Session: 14-06-01: Machine Learning and Deep Learning in Safety, Reliability, and Maintenance I

Paper Number: 150898

150898 - Predictive Modeling and Data Analytics for Preventing Excavator Stalling to Enhance Safety and Reliability in Mining Operations 

In mining and mechanical engineering, operational efficiency and safety are paramount concerns. One critical issue that can lead to significant downtime and operational inefficiencies is stalling in excavators, which occurs when the machine ceases to operate due to engine overload. This phenomenon not only disrupts the mining process but also poses safety risks to operators and can lead to costly delays. The motivation behind this research is to leverage data analytics and predictive modeling to understand and mitigate the occurrence of stalling in excavators, thereby enhancing both safety and reliability in operations.

The contribution of this work lies in its innovative approach to addressing a common yet challenging problem in engineering through the application of advanced data analytics techniques. By analyzing sensor data collected from excavators, this study aims to develop predictive models that can identify the conditions leading to stalling events. These insights can then be used to implement proactive measures, reducing the likelihood of stalling and its associated risks.

The methodology used in this research involves several key steps. First, extensive data collection was carried out, focusing on sensor data from excavators operating in a mining environment. This data includes parameters such as engine load, hydraulic pressure, and operational cycles, collected over a significant period in October 2023. The collected data was then preprocessed to remove noise and irrelevant information, ensuring high-quality inputs for analysis.

Following data preprocessing, advanced data analytics techniques were employed to uncover patterns and correlations within the data. Machine learning algorithms, including decision trees, random forests, and support vector machines, were applied to identify the key indicators of stalling events. These models were trained on historical data and validated using cross-validation techniques to ensure their accuracy and reliability.

Preliminary results from the data analysis have revealed several critical factors that contribute to stalling in excavators. High engine load combined with specific hydraulic pressure thresholds were identified as primary indicators. Additionally, operational patterns, such as extended periods of high-intensity use, were found to significantly increase the likelihood of stalling. These findings have been instrumental in developing a predictive model that can forecast stalling events with a high degree of accuracy.

The conclusions drawn from this study highlight the transformative potential of data analytics in mining engineering. By leveraging sensor data and advanced machine learning techniques, it is possible to gain a deeper understanding of operational issues and develop effective solutions to mitigate them. The predictive model developed in this research not only enhances the reliability of excavators but also contributes to the overall safety of mining operations by allowing for proactive intervention before stalling occurs.

In summary, this research demonstrates the importance of integrating data analytics and predictive modeling into engineering practices. The findings provide a robust foundation for future work aimed at further improving the safety and efficiency of mining operations. By continuing to explore and refine these techniques, the mining industry can achieve significant advancements in both operational reliability and risk management.

Presenting Author: Mateo Montenegro Defaz Missouri University of Science and Technology

Presenting Author Biography: Mateo Montenegro is currently pursuing his MS (Master of Science) degree in Mining and Mineral Engineering at Missouri S&T, where he also works as a Research Assistant, sponsored by ESCO. He earned his bachelor's degree in mechanical engineering with a Minor in Civil Engineering from USFQ in Ecuador, graduating with Summa Cum Laude distinction. Since his undergraduate studies, Mateo has been an active member of research teams, contributing to various projects and publications. He is particularly passionate about Mining Engineering and its critical role in modern society.

Authors:

Mateo Montenegro Defaz Missouri University of Science and Technology
Kwame Awuah-Offei Missouri University of Science and Technology