Method for controlling an internal combustion engine with stratified combustion to reduce knocking noise, and corresponding powertrain

Abstract

Method for controlling an internal combustion engine, comprising cylinders designed to enable stratified combustion to take place, comprising the following steps for each combustion cycle: • a. determination (31) of the number of injection and ignition patterns to be performed, • b. determination (32) of the quantity of fuel to be injected at each injection by dividing the quantity of fuel to be injected in the combustion cycle by the number of patterns to be performed, • c. determination (33) of the time of each pattern as a function of crankshaft angle, said time of each pattern being between intake bottom dead center and combustion top dead center of the combustion cycle, • d. control (34) of the fuel injection circuit and ignition means to produce patterns as a function of the crankshaft angle, the quantity of fuel to be injected per pattern and the timing of each pattern.

Description

[0001] DESCRIPTION

[0002] Method for controlling an internal combustion engine with stratified combustion to reduce knocking noise, and corresponding powertrain

[0003] Technical area

[0004] The technical field of the invention is direct-injection spark-ignition internal combustion engines.

[0005] Prior Art

[0006] Figure [Fig 1 ] illustrates a spark-ignition internal combustion engine 1 comprising at least three cylinders 2 connected by distributor, or intake manifold 3, to an intake circuit and by an exhaust manifold 4 to an exhaust circuit. In this example, the engine is supercharged by a turbocharger.

[0007] Each cylinder 2 is fitted with a fuel injector connected to a fuel supply circuit, intake valves opening into the distributor or intake manifold, and exhaust valves opening into the exhaust manifold. Each cylinder 2 is also fitted with a piston to compress the air / fuel mixture prior to ignition , and to transmit the momentum generated by ignition to an engine crankshaft. A spark plug is also provided to ignite the air / fuel mixture.

[0008] The intake circuit comprises a fresh air intake 5, generally fitted with an air filter 5a, connected to the inlet of a compressor 6a of a turbocharger 6. The outlet of the compressor 6a is connected to an intake valve 7, particularly of the throttle valve type, which in turn is connected to the distributor 3. A charge air cooler 8 is located between the compressor and the intake valve.

[0009] The exhaust manifold is connected to the inlet of a turbine 6b of the turbocharger 6, the outlet of the turbine 6b being connected to a vent 9 via a pollution control system 1 0 comprising a particulate filter and a three-way catalytic converter.

[0010] The state of the art consists of igniting a homogeneous air / fuel mixture as a function of the crankshaft angle. Fuel injection of direct injection type, is performed directly into the combustion chamber, defined by the volume of the cylinder bounded by the piston, into which the fuel injector, valves and ignition system open . Injection takes place at the beginning of the cycle, before the intake bottom dead center (BDC).

[0011] The quantity of fuel injected is controlled by an internal engine control unit ECU to ensure a stoichiometric air / fuel mixture (i.e. richness equal to 1 ) . This injection takes place, for example, fairly early in the cycle, during the intake stroke of the 4-stroke cycle, before the intake bottom dead center BDC. The resulting mixture is perfectly homogeneous at the moment of ignition , which takes place towards the end of the compression stroke. For example, ignition takes place at 20°(Vil) before the combustion top dead center (TDC). This ignition is carried out by the spark plug, which is also controlled by the engine control unit via an ignition coil.

[0012] Although very efficient, this type of control is subject to a combustion phenomenon known as "knocking", which is well known in spark-ignition engines. Knocking occurs above a certain engine load when cylinder pressure and temperature conditions are high. This typically happens when the pressure in the manifold, or intake manifold Pcol, exceeds an absolute value of 1 bar, which occurs at certain operating points of a supercharged engine. This abnormal phenomenon is self-ignition of the mixture after ignition at several points in the combustion chamber, before the flame front has burnt the fuel by so-called diffusion combustion , which is the normal combustion mode on sparkignition engines. This phenomenon can be seen on a graph showing cylinder pressure Pcyl as a function of crankshaft angle (°Vil). Strong pressure oscillations can be observed, superimposed on the expected pressure variation during the cycle. Figure [Fig 3] illustrates such a dashed curve, with reference 20. In particular, this phenomenon leads to the generation of noise, known as "knocking", which sounds "metallic" and is perceptible to the customer. Knocking is also the source of piston degradation or even destruction.

