WO2017076450A1

WO2017076450A1 - Engine control method and engine control device

Info

Publication number: WO2017076450A1
Application number: PCT/EP2015/075793
Authority: WO
Inventors: Tom Kaas; Matias PALMUJOKI
Original assignee: Wärtsilä Finland Oy
Priority date: 2015-11-05
Filing date: 2015-11-05
Publication date: 2017-05-11
Also published as: EP3371436A1; KR20180093883A; CN108350817A

Abstract

The present invention relates to an engine control method to be used for detecting misfire in an internal combustion engine having at least one pressure sensor for in-cylinder pressure detection, and an electronic control unit, wherein the method comprises the following steps: i) detecting an in-cylinder pressure, ii) determining an engine crank position, iii) calculating temperature index based on measured in-cylinder pressure and a cylinder volume as a function of measured engine position via the ideal gas law by the electronic control unit; and iv) determining from calculated temperature index within the current expansion stroke of the internal combustion engine whether or not a combustion event takes place in the current expansion stroke.

Description

ENGINE CONTROL METHOD AND ENGINE CONTROL DEVICE

BACKGROUND OF THE INVENTION

1. Field of the invention

The present invention relates to an engine control method and engine control device, a misfire detection method and misfire detection device, and particularly to a misfire detection suitable for four-stroke, medium-speed engines which are widely used in marine drive applications.

2. State of the Art

Misfire detection systems are becoming increasingly important due to recent environmental issues, for instance. In general, the occurrence of misfires reduce engine efficiency. An early misfire diagnosis also allows preventing damages to the exhaust emission system and consequent costs for the user. Today few low cost methods exists in order to precisely detect single misfires in real time, the majority in fact require the use of expensive sensors (e.g . pressure sensors) or dedicated circuits (e.g. ionization current sensing). A method can be based on the real time analysis of the crankshaft angular velocity and its variations. This is possible since each misfire event generates an abrupt perturbation of the crankshaft angular velocity that can be detected using an appropriate signal processing algorithm.

As mentioned above, misfire detection on internal combustion engines is also commonly based on in-cylinder pressure measurement. One way of detecting a misfire cycle is to calculate the work done by the cylinder during the corresponding cycle. This can be achieved by calculating gross-indicated mean effective pressure or net-indicated mean effective pressure (IMEP).

The indicated mean effective pressure (IMEP) is calculated by dividing the indicated work per cylinder per cycle by the swept volume per cylinder.

The pumping loss (pumping work) PMEP is calculated by dividing the Pumping work per cylinder per cycle by the swept volume per cylinder.

Accordingly, the net IMEP is calculated by subtracting the pumping loss PMEP from the gross IMEP. net IMEP = gross IMEP - PMEP

The indicated mean effective pressure (IMEP) is a mean effective pressure calculated from in-cylinder pressure which is averaged from an in-cylinder pressure over engine cycle (720° in a four-stroke, 360° in a two-stroke) . Direct IMEP measurement also requires combustion pressure sensing equipment.

The indicated mean effective pressure (IMEP) is calculated according to

where a is 360 for net IMEP and 180 for gross IMEP. Typically, in order to ensure that the control system is able to take action before the injection on the next cycle, the misfire status is needed before the next bottom dead centre after combustion is reached. By the nature of IMEP calculation this is not possible. In Figure 1, the IMEP based misfire detection method is shown.

SUMMARY OF THE INVENTION

1. Problem to be solved by the invention

In order to improve engine control, an alternative approach to the conventional art is required.

Furthermore, in order to overcome the problem of detecting a misfire cycles too late for the control system to make counter actions before next cycle injection starts, an alternative method for misfire detection is needed.

Namely, in Fig . 1 the problem underlying the present invention is shown. As mentioned above, (in-cylinder) pressure values from -180 degrees before the top dead centre / combustion to 180 degrees after the top dead centre / combustion are required for the calculation of e.g. the gross IMEP. However, as depicted in Figure 1, the calculation result should be present at a dead line which lies approximately 120 degrees after the top dead centre / combustion in order to ensure an appropriate engine control so as to prevent further misfiring in the next cycle. In other words, in case the detection or calculation result is provided after the dead line (e.g . 120 degrees after TDC (top dead centre)), it can not be assured that the engine control may prevent the misfire in the next cycle.

