DATA SEGMENTATION ALGORITHM FOR FUEL LEVEL MONITORING

Abstract

The article presents an algorithm for the automatic segmentation of signals from vehicle fuel level sensors based on derivative analysis and adaptive detection of changes in signal behavior. The method allows for sufficiently accurate identification of homogeneous segments, which serves as a basis for further noise characteristic analysis. The algorithm ensures an acceptable level of accuracy and demonstrates robustness against short-term anomalies and typical interferences commonly encountered in real operating conditions. A comparison of the segmentation results with empirical estimates confirmed the effectiveness and reliability of the proposed approach. The solution is sufficiently universal, making it suitable for application across various types of data acquisition systems, including those operating under complex and demanding conditions typical of modern vehicles.