[0013] It is then necessary to reduce ignition timing to counteract this phenomenon , which degrades engine efficiency (higher fuel consumption) and limits maximum performance.

[0014] With the advent of high-efficiency motors operating on so-called "Miller" or "Atkinson" cycles, the problem of knocking is even more prevalent at high and full loads. This is because these processes use high levels of Compression Volumetric Ratio (CVR) (between 1 2 and 14, as opposed to the usual 1 0) , which are highly unfavorable for knocking. These high levels increase pressure and temperature in the combustion chamber. The efficiency gain associated with the higher CVR is therefore reduced by the need to reduce ignition advance to counteract knocking in the critical zone. The full efficiency gains associated with these processes are therefore not realized. As a result, fuel consumption is less than optimal, as are carbon dioxide emissions. CO2emissions, which are increasingly subject to regulatory constraints.

[0015] So there's a need for knock reduction on internal combustion and spark-ignition engines.

[0016] In the prior art, EP 1 770256B1 discloses a method of operating an engine with stratified mixture, in which fuel is injected in two distinct successive sequences, during the compression stroke, before ignition.

[0017] Also known is document DE1 0200401 7988A1 disclosing a method of operating an engine with stratified mixture, in which the injection comprises a first injection sequence carried out during the intake stroke, followed by a second main injection sequence carried out during the compression stroke.

[0018] The main injection sequence can be divided into several consecutive phases (2 or 3).

[0019] Ignition takes place near top dead center at the end of compression.

[0020] When there are two injection phases, ignition can take place after both phases or between the first and second phases.

[0021] When there are three phases, ignition can take place after all three, or between the first and second phases, or between the second and third phases.

[0022] These documents provide for a single ignition .

[0023] The technical problem remains unchanged. Explanation of the invention

[0024] The invention relates to a method for controlling an internal combustion engine, comprising cylinders and pistons designed to enable stratified combustion to take place, comprising the following steps for each combustion cycle: a. determination of the number of injection and ign ition patterns to be performed in the combustion cycle, b. determination of the quantity of fuel to be injected at each injection prior to an injection and ignition pattern by dividing the quantity of fuel to be injected in the combustion cycle by the number of injection and ignition patterns to be performed in the combustion cycle, c. determination of the time of each injection and ignition pattern as a function of crankshaft angle, said time of each injection and ignition pattern being between intake bottom dead center and combustion top dead center of the combustion cycle, d. control of the fuel injection circuit and ignition means to produce injection and ignition patterns as a function of the crankshaft angle, the quantity of fuel to be injected per injection and ignition pattern, and the timing of each injection and ignition pattern .

[0025] Prior to the step of determining the number of injection and ignition patterns to be performed in the combustion cycle, the following steps can be performed: a. determination of the presence of the operating point in a knocking zone as a function of a predetermined mapping accepting as input the load and rotational speed of the internal combustion engine, b. if the operating point is in a knocking zone, control the internal combustion engine with injection and ignition patterns and continue the process with the step of determining the number of injection and ignition patterns to be performed per combustion cycle, the number of injection patterns being at least equal to two, each injection taking place during the compression stroke of the engine cycle and being followed by ignition.

[0026] The number of injection and ignition patterns in a combustion cycle can depend on the speed and load of the internal combustion engine.

[0027] The invention also relates to a powertrain for an automobile vehicle, comprising an internal combustion engine with cylinders designed for stratified combustion , a rotational speed sensor, a control coil for each ignition means, a fuel pump, a fuel injector actuator and an electronic control unit for the internal combustion engine designed to perform the steps of the control process as described above.

[0028] A fuel injector can be positioned laterally in the combustion chamber.

