Session: 15-05-01: Models and Methods for Probabilistic Risk Assessment

Paper Number: 166691

Mining Dumper Maintenance Schedule Estimation Using Weibull Model and Particle Filter 

Mining dumper, a heavy earth moving machinery, plays a critical role in the transportation of materials in open-pit coal mining operations from loading sites to dumping sites. This machinery operates under extreme environmental and mechanical conditions such as high loads, abrasive materials, and harsh terrains. Continuous operation of such equipment is essential for maintaining high productivity and minimizing operational costs. However, frequent exposure to heavy stress leads to mechanical wear, component degradation, and eventual system failures, resulting in unplanned downtime. Failure-induced downtime in mining machinery significantly disrupts production schedules, increases maintenance costs, and impacts overall mine efficiency. Breakdowns can be caused by various factors, including excessive vibrations. Predictive maintenance strategies, leveraging vibration-based condition monitoring, play a crucial role in mitigating these failures by scheduling timely maintenance. By integrating predictive analytics with real-time data from accelerometers, mining operations can minimize unexpected downtimes, extend machinery lifespan, and optimize resource utilization. Vibration analysis has been widely used in machine condition monitoring, but its application to large mining equipment is limited for maintenance scheduling. This study introduces a method for predicting dumper degradation using vibration signals recorded from an accelerometer mounted on the right frame over a period of time covering different operations performed by dumper on several mine terrains, particle filter and Weibull failure rate function (WFRF). The recorded vibration signals are pre-processed to remove noise and extract meaningful features for different operation of dumper. The Root Mean Square (RMS) value of the vibration signal is computed to quantify the dumper’s operational health. The RMS values are segmented for different operations and RMS values selected for idling operation using activity tracking methods. An increasing RMS trend signifies progressive component deterioration, highlighting the need for timely maintenance interventions performed by maintenance personnel to prevent unexpected failures. To accurately model and track vibration trends, a particle filter is applied. This Bayesian filtering technique estimates the state of the system by representing uncertainty in the vibration signal through a set of weighted particles. The WFRF is used to characterize the degradation pattern of the mining dumper. The Weibull distribution parameters are estimated from historical RMS data, allowing the prediction of the failure rate at different time intervals. This combined model (particle filter and WFRF) helps establish a probabilistic threshold for maintenance scheduling. The estimated vibration trend and probability density function (PDF failure probability) are combined to determine the next schedule of maintenance based on time required to reach critical RMS threshold.

Presenting Author: Nagesh Dewangan Indian Institute of Technology Kharagpur

Presenting Author Biography: Nagesh Dewangan is currently a Ph.D. student in the Acoustics and Condition Monitoring Laboratory, Mechanical Engineering Department, Indian Institute of Technology Kharagpur, India. He received his B.E. degree in Mechanical Engineering from Bhilai Institute of Technology Durg, India, in 2016, and M.Tech. degree in Maintenance Engineering & Tribology from Indian Institute of Technology Dhanbad, India, in 2019. His research interests are in the areas of Machinery, Condition Monitoring, Vibration, Sound, Current, Signal Processing, Fault Diagnosis, Fault Prognosis, Real-time Application, Internet of Things, Machine Learning, and Deep Learning for industry-oriented Product Design and Development.

Authors:

Nagesh Dewangan Indian Institute of Technology Kharagpur
Amiya Ranjan Mohanty Indian Institute of Technology Kharagpur