Binary particle swarm optimization-based T-S fuzzy predictive controller for nonlinear automotive application

In this paper, a robust evolutionary computing-assisted Takagi-Sugeno fuzzy predictive controller (T-S FPC) has been developed for nonlinear vehicle fuel injection and emission control. To strengthen the performance of T-S FPC, we have applied an enhanced evolutionary computing algorithm named binary particle swarm optimization (BPSO) that achieves optimal control variable by performing minimization of the cost function iteratively, where the cost function signifies the mean square error between reference data and the actual predicted data. To examine the efficacy of the proposed system, a case study was performed for an automotive vehicle to control its fuel injection, throttle angle, and emission control under nonlinear conditions. The simulation results affirmed that the proposed BPSO-based T-S FPC model exhibits optimal performance by achieving target performance with low mean square error between expected functions and prediction outcomes. The efficiency of the proposed BPSO T-S FPC model enables it to be used for online nonlinear control purposes for any type of the vehicle systems.