Research On Control Strategy Of Dual Fuel Engine Based On Torque

Marine diesel/natural gas dual-fuel engines have great development prospects in controlling pollutant emissions and reducing ship environmental pollution.However,due to the combustion characteristics of natural gas and the delay characteristics of the supply system,the control effect of marine dual-fuel engines is poor and the dynamic response is slow,the speed is unstable.Starting from the engine control strategy,this paper designs a torque-based engine control strategy to directly control the engine power output system to achieve more accurate and efficient dual-fuel engine performance optimization and operation control.First,a control-oriented diesel/natural gas dual-fuel engine simulation model is built.The model adopts the idea of modular modeling,including common rail fuel system,cylinder system,gas supply and intake and exhaust systems.The establishment of the model comprehensively uses empirical formulas with MAP diagram.Secondly,in order to solve the problem of complex operating conditions and difficult load measurement of marine engines,the research on engine torque demand estimation is carried out,and the torque estimator based on sliding mode theory and extreme learning machine(ELM)is respectively proposed to calculate the engine load torque.predict.Comparing the prediction effects of the two torque estimators,since the estimation effect of the sliding mode torque estimator depends on the selection of the K value,it cannot be guaranteed to be applicable to all engine operating conditions,while the ELM torque estimator can predict the load torque more accurately,and can be used for the development of engine control strategies.The framework of the control strategy is determined by analyzing the control requirements of the dual-fuel engine,and researches are carried out from the aspects of torque-based speed closed-loop,rail pressure closed-loop,air-fuel ratio open-loop control,and fuel management.The engine feedback algorithm based on Elman neural network and the engine feedforward algorithm based on torque estimator are designed to optimize the engine speed control;the rail pressure closed-loop algorithm of PCV/VCV dual valve coordinated control is designed to optimize the rail pressure control effect;The engine fuel management strategy is designed to shorten the switching time and reduce the speed fluctuation;and the control strategy is verified offline by using the engine simulation model.Finally,the control strategy is verified on a semi-physical simulation platform,and a rapid control prototype system based on c RIO and a real-time simulation platform based on NI PXIe are built.In the semi-physical simulation environment,the steady-state fluctuation of the engine is in line with expectations,the speed error is within 2.5% under transient conditions,and the load response is greatly improved in dual-fuel mode;the steady-state fluctuation of rail pressure is significantly improved;the fuel switching process is within the specified time It is completed within and the speed fluctuation is extremely small.