The present invention has been conceived in view of the above-described problem. 2. Solution of the problem

The above indicated problem is solved by an engine control method according to claim 1 or claim 2 and an engine control device according to claim 9. Further advantageous developments are subject-matters of the dependent claims.

An engine control method to be used for detecting misfire in an internal combustion engine having at least one pressure sensor for in-cylinder pressure detection, and an electronic control unit comprises the following steps: i) detecting an in-cylinder pressure, ii) determining an engine crank position, iii) calculating temperature index based on measured in-cylinder pressure and a cylinder volume as a function of measured engine position via the ideal gas law by the electronic control unit; iv) determining from calculated temperature index within the current expansion stroke of the internal combustion engine whether or not a combustion event takes place in the current expansion stroke.

A further aspect of the invention also corresponds to an algorithm for detection of a misfire event closer to the combustion top dead centre compared to known methods, thus allowing the control system to take counteraction for misfiring before the next cycle injection command .

The invention may depend on measurement of cylinder pressure, engine position and knowledge of engine geometry in order to calculate the cylinder volume as a function of engine position.

The invention can be used to create a misfiring indication already close to the TDC of the firing cylinder. Using the misfire indication counteractions can be taken by the control system already before the next combustion cycle. Counter actions can e.g . combustion timing (spark timing, pilot timing and fuel injection timing), fuel duration and/or valve timing .

Another aspect of the invention provides an engine control method for detecting misfiring which comprises

detecting in-cylinder pressure, determining an engine position, and

determining misfiring in one cycle, if a significant temperature rise after the top dead centre position of the corresponding cylinder at which combustion should be initiated is not detected based on the calculated temperature index.

This temperature index is calculated based on the ideal gas law. The temperature at each single crank angle can be calculated based on the respective in cylinder pressure which is measured and the respective volume of the cylinder which depends on the crank angle. Due to this, a differential temperature between two measurements, i.e. pressure measurements, and the correspondingly known cylinder volumes can be calculated . Accordingly, a product of the calculated differential temperature and a factor nR which relates to the trapped mass inside the cylinder provides the temperature index.

The temperature index can be used to measure or rather indicate the temperature rise scaled with gas amount, due to combustion of fuel . This means that if the change in temperature index is >0, combustion of fuel in the cylinder has increased the temperature. If the change in temperature index =0, the combustion of fuel in the cylinder has not increased the temperature. The change in temperature index could also be <0, if more energy leaks out from the cylinder than the energy has been created by the combustion of fuel. In any case a change in temperature index < =0 would then be interpreted as a misfire and there is no significant temperature rise due to combustion.

In other words, in contrast to the conventional misfire detection which is directly associated with the measured pressure value, the misfire detection according to the present invention can more quickly provide a determination whether or not misfire took place. This can be achieved by using the temperature index which is calculated based on the actually cylinder volume. Therefore, an early discrimination can be made due to the fact that the temperature indexes of an misfire-event and a non-misfire-event drift apart directly after the TDC (or the instructed ignition timing). "Directly after the TDC" means within 10 to 100, preferably 10 - 75, and more preferably 10 - 50 degrees crank angle after the TDC as can be gather best from Fig . 3. In contrast to the above-mentioned IMEP detection, the method according to the present invention allows for a quicker determination whether or not misfiring occurs in the current cycle. In detail, a misfire detection can be thus achieved before the dead line at approximately 120 degrees after the predicted combustion, which basically equals 120 degrees after the top dead centre (TDC). In two-stroke engines, each top dead centre should provide a combustion initiation no matter whether derived by self-ignition or spark-ignition. In four- stroke engines only every second passing of the top dead centre should provide a combustion initiation. With respect to a spark-ignition engine, a misfire determination method could also be used for pre-ignition detection in case a significant temperature rise is determined or calculated before the fuel should have been ignited by the spark plug .

In contrast to the conventional IMEP calculation which can only be performed after the passing of 180 degrees crank angle after the TDC, the present invention allows misfire detection between the TDC and at least 120 degrees crank angle after the TDC at which an ignition/start of combustion has been supposed . Hence, the misfire detection of the present invention can be made earlier than in the conventional art including the IMEP calculation.

