Author: Qо‘ziyev, Nodir Murodullayevich; Mengliqulov, Jonibek Ramazon o‘g‘li

Annotation: In modern vehicles, the process of assessing the technical condition of the engine is becoming increasingly complex, rendering traditional diagnostic methods insufficiently effective. Specifically, the diagnostic trouble codes (DTC) identified through the OBD-II system only indicate pre-defined faults, limiting the ability to detect latent or early-stage engine malfunctions. This study proposes a diagnostic approach based on OBD-II sensor data for analyzing engine faults using artificial intelligence. During the study, key technical parameters such as engine RPM, coolant temperature, air temperature, fuel consumption, and vehicle speed were analyzed. After initial data preprocessing, engine faults were detected using a model based on artificial neural networks. Experimental results demonstrate that the proposed AI-based diagnostic algorithm achieves higher accuracy and reliability compared to traditional methods. The study findings are significant for the early detection of engine faults and for improving the efficiency of vehicle maintenance.

Keywords: OBD-II, artificial intelligence, engine diagnostics, fault detection, sensor data.

Pages in journal: 497 - 503