[0029] An ignition means can be centrally located in the combustion chamber.

[0030] Each piston head can have an adapted, bowl-shaped profile with a concavity facing the fuel injector and ignition means, each piston head creating a rotary movement perpendicular to the cylinder axis.

[0031] The control method described above is not significantly more expensive than a prior art control method, and reduces or even eliminates knocking in critical areas. It eliminates the need to reduce ignition timing, which has a negative impact on engine efficiency and performance. Fuel consumption and CO2 emissions are reduced.

[0032] Brief description of the drawings

[0033] Further aims, features and advantages of the invention will become apparent from the following description , given solely by way of non-limiting example and made with reference to the appended drawings in which :

[0034] - Figure [Fig 1 ] illustrates the main components of a supercharged internal combustion engine,

[0035] - Figure [Fig 2] illustrates the main components of a combustion chamber suitable for stratified combustion ,

[0036] - Figure [Fig 3] illustrates the evolution of cylinder pressure as a function of crankshaft angle for an engine controlled according to the state of the art and for an engine controlled according to the invention , and

[0037] - Figure [Fig 4] illustrates the main steps in a process for controlling an internal combustion engine with stratified combustion .

[0038] Detailed description

[0039] Recent engine control systems are increasingly sophisticated, allowing the use of multi-injection (several injections per cycle instead of just one, typically up to 5) and multi-spark (several ignitions per cycle instead of just one, typically up to 5) strategies.

[0040] To reduce or even eliminate knocking in the speed / critical load zone, a so- called "multi-stratified" injection / ignition scheme is used, involving several injections and ignitions per combustion cycle, depending on the crankshaft angle.

[0041] To avoid knocking , fuel is injected progressively in several injections during the compression stroke of the cycle, each followed by an ignition. As the injection is split, each time a smaller quantity of fuel is injected than that required to obtain a mixture with an average richness of 1 . There are at least two injections. For example, if the injection is cut into 3, the first injection has an average richness of around 0.33, which cannot burn because the air-fuel mixture, if homogeneous, is too lean.

[0042] Combustion chamber architecture must therefore be designed to enable such "stratified" combustion , i.e. one in which the richness of the air-fuel mixture is not homogeneous.

[0043] Figure [Fig 2] illustrates a cylinder 2 comprising a cylinder head 1 1 and a piston 12 together delimiting the combustion chamber 13. At least one intake valve 16, an ignition means 1 5, at least one exhaust valve 14 and a fuel injector 17 arranged in the cylinder head 1 1 are also illustrated. The special feature of the combustion chamber is that the fuel is injected as close as possible to the spark plug via the interaction between the fuel jet (referenced 1 8) , the shape of the piston head 1 2 and the internal aerodynamics.

[0044] More specifically, the piston head has an adapted, bowl-shaped profile with a concavity facing the fuel injector 17 (in a lateral position in the combustion chamber) and the ignition means 1 5 (in a central position in the combustion chamber) .

[0045] Internal aerodynamics involve the creation of a rotary motion , known as "tumble", perpendicular to the cylinder axis, due to the shape of the piston crown . This combustion chamber architecture makes it possible to locally obtain a mixture with a richness substantially equal to 1 , which can burn as soon as ignited by the spark, whereas the richness over the entire combustion chamber is less than 1 .

[0046] Other engines suitable for this type of combustion are known to the state of the art, in particular the devices disclosed in documents EP142901 0A1 or EP1770256B1 .

[0047] The curve showing the evolution of the cylinder pressure Pcyl as a function of the crankshaft angle °Vil in a cylinder of an internal combustion engine controlled by the method according to the invention is illustrated in solid lines in figure [Fig 3] and bears the reference 21 . In contrast to the same curve for a homogeneous-combustion internal combustion engine (dashed line, reference 20) , this curve no longer exhibits the oscillations associated with knocking . This staged combustion allows a gradual build-up of pressure in the cylinder, avoiding any knocking. The number of "injection + combustion" patterns can be optimized according to engine speed and load. Outside the knocking zone, a classic "homogeneous" injection / ignition strategy can be maintained. In addition to the pressure criteria defined above, the knocking zone is defined as a high-load zone, notably from 50% of full load for a given engine speed. Alternatively, this rate can vary according to rotational speed, being higher for low rotational speeds and lower for high rotational speeds.

