Gray based Fuzzy Gain-Scheduling PID Controller Design for AirFuel System Under Variable Engine Operating Conditions

Abstract

In this study, the problem of regulation of air-fuel ratio (AFR) in gasoline engines under different engine operatingconditions is discussed. Firstly, the mean value mathematical model of the AFR system has been created. Then,two different approaches named with classical proportional-integral-derivative (PID) and a fuzzy logic gainscheduling PID controller combined with gray system modelling approach (Gray GS-PID)have been used toimprove the performance of the engine to monitor stoichiometric conditions. The parameters of classical PIDparameters are determined by the pattern search algorithm. The design procedures for both controllers havebeen presented in detail. In order to evaluate the performance analysis for both of the proposed controllers,variable conditions were established based on engine speed and throttle opening ratios in the US06 and UDDSdriving conditions and validated by simulation results. According to the results, Gray GS_PID is more powerfulthan optimally adjusted PID in terms of reducing the amount of deviation of AFR from stoichiometric valueunder variable engine operating conditions. The most important contribution of this study is that, unlikeconventional AFR regulation, the prediction of future error value relative to the previous AFR error values usingthe gray prediction algorithm, and the design of the control algorithm that determines the control action for thenext step depending on the predicted error value before the error occurs and sets the gain parameters.