Which are the four important control loops of an spark ignition (SI) engine?

Transcription

1 Engine Systems (HS 2017) Exercise 1 Topic: Lecture 1 Johannes Ritzmann (jritzman@ethz.ch), Raffi Hedinger (hraffael@ethz.ch); October 13, 2017 Problem 1 (Control Systems) Why do we use control systems in cars? Solution 1 Control systems in cars help reduce the emission of pollutant species and fuel consumption, increase safety, and improve the drivability and comfort of passenger cars. In addition, control systems allow the ECU (Electronic Control Unit) to adjust to changes of certain components such as ageing of the injection valves and to react to environmental disturbances or failures of components. Problem 2 (Control Loops of an SI Engine) Which are the four important control loops of an spark ignition (SI) engine? Solution 2 Fuel injection feedforward loop Air/fuel ratio feedback loop Ignition angle feedforward loop Knock feedback loop Problem 3 (Knocking) What is knocking and how can it be controlled? Solution 3 The phenomenon of knock has been a major limitation for SI engines since the beginning of their evolution. Engine knock has its name from the audible noise that results from autoignitions in the unburned part of the gas in the cylinder that leads to local, very high pressure peaks that can destroy the rim of the piston and other parts in the cylinder. Knocking occurrs mainly in the region of low speeds and high torques. This is because lower speeds give the fuel more time to autoignite and high torques lead to higher pressures and temperatures in the cylinder. In order to avoid knocking, the compression ratio must be kept below a safe value and ignition timing must be optimized off-line as well as on-line. Problem 4 (Ignition Angle) How is the ignition angle controlled? Explain the underlying basic concept.

2 Solution 4 A first optimization takes place during the calibration phase (experiments on engine or chassis dynamometers) of the engine development process. The nominal spark timing data obtained is stored in the ECU. An on-line spark timing control system is required to handle changing fuel qualities and engine characteristics. The key to this component is a knock sensor and the corresponding signal processing unit that monitors the combustion process and detects the onset of knocking. If knocking occurs, the crank angle at which the spark is generated is incrementally shifted towards TDC. (The angle of ignition is before TDC in order to compensate for the delay ignition-to-start-of-combustion in the cylinder and to optimize the produced torque.) Problem 5 (Engine Efficiency) What is the tank-to-wheel efficiency of actual passenger cars? Explain why it remains substantially below the efficiency that could be reached with the engine running at its maximum efficiency. Solution 5 The most severe drawbacks of current SI engines are evident in part-load operating conditions. As Fig. 1 shows, the average efficiency even of modern SI engines remains substantially below their best bsfc 1 values. This is a problem because most passenger cars on the average (and also on the governmental test cycles) utilize less than 10% of the maximum engine power. 2 Not surprisingly, cycle-averaged tank-to-wheel efficiency data of actual passenger cars are between 12% and 18% only. The next step in the development of SI engines therefore will be a substantial improvement of their part-load efficiency. 1.4 Overview of Control Problems in CI Engines 11 p me [bar] η= c m [m/s] Figure Fig. 1: 1.7. Engine Engine map (mean map (mean effective effective pressure pressure versus mean versus piston meanspeed) pistonof speed) an SI engine, of a gray area: modern part-load SI zone, engine, η =const: gray area: iso-efficiency part-load zone, curves. η =const: iso-efficiency curves. For the definition of p me and c m see Sect variable compression ratio engines; and engines with improved thermal management. 1 Brake-specific fuel consumption (usually in g fuel/kwh mechanical work). 2 The maximum engine power is mainly determined by the customer s expectation of acceleration performance and is therefore very much dependent on the vehicle s mass. It is reached in the upper right corner of the engine map. These systems reduce the pumping work required in the gas exchange part of the Otto cycle, reduce mechanical friction, or improve the thermodynamic efficiency in part-load conditions. Another approach to improving part-load efficiency is to include novel 6 7

3 Problem 6 (Control Paths of a CI Engine) Which are the three paths in a compression ignition (CI) engine that are important from a control-engineering point of view? What are the corresponding control actuator(s)? Solution 6 Path Fuel path Air path Exhaust gas recirculation (EGR) path Actuators High pressure fuel pump, injectors Waste-gate, VGT (variable geometry turbocharger) actuator EGR valve Problem 7 (GDI) What are the benefits of gasoline direct injection (GDI)? Solution 7 Since the fuel is injected directly into the combustion chamber, no wall wetting dynamics need to be considered and no fuel needs to be flow past the intake valves allowing more air to be inducted per stoke. Further, the cooling effect of the gasoline evaporation taking place inside the cylinder causes a small increase in the volumetric efficiency. The degrees of freedom with regard to the injection timing and the amount of fuel injected can be used to optimize the combustion in terms of fuel efficiency, exhaust gas emissions, and comfort. Especially the lean combustion mode (using stratified combustion) can be optimized, reducing throttling requirements, increasing the intake pressure and reducing pumping losses. Problem 8 (SI Engine Concepts) Describe the two different types of charge mixing that are used in gasoline engines. Solution 8 Homogeneous charge mode (typically used at higher loads or speeds), with injection starting before air intake, and burning stoichiometric air/fuel mixtures (Port fuel injection (PFI) and direct injection (DI)) Stratified charge mode (used at low to medium loads and low to medium speeds), with late injection and lean air/fuel mixtures (DI only) Problem 9 (Diesel Engine Model) Which additional reservoirs have to be modeled in a Diesel engine system (compared to a gasoline engine system)?

