Electrical Fault Classification Strategies for Maintenance Models Using Machine Learning Algorithms

Mining equipment suppliers are investigating new maintenance models for mining truck monitoring and control systems that focus on process efficiency due to the high costs involved. The aim of this article is to generate a predictive maintenance model for electrical faults through machine learning algorithms that predict and classify the severity of faults in harnesses. The study is carried out in the harnesses of the control system of the diesel engine component of a fleet of trucks. Predictive analysis considers two factors: number and trends of failures associated with a proposed classification. Neural networks and decision trees perform better. These results determine a CBM (Condition Based Maintenance) model that makes the maintenance program more efficient. © 2022 IEEE.