Optimization of SI and CI engine control strategies via integrated simulation of combustion and turbocharging

Combustion engines have been for a long time the most important prime mover for transportationglobally. A combustion engine is simple in its nature; a mix of fuel and air is combusted, and workis produced in the operating cycle. The amount of combusted air and fuel controls the amount ofwork the engine produces.The engine work has to overcome friction and pumping losses, and a smaller engine has smallerlosses and is therefore more efficient. Increasing engine efficiency in this way is commonly referredto as downsizing. Downsizing has an important disadvantage; a smaller engine cannot take in asmuch air and fuel as a larger one, and is therefore less powerful, which can lead to less customeracceptance.By increasing the charge density the smaller engine can be given the power of a larger engine,and regain customer acceptance. A number of charging systems can be used for automotiveapplication, e.g. supercharging, pressure wave charging or turbocharging. Turbocharging hasbecome the most commonly used charging system, since it is a reliable and robust system, thatutilizes some of the energy in exhaust gas, otherwise lost to the surroundings.There are however some drawbacks and limits of a turbocharger. The compressor of a singlestage turbo system is sized after the maximum engine power, which is tightly coupled to themaximum mass flow. The mass flow range of a compressor is limited, which imposes limits on thepressure build up for small mass flows and thereby engine torque at low engine speed. Further, aturbo needs to spin with high rotational speed to increase air density, and due to the turbo inertia ittakes time to spin up the turbo. This means that the torque response of a turbocharged engine isslower than an equally powerful naturally aspirated engine, which also lead to less customeracceptanceA two stage turbo system combines two different sized turbo units, where the low mass flowrange of the smaller unit, means that pressure can be increased for smaller mass flows. Further, dueto the smaller inertia of the smaller unit, it can be spun up faster and thereby speed up the torqueresponse of the engine. The smaller unit can then be bypassed for larger mass flows, where insteadthe larger turbo unit is used to supply the charge density needed.In the dissertation, the value of engine system modeling has been discussed. It was shown howmodeling in-cylinder processes and turbocharger can aid the development of the control strategiessaving time and money efforts. All the developed models were experimentally validated and appliedfor optimization analysis or real-time control.Particularly the model based optimization of the engine control variables of an automotiveturbocharged Diesel engine has been presented. The model structure is based on a hybrid approach,with a predictive multi-zone model for the simulation of in-cylinder processes (i.e. combustion andemissions formation) integrated with a control-oriented turbocharger model to predictintake/exhaust processes. Model accuracy was tested via comparison between measured andsimulated in-cylinder pressure and engine exhaust temperature on a wide set of experimental data,measured at the test bench. Validation results exhibit a correlation index R2 equal to 0.995 and0.996 for IMEP and exhaust temperature, respectively. The optimization analysis was aimed atminimizing NO emissions in four steady state engine operating conditions, selected among those ofinterest for the ECE/EUDC test driving cycle. Constraints were introduced to prevent from increaseof soot emissions and low exhaust temperature which would have a negative impact on theefficiency of the after-treatment devices. The optimization results evidence a significant reductionof engine NO emissions by means of increased EGR rate and earlier main fuel injection.A model-based optimization was also applied for a CNG heavy-duty engine, equipped withturbocharger and EGR. The optimization analysis was addressed to design the set-points of enginecontrol variables, following the implementation of an EGR system aimed at reducing the in-cylindertemperature and preventing from the thermal stress of engine components (i.e. head and valves). Aco-simulation analysis was carried out by coupling a 1-D engine commercial code with a classicalconstrained optimization algorithm. The 1-D model accounts for intake and exhaust gas flowarrangement, comprehensive of EGR system and turbocharger, while an empirical formulationbased on the classical Wiebe function was implemented to simulate the combustion process. Anintensive identification analysis was performed to correlate Wiebe model parameters to engineoperation and guarantee model accuracy and generalization even in case of high EGR rate. 1-Dmodel and identification results were successfully validated against a wide set of experimental data,measured on the test bench. The results of the optimization analysis, aimed at minimizing fuelconsumption while preventing from thermal stress, showed an increase of fuel economy up to 4.5%and a reduction of the thermal load below the imposed threshold, against five engine operatingconditions selected among the most critical of the reference European Transient Cycle (ETC).Particularly, the effectiveness of the co-simulation analysis is evidenced in pursuing the conflictinggoal of optimizing engine control while reducing the recourse to time consuming and expensiveexperiments at the test bed. This latter point is becoming more and more critical as the number ofcontrol variables is increasing with engine complexity.Both the presented optimization analyses evidenced the key-role of the turbocharger to face withenergy and emissions issues. Particularly the impact of the turbocharger management via wastegateor VGT control was evidenced. Indeed, by acting on these components, the amount of exhaust gasesevolving in the turbine can be managed thus regulating the supercharging degree and the boostpressure. This allows keeping the throttle valve fully open with significant decrease of pumpinglosses. The wastegate position is defined by a pneumatic actuator in which the pressure is regulatedby a solenoid valve fed by a PWM signal. The drawback of this system is the dependence of thePWN signal, and afterwards of the performance, from the system supply voltage. During the thesisthe development of a wastegate actuator model was carried out in order to compensate the actuatorPWM signal to improve boost pressure control. The compressible flow equations were found to besufficient to describe the actuator system mass flow and both discharge coefficient and staticactuator chamber pressure were modeled using polynomials in PWM signal. Furthermore a simplefriction model was implemented to simulate the actuator system. The boost pressure controllerbased on the developed compensator has shown to give limited undershoot and overshoot and isfurther able to reject the disturbance in supply voltage. The compensator was incorporated into aboost pressure controller and the complete control system has shown to reject system voltagevariations and perform good boost pressure control in both simulations analyses and experimentaltests on the engine test stand. Model simulations evidenced the need to ensure low enough vacuumpressure to enable fully closed and open actuator while a switch type controller was proved to besufficient for vacuum tank pressure control. [edited by Author]