The future of mobile working machines will see increasing number of assistance systems while it gets more and more difficult to engage skilled machine operators able to check on the condition of machines. So, the need for machines monitoring their current state and deciding on necessary maintenance will increase. This contribution seeks to add a new approach in the field of Predictive Maintenance (PM) strategies working with the example of checking the clearance state of friction bearing joints on excavators. A great challenge in using Machine Learning (ML) for that purpose is to have enough meaningful data. The data islands problem often prevents solutions to be derived by not having enough data of certain use cases.

    By the use of Convolutional Neural Network Auto Encoders, it is possible to deal with the problem of data islands so that enough data of the bearing joint states can be generated using the signals of Inertial Measurement Units as input. In this contribution the monitoring of the clearance of the main friction bearing of the tip of an excavator was chosen as use case with different sensor positions to be investigated. The clearance state can be estimated generally well with the best results obtained when the sensor is placed in near environment of the bearing.

    In future works the technique is intended to be used on more use cases like transferring failure signals between derivatives of machines so that learning from failure is enabled within a fleet of machines.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    The (important) Role of Digital Twins and AI Methods within Predictive Maintenance Strategies


    Contributors:

    Conference:

    International Commercial Vehicle Technology Symposium ; 2024 ; Kaiserslautern, Deutschland March 13, 2024 - March 14, 2024


    Published in:

    Commercial Vehicle Technology 2024 ; Chapter : 19 ; 322-341


    Publication date :

    2024-10-02


    Size :

    20 pages




    Type of media :

    Article/Chapter (Book)


    Type of material :

    Electronic Resource


    Language :

    German




    Digital Twins for Automotive Predictive Maintenance

    Mohanraj, E. / N, Eniyavan / S, Sidarth et al. | IEEE | 2024


    Predictive Digital Twins for Autonomous Ships*

    Hasan, Agus / Widyotriatmo, Augie / Fagerhaug, Eirik et al. | IEEE | 2023


    Digital Twins in Operation and Maintenance(O&P)

    Wang, Wenjuan / Zaheer, Qasim / Qiu, Shi et al. | Springer Verlag | 2023



    Digital Twins

    Brucherseifer, Eva / Fay, Alexander | German Aerospace Center (DLR) | 2021

    Free access