A significant temperature rise can be clearly distinguished from an adiabatic temperature rise within the cylinder.

The engine control method preferably includes taking at least one out of a plurality of counteractions for preventing misfiring before the next cycle injection command. Such counteractions may include adjustments of combustion timing or spark timing, and/or adjustments of pilot injection timing and/or main fuel injection timing, adjustments of fuel injection amount and/or fuel injection duration, and/or valve timing . Basically the possible counteractions vary slightly in accordance with the basic structure of the engine. In other words, the counteractions for four-stroke-gasoline engines may be somewhat different from the counteractions for two-stroke-diesel engines. The crank angle may be detected by a crank angle sensor. However, the engine position may also be determined from other measurements, for example, cylinder pressure measurement.

The engine control method for detecting misfiring may further preferably provide that the determining whether or not misfiring occurs in the present cycle is made within an angular range from the top dead centre of the corresponding cylinder to a dead line which lies approximately 120 degrees after the top dead centre or combustion in order to ensure an appropriate engine control, preferably within a range from the top dead centre of the corresponding cylinder to 120 degrees after the top dead centre, and more preferably within a range from the top dead centre of the corresponding cylinder to 100 degrees after the top dead centre, and most preferably within a range from the top dead centre of the corresponding cylinder to 75 degrees after the top dead centre. This dead line is shown in Figure 1 between the TDC and 180 degrees crank angle at the upper right side in the diagram at approximately 120 degrees after the TDC, for instance.

In contrast to the prior art in which the pressure variation sequence has to be continued until 180 degree after the TDC (cf. gross IMEP), the determination according to the present invention can be made before reaching 120 degree after the TDC. Hence, in contrast to the prior art, the electronic control unit of the present invention which is used for controlling the engine has sufficient time for taking countermeasures such that no further misfiring occurs in the subsequent cycle. In the misfire detection in line with the IM EP calculation, the electronic control unit does not have sufficient time for taking countermeasures such that no further misfiring occurs in the subsequent cycle. Therefore, engine condition cannot be changed for the next cycle in time such that the next cycle will most likely show misfiring as well. The changes of the electronic control unit for preventing misfire will not effect for the next cycle. In other words, compared to the engine control method of the present invention, the conventional engine control includes two misfiring cycle instead of one misfiring cycle in the engine control of the present invention. The engine control method for detecting misfiring may provide, in case a misfire is detected in the actual cycle, that the engine control keeps the inlet valve and/or outlet valve shut, so as to keep the actual air fuel mixture within the cylinder, and tries to achieve combustion in the next cycle by correcting the fuel injection amount. Such a quick control for the subsequent cycle may be achieved due to the early detection of misfire according to the present invention. This kind of engine control provides that the misfire condition will be canceled for the next cycle and that the engine efficiency is improved, since the non-combusted air fuel mixture of the misfire cycle is not only exhausted, but re-used in the subsequent cycle with some adjustments in the engine control for the subsequent cycle, e.g. the additional injection of fuel to the existing air-fuel mixture so as to achieve a combustible mixture in the next cycle. Accordingly, such a control achieves a larger efficiency compared to the conventional art.

As a second aspect of the present invention an engine control device for detecting misfiring which comprises an engine with at least one cylinder, engine crank position determination means, a pressure sensor capable of sensing the in- cylinder pressure within the at least one cylinder of the engine, and an electronic control unit. The electronic control unit is capable of determining misfire within the current expansion stroke based on a temperature index calculated by the electronic control unit from the in-cylinder pressure and the engine crank position, such that the electronic control unit can prevent misfiring in the next cycle.

