A simple but effective system has been demonstrated for monitoring hour meters remotely over a cellular network, with a simple web-based user interface for generating reports. In the first stage of the project, system specifications were developed and a basic architecture was designed for a distributed wireless fleet monitoring system. The initial implementation considered hour metering only, but location and machine health status indicators were also added. Next, a preliminary proof-of-concept prototype was prepared and demonstrated. Reviewing the initial specifications after completing the proof of concept prototype introduced some changes into the system requirements. A complete embedded prototype was designed and developed. Two prototypes were built and underwent basic testing. Firmware implementing device functionality was written in parallel with hardware development. Firmware was initially tested on lab development hardware, and later verified on the alpha prototype. A database was developed on a server that received information from the onboard units through cellular modems. A web client accessed the database and generated simple reports. The next steps in development are to take the working alpha stage prototype and produce a beta product for testing at actual sites where forklifts are operating. Firmware will be tested to ensure that it performs well under the full range of conditions without missing events, even if there is a loss of power to the system. The hour meter counting strategy will be critiqued, with more polling to avoid missed state changes. Parsing of AT commands will be implemented to check for errors, and message structure will be modified and moved to HTTPS. A more robust cell driver will reduce the possibility of dropped messages. Additional features that need to be added include an over-the-air boot loader for remote firmware upgrades, data encryption, and stress testing. Additional hardware testing remains to be done to ensure antenna reliability and suppress a wider range of possible power input spikes. The next stage of development would involve building a beta prototype for limited deployment and long term testing on a forklift truck fleet, with reduced power consumption, smaller board size and reduced component count. A ruggedized case and connector set, and other features for durability. After testing of the beta prototype, additional steps will be required to move into hardware production. Several development steps are also needed for a commercial deployment of the software, including certification for a custom cell implementation. The software, communication system, and database must each be scalable for more features, multiple companies, and large fleets, with security features for each company's information. Any third-party software will have to be licensed. This technology offers promise to deliver reasonable quality operational and condition monitoring information for a variety of distributed, remote assets, such as light commercial vehicles, dedicated process facilities that lack sensing and communication infrastructure (pipelines, railway), and environmental monitoring stations. The architecture even allows for actuation, which would be useful for closing valves, taking samples, and other simple actions.


    Access

    Access via TIB

    Check availability in my library


    Export, share and cite



    Title :

    Design of a modular, cost-effective embedded wireless condition monitoring system for asset management of distributed fleets of light commercial vehicles




    Publication date :

    2014


    Size :

    11 Seiten, Bilder, 26 Quellen


    Type of media :

    Conference paper


    Type of material :

    Print


    Language :

    English