[0048] Figure [Fig 4] illustrates the control procedure for an internal combustion engine with combustion chambers designed for stratified combustion . For each combustion cycle, the following steps are performed :

[0049] In a first step 31 , the number of injection and ignition patterns to be performed per combustion cycle is determined. In particular, the number of injection and ignition patterns is predetermined and stored.

[0050] In a second step 32, the quantity of fuel to be injected at each injection prior to an injection and ignition pattern is determined by dividing the quantity of fuel to be injected per combustion cycle by the number of injection and ignition patterns to be performed per combustion cycle. In a third step 33, the time of each injection and ignition pattern is determined as a function of the crankshaft angle °Vil. The time of each injection and ignition pattern lies between BDC intake and TDC combustion . In a fourth step 34, the fuel injection circuit and the ignition means are controlled to produce the injection and ignition patterns as a function of the crankshaft angle °Vil, the quantity of fuel to be injected per injection and ignition pattern and the time of each injection and ignition pattern . In a particular embodiment, the control process comprises a preliminary step 30, in which it is determined whether the knocking zone is present by comparing the load and rotational speed with a predetermined mapping .

[0051] If there is no knocking zone, the internal combustion engine is controlled in homogeneous combustion .

[0052] If a knocking zone is present, the internal combustion engine is controlled in stratified combustion , and the control process continues with the first step 31 .

Claims

CLAIMS1 . Method for controlling a spark-ignition internal combustion engine comprising cylinders and pistons designed to enable stratified combustion to take place, comprising the following steps for each combustion cycle: a. determination of the presence of the operating point in a knocking zone as a function of a predetermined map accepting as input the load and rotational speed of the internal combustion engine, b. if the operating point is in a knocking zone, control the internal combustion engine with injection and ignition patterns and continue the process with the step of determining the number of injection and ignition patterns to be performed per combustion cycle, c. determination (31 ) of the number of injection and ignition patterns to be performed in the combustion cycle, an injection and ignition pattern comprising a set of injections each followed by an ignition , the number of injection and ignition patterns being at least equal to two, d. determination (32) of the quantity of fuel to be injected at each injection prior to an injection and ignition pattern by dividing the quantity of fuel to be injected in the combustion cycle by the number of injection and ignition patterns to be performed in the combustion cycle, e. determination (33) of the timing of each injection and ignition pattern as a function of crankshaft angle, said timing of each injection and ignition pattern being between intake bottom dead center and combustion top dead center of the combustion cycle, each injection taking place during the compression stroke of the engine combustion cycle and being followed by ignition, f. control (34) of the fuel injection circuit and the ignition means in order to perform the injection and ignition patterns as a function of the crankshaft angle, the quantity of fuel to be injected per injection and ignition pattern and the timing of each injection and ignition pattern.

2. Control method according to claim 1 , in which the number of injection patterns in a combustion ignition cycle depends on the speed and load of the internal combustion engine.

3. Automobile vehicle powertrain, comprising an internal combustion engine with cylinders designed for stratified combustion , a rotational speed sensor, a control coil for each ignition means, a fuel pump, a fuel injector actuator and an electronic control unit for the internal combustion engine designed to perform the steps of the control method as claimed in claims 1 or 2.

4. The powertrain of claim 3, in which a fuel injector (1 7) is positioned laterally in the combustion chamber.

5. Powertrain according to claim 3 or 4, in which an ignition means ( 15) is centrally located in the combustion chamber.

6. Powertrain according to any one of claims 4 to 6, in which each piston head has an adapted, bowl-shaped profile with a concavity facing the fuel injector ( 1 7) and ignition means ( 15), each piston head enabling a rotary movement to be created perpendicular to the cylinder axis.

Images