Fatigue Damage Monitoring for Mining Vehicles using Data Driven Models

The life and condition of a mine truck frame are related to how the machine is used. Damage from stress cycles accumulated over time, measurements throughout needed monitor condition. This results in high demands on durability sensors, especially harsh mining application. To make monitoring system cheap robust, sensors already available vehicles preferred rather than additional strain gauges. main question this work whether existing on-board can give required information estimate signals calculate damage frame. Model complexity requirements selection also considered. A final be used for prognostics increase reliability. investigation performed using large data set two operating real applications. Coherence analysis, ARX-models, rain flow counting techniques show that low number like load cells, damper cylinder positions, angle transducers enough recreate some measured. models significant differences usage by different operators, its effect damage.