The engine position determination means may be a crank angle sensor or a device which may conclude a cylinder position in line with the in-cylinder pressure. However, it shall be noted that the engine position, i.e. the current state of the crank shaft or the current state of the respective cylinder may also be determined by analyzing the detection results of the pressure sensor. For example, since the electronic control unit may be able to calculate the adiabatic temperature changes, the electronic control unit may determine the engine position from the detection results of the pressure sensor which are set in relation with the changing cylinder volume which is depend of the crank angle. For that reason, the "engine position determination means" may be constituted by an electronic control unit and an in-cylinder pressure sensor. These arrangements are suitable for determining the engine position, especially the top dead centre and the bottom dead centre. In other words, the provision of a crank angle sensor is not essential. Since usually the inlet valve is opened or closed at the bottom dead centre or the top dead centre, the respective control signals for the valve actuation, i.e. Valve opening or closure, can be obtained when the "engine position determination means" recognizes the respective engine position. In other words, a crank angle sensor is not essential and a considerable control work can be based on the in-cylinder temperature calculation .

The engine may also be a two-stroke engine. In a two-stroke engine, the complete time between two subsequent injection commands or ignition commands, i.e. two subsequent combustions, is considerably smaller than in a four-stroke engine. Namely, in a four-stroke engine, combustion is executed every 720 degrees, whereas in a two-stroke engine combustion is executed every 360 degrees.

The engine may be a medium-speed engine. In this context, "medium speed" means an engine having 500 - 1500 revolutions per minute. The engine may also be a low-speed engine. In this context, "low speed" means an engine which does less than 500 revolutions per minute. For example, marine engines used for big ships such as tankers or bulk carriers, and to which the present invention is very suitably adapted, are usually low speed engines which have 60 to 250 revolutions per minute. Due to the comparatively large time period between the combustion events, compared to high speed engines having several thousands of revolutions per minute, the electronic control unit can follow the temperature changes in the cylinder with a regular in-cylinder pressure measurement rate in order to provide an appropriate resolution for making a misfire determination.

BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 is a schematic diagram which indicates the availabilities of misfire statues according to the present invention and according to the IMEP calculation of the related art;

Figure 2 shows a typical signal trace for the temperature index during an engine cycle;

Figure 3 shows a comparison of signal trace of temperature index from a normal combustion cycle without misfire and a motored cycle with misfire;

Figure 4 shows a flow chart and a diagram for explaining the structure of the present invention; and

Figure 5 shows an inlet valve timing estimation using the temperature index.

DETAILED DESCRIPTION OF THE PRESENT INVENTION

The engine control method according to the present invention includes a misfire detection method based on a temperature index which is calculated in line with in-cylinder pressure measurement. The temperature in the cylinder increases as the trapped mass is compressed and combusted . If neglecting the heat transfer from the cylinder walls the temperature increase for a short segment of the work cycle can be expressed as

(2)

It is known that the Poisson's equations for an adiabatic process of an ideal gas can be used to describe the pressure and temperature in the cylinder during compression and expansion. Hence during compression and expansion the following relation applies

(3)

Hence the temperature increase in the cylinder orig inating from volume change can be expressed as

The temperature increase originating from combustion can be calculated as the total temperature change subtracted by the temperature change due to volume change.

From the ideal gas law it is known that

(6)

Inserted in the equation for temperature rise originating from combustion we get

(7)

As the trapped mass inside the cylinder is not known, the equation is multiplied by the factor nR.

Equation (8) provides the final expression for a temperature index T_index - In Figure 2, a typical signal trace of the temperature index is shown for a normal engine cycle. Moreover, Equation (8) provides that the temperature index T_index is obtained by multiplying the temperature rise originating from combustion AT_COmbustion with the factor nR. In other words, the temperature index T_index is able to reflect or indicate the combustion independent from any geometric dimensions of the engine, i.e. the combustion chamber parameters.

In case misfire appears during a cycle, the level change of the temperature index moving from compression to expansion does not appear. In Figure 3, a comparison is shown between the calculated temperature index for a combustion cycle (= no misfire) and a motored cycle (= misfire).

As shown in Figure 3, the fact that the level of the temperature index trace does not change even if combustion should have happened should be interpret as an indication of misfire. It should be noted that it is possible to detect the misfire event already close to the combustion top dead centre (TDC). Close to the TDC means that a detection or determination can be made within a range of 120 degrees, preferably within a range of 100 degrees, and more preferably within a range of 75 degrees, after the TDC .

