Combustion timing control of HCCI engine based on FNN-PID and black-box model

In order to obtain benefits of Homogenous Charge Compression Ignition (HCCI) engine, the most significant challenge is to control the combustion timing. Therefore, this study develops a black-box model based on BP neural network to estimate the combustion timing, where speed of engine, fuel equivalence ratio, temperature and pressure of gas at intake manifold, and intake valve closing timing are chosen as the system inputs. Then, a PID controller, whose parameters are adjusted by fuzzy neural network, is designed to control combustion timing by controlling the intake valve closing timing. Results from numerical experiments show that the proposed model could estimate combustion timing accurately with the error under 0.4(CAD); and meanwhile, compared with the traditional PID controller, the proposed controller has a better performance with respect to the overshoot, setting time, and the robustness.