4 Solution 9 Because Diesel engines are almost always supercharged, there are several additional reservoirs to be modeled in a Diesel engine system. In the intake and exhaust manifolds, for instance, not only masses, but also thermal energy is important. Accordingly, two level variables (pressure and temperature) form the output of these blocks. The turbocharger s rotor, which stores kinetic energy, is an additional reservoir. If EGR and intercooling are modeled as well, the cause and effect diagram has a similar, but even more complex, structure. The most important additional variable is the intake gas composition (the ratio between fresh air and burnt gases in the intake). If perfect mixing can be assumed, this leads to only one additional reservoir. Problem 10 (Particulate Matter) What is particulate matter (PM) and how can its amount in the exhaust gas be lowered? Solution 10 Diesel particulate matter consists principally of combustion-generated soot absorbing organic compounds. Lubricating oil contributes to the formation of particulate matter. The amount of particulate matter produced during combustion depends on oxygen availability, spray formation and oxidation conditions towards the end of the combustion process. Early start of injection with its fast, hot and complete combustion produces low amounts of particulate matter. Additionally, with an early start of injection, conditions for soot oxidation are good during the long period of the expansion stroke. Due to the good influence on spray formation, high injection pressures also are beneficial for obtaining low amounts of PM. Unfortunately, the high-pressure fuel pump introduces an additional load and thus reduces the engine s overall efficiency. Problem 11 (PM-NO x Trade-off) Explain the existence of the PM-NO x trade-off.

5 Solution Overview of Control Problems in CI Engines 17 PM [g/kwh] NOx [g/kwh] bsfc [g/kwh] = 800 bar preferred range of operation start of injection [deg CA] = 1100 bar = 800 bar = 800 bar = 1100 bar start of injection [deg CA] = 1100 bar start of injection [deg CA] Figure 2: Fig. Influence Influence of start of injection start of injection bsfcon and bsfc onand theon emission the emission of PMof and PMNO and x, with NO x, with representing the injection pressure. representing the injection pressure. Particulate matter (PM): Early start of injection with its fast, hot and As shown in complete Fig. 2, itcombustion is difficult to produces state clear low control amountsobjectives, of particulate sincematter. nearly Addi- bad effect with onan thearly outputs startof ofinterest. injection, conditions for soot oxidation are every input has a good and ationally, good during the long period of the expansion stroke. Due to the good influence on spray formation, high injection pressures also are beneficial for obtaining low amounts of PM. Unfortunately, the high-pressure fuel pump introduces an additional load and thus reduces the engine s overall Problem 12 (Tendencies in the Influence of Control Inputs) efficiency. What are thethe effects tendencies of of various input-output relations are summarized in Table 1.1. It shows the difficulty of stating clear control objectives, since nearly every input has a good and a bad effect on the outputs of interest. A well-known a) Early method start of forinjection? dealing with the PM-NO x trade-off is to vary the start of injection from early (e.g., 30 degrees before TDC) to late (e.g., eight degrees after TDC). The optimal injection timing is selected as the one where the trade-off curve b) Highcrosses rail pressure? the acceptable emission limits defined by the corresponding emission regulations. This approach, however, does not take into account that bsfc should be as low as possible. c) Pilot injection? d) Smaller variable nozzle turbine (VNT) area? e) Increased exhaust gas recirculation (EGR)?

6 Solution 12 Control Input early start of injection high rail pressure pilot injection(s) smaller VNT area increased EGR Result good bsfc low particulate matter higher NO x increased NO x low PM slightly improved bsfc low noise improved bsfc lower PM higher NO x lower NO x danger of higher PM improved noise equal or slightly increased bsfc Problem 13 (Chemical Properties of Gasoline and Diesel) a) Write down the chemical forumlas of gasoline and Diesel. b) Based upon the formulas derived above and your basic chemical background, can you guess which fuel has a higher boiling temperature? c) Diesel has an ignition temperature of approximately 250 C while the one of gasoline is significantly higher, around 400 C. Combining this knowledge with the differences in the boiling temperatures from above, can you explain why different ignition techniques are used? Solution 13 a) Both gasoline and Diesel basically consist of hydrocarbons. gasoline : C 5 H 12 to C 9 H 20 Diesel : C 14 H 30 to C 18 H 38 It is important to notice that both fuels have a H/C ratio of 2. This value plays a key role when different fuel are compared in terms of pollutant formation and CO 2 production. b) The hydrocarbon chains that form Diesel are longer than those that form gasoline which leads to a larger surface of a single molecule. Accordingly, the van der Waals forces that make different molecules stick together are stronger and therefore the boiling temperature of Diesel is higher: Boiling Temperature of gasoline Boiling Temperature of Diesel : o C : o C Another effect of those forces is that Diesel has a higher density ( 0.85 kg ) than gasoline dm 3 ( 0.75 kg ). Because the lower heating values of both fuels are almost identical ( dm MJ kg ), Diesel has an energy density per liter that is 13% higher than the one of gasoline. This is something that must be kept in mind when the fuel consumptions of cars with different engine (SI/CI) types are compared.

7 c) Due to its relatively low ignition temperature, Diesel is well suited for the use in CI engines. Because of its relatively high boiling temperature, it can t be used in SI engine applications, where the fuel usually has to evaporate and mix with the air. Looking at the properties of gasoline, the opposite holds. It evaporates easily and due to the high ignition temperature, the knocking phenomenon can be kept under control relatively easy. Problem 14 (TPU) What is the job of the time processing unit (TPU)? Solution 14 The TPU synchronizes the engine control commands with the reciprocating action of the engine.