Technology for extending the life cycle of agricultural machinery based on software training

Introduction. The average age of machinery in the agro-industrial and mining complexes often exceeds its service life, which requires maintaining operability and improving technical and economic indicators. Up to 40% of equipment failures are related to the power plant and drive system, where oil plays a key role. Purpose of the research. This study is aimed at solving the most important problem of maintenance of agricultural and industrial machinery in conditions of increased load and aging of equipment. The main focus is on monitoring the lubrication systems of internal combustion engines and transmissions, which account for up to 40% of equipment failures. To improve operational efficiency and reduce wear, it is proposed to develop a telematics system for evaluating the properties of lubricants in real time. Materials and Methods. The research combines the approaches of tribology and data science. Existing real-time oil monitoring methods such as SOAP and JOAP were analyzed and limitations such as cost and data volume were identified. A new hardware and software solution has been developed, including sensors for measuring the acidity of oil and the concentration of solid particles. Using machine learning algorithms, the system predicts optimal oil change intervals and evaluates wear without the need for disassembly. Results. Preliminary results show that the system effectively monitors key properties of the lubricant, including acidity and particle size. Real-time data processing revealed patterns of deterioration in oil quality associated with the degree of wear, which allows timely maintenance. The predictive model, developed on the basis of historical data, provides high accuracy in predicting lubricant characteristics and wear thresholds. Initial modeling suggests that the service life of the equipment may increase by 15–20%, and downtime may be reduced by up to 30%. Conclusions. The developed system offers a comprehensive solution for quality management of lubricants and prevention of critical equipment failures. Thanks to real-time monitoring and machine learning, this approach provides increased reliability and reduces maintenance costs. Future work includes field tests of agricultural and industrial equipment to confirm these findings.