Research On Combustion Timing Control Methods For HCCI Engine Under Complex Working Conditions

The continuous increase in the number of vehicles worldwide has caused more and more serious environmental pollution and energy loss.Research and improvement of automobile engines are needed to alleviate the huge pressure of automobiles on the environment and energy.Homogeneous Charge Compression Ignition(HCCI)engine combustion technology combines the characteristics of diesel and gasoline engines.The mixture of ordinary fuel and air is used as fuel for HCCI engines and is ignited by compression,which has the advantages of high efficiency and low emissions.HCCI engine technology is a key research direction in the future research field of internal combustion engines.Due to the lack of a direct ignition mechanism for HCCI engines,the combustion timing of the fuel in the cylinder is completely determined by the chemical components in the cylinder,which makes it impossible to directly measure the combustion timing.Too early or too late combustion will affect the engine's emissions performance,so accurate estimation and control of combustion timing is the key to HCCI engines.In particular,engines often work under complex operating conditions,which makes the estimation and control of combustion timing more difficult.Therefore,it is of great research significance to accurately estimate and control the combustion timing under complex conditions.In this thesis,a variety of estimation models and control algorithms are designed for HCCI engine combustion timing estimation and control under complex conditions.The main work of this thesis includes:1.The research about combustion timing estimation model for HCCI engine under complex conditionsAiming at the problem that the combustion timing of HCCI engine cannot be measured under complex conditions.In this thesis,the combustion process of the engine is first analyzed.A non-linear mean model is established to estimate the combustion timing.Considering the accuracy of the model estimation,a gray box model is established based on the nonlinear mean model and neural network to improve the estimation accuracy.In view of the cyclic combustion characteristics of the engine and the difficulty of establishing a nonlinear mean model under complex operating conditions,an Elman neural network black box model is designed.The experimental results show that the Elman neural network black box model has the best estimation of HCCI engine combustion timing under complex conditions.2.Combustion timing controller design of HCCI engine under complex conditionsAiming at the combustion timing control problem under complex working conditions,neural network predictive control algorithm is designed to control the combustion timing.Model predictive control is suitable for non-linear systems.The neural network has the ability to fit nonlinear functions.Combining the two forms a neural network predictive control to achieve precise control of combustion timing.Considering that using a neural network to predict such a feedback controller has a certain lag,a B-spline fuzzy neural network control algorithm is designed as a feedforward controller.Simulation experiments show that the designed composite controller can achieve precise control of HCCI engine combustion timing under complex operating conditions,and has the characteristics of strong robustness and fast response.3.Consistency control of multi-cylinder HCCI engine combustion timingFrequent changes in engine speed under complex operating conditions cause thermal disturbances in the cylinder of the HCCI engine.This leads to inconsistent combustion timing in each cylinder of multi-cylinder HCCI engines.After solving the combustion timing control problem of single-cylinder HCCI engine,the PID controller is selected to control the combustion timing of the multi-cylinder HCCI engine to ensure that the combustion timing of the multi-cylinder is consistent.Considering that the traditional PID control parameters cannot be adaptively adjusted,this thesis uses a neural network and gradient descent method to dynamically adjust the PID parameters.Simulation experiments show that the designed PID controller can control the combustion timing of the multi-cylinder HCCI engine after the parameters of the neural network are set.