the nodes. We further study the possibility of using standard lossless data compression techniques in our application. The transport protocol developed is SACK based and is used for reliable transfer of vibration data in all the cases mentioned above. In our application the mobile data transfer is time critical in the sense that we need to reliably transfer the data to mobile train within limited contact duration. The protocol achieves this gracefully with minimum latency. Also it satisfies all other requirements of the application. The debugging tool is used for debugging the application. The data analysis tool we have developed has user friendly GUI and it meets all the requirements of the structural engineers for data analysis.
Finally we discuss the integration work which we did to integrate our work with other components of the BriMon which were developed by a fellow colleague. These components are routing, time synchronization and event detection.