Vehicle maintenance management based on machine learning in agricultural tractor engines

• DOI: 10.15446/dyna.v90n225.103612
• Publication Date: 2023-03-31T17:06:33Z
• Authors: Carlos Nolasco Ma...; Cesar Fabricio Mo...; Erik Paul Hernánd...; Ignacio Bayardo B...
• Keywords: Tractor; Agricultural machinery

The objective of this work is to use the autonomous learning methodology as a tool in vehicle maintenance management. In obtaining data, faults fuel supply system have been simulated, causing anomalies combustion process that are easily detectable by vibrations obtained from sensor engine an agricultural tractor. To train classification algorithm, 4 states were used: BE (optimal state), MEF1, MEF2, MEF3 (simulated failures). applied supervised type, where samples initially characterized and labeled create database for execution training. results show training carried out within algorithm has efficiency greater than 90%, which indicates method used applicable management predict failures operation.