Research On Control Method Of Combustion Process Based On Fuzzy Control Theory Of Diesel Engine

Nowadays,all kinds of fossil energy reserves are declining year by year,and environmental deterioration is a pressing issue that needs to be addressed urgently.Energy conservation and emission reduction are being implemented in different industries.Especially in the road traffic,emissions laws are becoming more stringent.Although new energy vehicles are booming in the automotive field,pure fuel vehicles still dominate according to the proportion of vehicle ownership.Compression-ignition internal combustion engine can have higher compression ratios because they are not limited by knocking,so that it has high thermal efficiency,low fuel consumption,HC,CO and other pollutants emission.So it become the best choice to implement energy conservation and emission reduction in the automotive field.This paper chooses diesel engine,a typical representative of compression-ignition internal combustion engine,as an example to carry out follow-up research.Diesel engine currently has serious problems of NO_x emission and particulate emission,Since the emission characteristics of diesel engine are closely related to its actual combustion process.This paper proposes a control method for the actual combustion process of diesel engines,which is based on the target heat release regularity curves.In the working process of diesel engine,some key control parameters affecting the combustion process may fluctuate around their target values and are not controllable.In order to ensure that the actual heat release regularity curves always fluctuates within the acceptable range around the target heat release regularity curves,it is necessary to analyse in real time whether the impact of the fluctuation of the control parameters on the combustion process is within the acceptable range.If they are outside the acceptable range,the control parameters need to be adjusted in time.If they are within the acceptable range,no adjustment is required to avoid repeated adjustments leading to overshooting of certain parameters,so as to improve the robustness of the entire control system.In order to judge whether the performance indicators meet the requirements of emission regulations during the actual operation of the diesel engine,and at the same time put forward certain requirements on the control system control accuracy,this paper builds a performance open-loop prediction model based on MATLAB/SIMULINK software and GT-Power software,and the main work is as follows.Firstly,using the results of orthogonal tests at different speeds and loads,plot the engine heat release regularity curves under different operating conditions,classified the heat release curves into two categories,namely single-peak and double-peak,based on the characteristics of the heat release regularity curves under different operating conditions,and the actual operating conditions of the engine were classified and the limits of the operating conditions were obtained.Secondly,select one representative heat release regularity curve with a high NO_x emission at 2200r/min and 60% load,and the heat release regularity curve under this condition was reasonably quantified and evaluated,and the evaluation index was quantified for the double-peak heat release regularity curve.By using the onedimensional simulation software GT-Power,the performance indicators such as pressure rise rate,average in-cylinder temperature and fuel consumption rate are obtained by varying the fuel injection time and injection pressure.Gets the quantitative index of heat release regularity curve under corresponding condition.Fitting calculation to obtain the relationship between the quantitative indicators and the performance indicators.Based on the limits for each performance indicator in the emission regulations,the boundary values for pressure rise rate,average in-cylinder temperature and fuel consumption rate are calculated.Use the fitted relationships to derive the target control values for start of combustion,premixed combustion rate,diffusion combustion rate and duration of combustion,as well as the permissible fluctuation range for each parameter,and the concept of target heat release regularity curve is thus proposed.Finally,GT-Power and SIMULINK are used to build an open-loop prediction model of the heat release regularity.Use the single variable principle to investigate the degree of influence on the quantified index when the rail pressure,inlet pressure,inlet temperature and speed fluctuate to different degrees,and this influence is transformed into fuzzy language variables.Use the simulation data to continuously optimize the fuzzy controller until the error is acceptable.Eventually the fuzzy controller is used to establish an open-loop prediction model of the heat release regularity.The results of the study show that(1)the heat release regularity curves under different working conditions can be divided into single-peak heat release regularity curves and double-peak heat release regularity curves according to the presence or absence of "flat-topping period" of the heat release regularity curves.(2)According to the actual fuel mixing and combustion mechanism in the cylinder,the theoretical combustion process of diesel engine is divided into two stages: premixed combustion and diffusion combustion,in which the diffusion combustion stage can be divided into diffusion combustion stable period and diffusion combustion late period according to the different states of the fuel in the combustion chamber.Extract four quantitative indicators from this: start of combustion,premixed combustion rate,diffusion combustion rate and duration of combustion.Under the study conditions when the SOC is not earlier than-3.5°CA,the pressure rise rate will meet the limit of 0.4MPa/(°CA),and to ensure a lower fuel consumption rate,the combustion start point should be controlled in the range of-3.5°CA to-1.5°CA,in summary,the target control value of the combustion start point is-1.5°CA;When the premixed combustion rate is lower than 5.81J/(°CA)~2,the pressure rise rate will meet the limit condition,considering the limit of the average temperature in the cylinder not exceeding 1606 K,the premixed combustion rate should not be higher than 6.12 J/(°CA)~2,considering 5.81 J/(°CA)~2 as the target control value of premixed combustion rate;when the diffusion combustion rate is lower than 3.315 J/(°CA)~2,the average temperature in the cylinder will meet the limit value of 1606 K,in order to take into account the oil consumption rate not to exceed 230 g/(k W·h)limit condition,the diffusion combustion rate should not be lower than 3.259 J/(°CA)~2,finally determined 3.3 J/(°CA)~2 as the target control value of diffusion combustion rate;when the combustion duration is lower than 41.17°CA,the oil consumption rate will meet the 230 g/(k W·h)limit condition,therefore determined41.17°CA.is the target control value for the combustion duration.(3)The output variables of the fuzzy controller are selected as four quantitative indicators,and the output domain of the fuzzy controller is divided into "NN","PN" and "Y" according to whether the output value of each quantitative indicator meets the boundary conditions under the working condition.The endpoint values of the interval of the domain are the same as the boundary values of their regulatory limits,and the affiliation functions are all "bell-type" affiliation functions;The rail pressure,intake pressure,intake temperature and speed fluctuations are the input variables,and the fluctuations are set in the respective domain intervals according to their values in descending order.Seven groups of fluctuations "NB,NM,NS,ZE,PS,PM,PB" are set in the respective domain,and "trapezoidal" and "bilateral Gaussian" are selected as the shape of the subordinate degree function.The optimization and verification of the fuzzy control system is completed by comparing the GT-Power output with the calculation results of the fuzzy control system,make sure the error is within 5%.An open-loop prediction model of the heat release rate was constructed and the accuracy of the performance prediction model was verified by varying the operating parameters within a reasonable range.In order to ensure that the four quantitative indicators meet the boundary conditions,the rail pressure fluctuation should be controlled at[-4MPa,4MPa],the inlet air temperature fluctuation should be controlled at [-5℃,5℃],the inlet air pressure fluctuation should be controlled at [-3k Pa,3k Pa],and the speed fluctuation should be controlled at [-20r/min,20r/min].min],thus putting certain requirements on the control system’s control accuracy.