Research On Dynamic Response Characteristics And Control Strategy Of Natural Gas Engine

In solving the problems of energy shortage and environmental pollution,China has paid more and more attention to the development of natural gas.Vehicles and ships are one of the key areas for accelerating the application of natural gas.Therefore,the research on natural gas engine technologies has received much attention;Natural gas engines have poor dynamic response due to slow intake of air and slow flame propagation.Improving dynamic response performance is of great significance for propelling natural gas engine applications.This paper mainly discusses the ways to improve the dynamic response on the control.Firstly,the quasi-dimensional model of natural gas engine is built by using GT-Power software.Based on the model verification,the multi-objective optimization research on the control parameters in the dynamic process is carried out.On the basis of the influence of time and lambda on engine combustion,power,economy and knocking phenomenon,the multi-objective optimization function is constructed with the aim of power and economy,and adding constraints such as exhaust temperature and knock.Based on the experimental design method,the genetic algorithm is used to solve the multi-objective optimization function,and finally the optimized control parameter MAP is obtained,which lays a foundation for the precise control of dynamic process.The results show that the ignition timing and lambda are coupled to the engine’s power and economy effects.The multi-objective optimization study can obtain the optimal values of ignition timing and lambda under the same working conditions.In the dynamic process,the air-fuel ratio and the speed are the main control variables of the natural gas engine.By analyzing the control accuracy and response speed of the air-fuel ratio and speed under dynamic conditions,the air-fuel ratio is controlled by the injected mass per cycle,the speed is controlled by throttle angle,and PID algorithm is used to design a reasonable control framework for speed and air-fuel ratio closed-loop control.Then,the control parameters optimization research is carried out for the start-up process of open-loop control,and the feasibility of closed-loop control of speed and air-fuel ratio is verified during the process of load addition and subtraction.Based on the common control strategies of speed and air-fuel ratio,in order to further improve the dynamic performance,an intelligent control algorithm research is carried out,changing the air-fuel ratio MAP table feed forward into neural network feed forward in order to improve the accuracy of injected mass prediction,and changing the PID control of speed into BP-PID control based on neural network to realize online self-tuning of PID parameters,finally improve dynamic performance of engine.The results show that,when the injected mass per cycle is 78mg/cycle,the ignition timing is-10°CA,and the throttle angle is greater than 12%,the time to start acceleration to idle is the shortest;compared with non-neural network control,the transient speed regulation is reduced and speed recovery time is shortened with the neural network control.