A Study on the Anomaly Detection of Engine Clutch Engagement/Disengagement Using Machine Learning for Transmission Mounted Electric Drive Type Hybrid Electric Vehicles

Transmission mounted electric drive type hybrid electric vehicles (HEVs) engage/disengage an engine clutch when EV↔HEV mode transitions occur. If this engine clutch is not adequately engaged or disengaged, driving power is not transmitted correctly. Therefore, it is required to verify whether engine clutch engagement/disengagement operates normally in the vehicle development process. This paper studied machine learning-based methods for detecting anomalies in the engine clutch engagement/disengagement process. We trained the various models based on multi-layer perceptron (MLP), long short-term memory (LSTM), convolutional neural network (CNN), and one-class support vector machine (one-class SVM) with the actual vehicle test data and compared their results. The test results showed the one-class SVM-based models have the highest anomaly detection performance. Additionally, we found that configuring the training architecture to determine normal/anomaly by data instance and conducting one-class classification is proper for detecting anomalies in the target data.

Dynamic analysis of small gasoline car model powertrain using MATLAB / SIMDRIVELINE

The powertrain model of a vehicle using a small gasoline engine is designed based on the analysis results of the Matlab/Simdriveline simulation model. In which, the vehicle's powertrain model parts include: engine, clutch, gearbox, differential and wheels, and overall vehicle modeled by Matlab/ Simdriveline. Mathematical basis of the corresponding models for systems or components are used to build simulated models for the entire vehicle's powertrain system. The input parameters for the simulation problem include parameters of the size, mass, structural and technical parameters of each system such as transmission ratio, power, velocity, efficiency, determined based on actual vehicle model and empirical calculations. The simulation calculation process is done on the basis of the variation of the engine power, from which, the corresponding input and output kinetic and dynamic parameters of each system in the powertrain system consists of clutch, gearbox, differential, are obtained in the time domain. The results of simulation calculation of the kinematics and dynamics of each system in the vehicle's powertrain are analyzed as a basis for design improvement to improve the dynamic performance of the vehicle model.

Adaptive Engine Torque Compensation with Driveline Model

Engine net torque is the total torque generated by the engine side, and includes the fuel combustion torque, the friction torque, and additionally the starter motor torque in case of hybrid vehicles. The engine net torque is utilized to control powertrain items such as the engine itself, the transmission clutch, also the engine clutch, and it must be accurate for the precise powertrain control. However, this net torque can vary with the engine operating conditions like the engine wear, the changes of the atmospheric pressure and the friction torque. Thus, this paper proposes the adaptive engine net torque compensator using driveline model which can cope with the net torque change according to engine operating conditions. The adaptive compensator was applied on the parallel hybrid vehicle and investigated via MATLAB Simcape Driveline simulation.