Design of an intelligent diagnostic system for powertrains in an electric-drive vehicle
Abstract
This study examines the process of detecting hidden faults and complex relationships in the operation of an internal combustion engine, which cannot always be established in a timely manner using conventional diagnostic approaches. Hidden faults in the fuel system are determined by analyzing deviations between the calculated torque of the internal combustion engine (ICE) and the generator torque under an energy recovery mode. The engine torque is determined by analyzing the generator torque when charging a traction battery in a hybrid vehicle. This approach makes it possible to obtain additional information about the technical condition of the powertrain and increase the reliability of diagnostic conclusions. The results have been used as a component of intelligent diagnostic systems for hybrid vehicles that operate under the mode of continuous monitoring of the technical condition. The architecture of an intelligent system for diagnosing the technical condition of a hybrid powertrain has been designed. The key diagnostic signs of fuel system degradation as part of the integrated vehicle power system are described, which are based on a comparison of the calculated torque of the internal combustion engine and the generator torque under the energy recovery mode. The deviation between these values is used as a diagnostic sign to detect hidden malfunctions of the fuel system. A feature of the proposed solution is a comprehensive approach to assessing the operation of a hybrid powertrain, which takes into account both the technical parameters of its functioning and the economic aspects of vehicle operation. This provides a more objective assessment of the state of the power system and powertrain compared to conventional control methods. The practical implementation of an intelligent diagnostic system involves its integration into the vehicle's on-board information and analytical systems using automated data collection, continuous monitoring, and intelligent information processing