Monitoring internal combustion engine wear from engine oil using linear regression

This paper deals with the wear monitoring of an internal combustion engine in a heavy combat vehicle. For the purpose of the study, engine oil samples were collected and then checked using a technique called atomic emission spectroscopy (AES). This is one of the modern ways of measuring metallic elements in oil, including iron. The amount of iron in engine oil is expressed in parts per million (ppm). This indicates how much the metal parts of an internal combustion engine are worn. Engine oil samples from 15 combat vehicles over several years were used for the experiment. The aim of this work was to predict the iron concentration in engine oil using a suitable mathematical method. Robust linear regression with MM estimator is used in this paper. It is a method that combines high robustness to outliers and minimizes a weighted loss function using an iterative optimization approach. The second method used is the least squares method, which is a mathematical technique that works by minimizing the sum of the squares of the differences between the actual values and the values predicted by the model.