Research On Control Method Of Two-stage Turbocharging System For Diesel Engine Working At Different Altitudes

Chinese geographical environment is complex and diverse.Mountainous and plateau topography account for a large proportion of the country's land area.The deterioration of the intake of diesel engines in the plateau region has greatly affected its performance.Due to the drop of the air density in highland,the diesel engine may not work normally.In recent years,with the development of engineering transportation,national defense equipment,there was an objective request for enhancing the diesel engine adaptability at different altitudes.The two-stage turbocharging system can provide high intake pressure,and its high-stage bypass valve allows the pressure ratio to be adjusted according to the change of operation conditions.It has been proved to be a good way to raise the engine power at plateau environment.In this dissertation,the two-stage turbocharging system was taken as the research object,and the control method of the two-stage turbocharging system for diesel engine under variable altitudes was studied.The target was to improve the performance of diesel engine at variable altitudes.The detailed dynamic model was established and the parameter influence law was studied.Based on the dynamic characteristics of the research object,a dynamic simulation model of the two-stage turbocharged diesel engine was built and the model was verified by experimental data.Based on steady state model simulation,the influence laws of the bypass valve opening,injection advance angle and circulating fuel injection amount on engine performance were analyzed,and factors which affected the intake pressure during the engine dynamic process were analyzed.All these works laid the foundations for the following study.In order to enhance the performance of the two-stage turbocharged diesel engine working at different altitudes,a fuzzy PID optimization approach for calibrating the fuel injection parameters and bypass valve opening at the full load operation area was proposed.Firstly,in the full load area,the steady-state process control parameters(opening degree of the turbine bypass valve)and the oil supply parameters(circulating fuel injection amount and injection advance angle)of the two-stage system were optimally matched.The bench test was carried out according to the optimization results,and the experiment results showed that the rate power of the diesel engine working at 4500m can reach 266k W,which verified the effectiveness of the fuzzy optimization algorithm.Secondly,in the partial load area,the influence of different bypass valve opening on diesel engine power was analyzed by simulation,and the optimum bypass valve opening degree and optimal boost pressure of partial load were selected.Finally,linear interpolation was used to obtain the optimal bypass valve opening and intake pressure map under full working conditions.The bypass valve opening map was chosen as the output of the open-loop control.The intake pressure map was chosen as the control target of the closed-loop control.Controlling of the actuator of the regulating executor for the two-stage turbocharging system was studied.The test bench and data collection system were built.The linear parameter variation transfer function of pressure regulating electro valve was built based on the parameter identification experiments.The dynamic system model of electro valve,the actuator of the regulating executor and the whole adjustable actuator were established.To address the PID(Proportion Integral Differential)shortcomings of control performance and robust stability on variable parameter problem,a mixed H₂/H_?controller was designed based on the system model,and the control effect was analyzed through simulation.The design of closed-loop control algorithm for intake pressure of two-stage turbocharging system was studied.To decrease the large overshoot of the intake pressure due to the rapid cycle fuel injection change,the optimization results got from the steady state simulation were chosen as the closed-loop control objectives.Two closed-loop control algorithms were designed following.Base on the traditional gain schedule PID algorithm,the P gain and I gain were designed not only according to the variation of the working conditions,but also according to the variation of the cycle fuel injection mass.Simulation results showed that the improved PID algorithm can reduce the overshoot by 56%.The Neural Network based model predictive controller was designed based on the external regression neural network.The identification data was got through the dynamic condition identification of different altitudes.The predictive model was built up by using a regression network with input and output delays.The objective function was calculated by using Newton Raphson Iteration.Simulation results showed that the predictive control method can reduce the overshoot by 33%.This method can not only help solving the multi-input and single-output intake pressure control problems,but also reduce the complexity of the controller design process.