Next, the use of the temperature index for valve timing estimation will be explained. As can be seen from the temperature index signal trace in Figure 2 the level of the temperature index is constant during compression (before combustion) and during expansion (after combustion). By inspecting the early part (-360 to -200 crank angle degrees) of the temperature index trace it is evident that there is a state change from a rising temperature index to a constant temperature index. As no combustion takes place in this particular interval of the combustion cycle it is likely that the state change originate from inlet valve closure. Hence using the same temperature index, as used for early misfire detection, valve timing estimation can be achieved. In Figure 5 the signal trace of the temperature index for a typical engine cycle is shown with the estimated inlet valve timing .

As a further extension also the exhaust valve timing could be estimated using similar techniques. The information which can be achieved by the evaluation of the temperature index, may also provide correlated information on the status of the inlet valve and/or the exhaust valve. For example, this information may give information about any failure of the inlet valve and/or the exhaust valve. In other words, the use of a calculated temperature index may result in an omission of feed-back controlled inlet and outlet valves. Hence, single controlled valves may be used as inlet and outlet valve, whereas the information regarding the opening and closure states of the valves may be gained by interpreting the temperature index-crank angle curve, as can be seen in Figure 5.

Beside the additional information with respect to opening and closure state of the inlet and outlet valve which can be advantageously used in any engine control, the engine control with the misfire detection can take advantage of this information as well. Namely, in case there is a misfire detected in the present cycle which occurs early in the cycle, e.g. within the first 100 degrees crank angle after the intended combustion (TDC), the engine control has still enough time so as to take measures such that no misfire occurs in the subsequent cycle. The electronic control unit may keep closed the inlet valve and/or the outlet valve during the misfire cycle such that basically the same air-fuel mixture is present in the corresponding cylinder for the subsequent cycle for which adjustments can be made in order to achieve combustion in the subsequent cycle. These adjustments may include that the fuel injection amount which is additionally injected for the subsequent cycle is injected into the non-combusted air-fuel mixture of the preceding misfire cycle. This improves the engine efficiency compared to an engine control in which the non-combusted air-fuel mixture of the preceding misfire cycle is conventionally exhausted and a completely new (adjusted) air-fuel mixture is injected into the cylinder in which the misfire has occurred in the preceding cycle.

Claims

1. Engine control method to be used for detecting misfire in an internal combustion engine having at least one pressure sensor for in-cylinder pressure detection, and an electronic control unit, comprising

detecting an in-cylinder pressure,

determining an engine crank position,

calculating temperature index based on measured in-cylinder pressure and a cylinder volume as a function of measured engine position via the ideal gas law by the electronic control unit;

determining from calculated temperature index within the current expansion stroke of the internal combustion engine whether or not a combustion event takes place in the current expansion stroke.

2. Engine control method for detecting misfiring, comprising

detecting an in-cylinder pressure,

determining an engine position, and

determining misfiring in an expansion stroke, if a significant temperature rise after the top dead centre position of the corresponding cylinder at which combustion should be initiated is not assumed based on the calculated temperature index.

3. Engine control method for detecting misfiring according to claim 2, wherein the significant temperature rise can be clearly distinguished from an adiabatic temperature rise within the cylinder.

4. Engine control method for detecting misfiring according to claim 1, 2 or 3, further comprising :

taking at least one out of a plurality of counteractions for preventing misfiring before the next cycle injection command .

5. Engine control method for detecting misfiring according to claim 4, wherein the plurality of counteractions include adjustments of combustion timing or spark timing, and/or adjustments of pilot injection timing and/or main fuel injection timing, adjustments of fuel injection amount and/or fuel injection duration, and/or valve timing.

6. Engine control method for detecting misfiring according to any of claims 1 to

5, wherein the determining whether or not misfiring occurs in the current expansion stroke is made within an angular range from the top dead centre of the corresponding cylinder to a dead line which lies approximately 120 degrees after the top dead centre or combustion in order to ensure an appropriate engine control, preferably within a range from the top dead centre of the corresponding cylinder to 80 degrees after the top dead centre, and more preferably within a range from the top dead centre of the corresponding cylinder to 50 degrees after the top dead centre.

7. Engine control method for detecting misfiring according to any of claims 2 to

6, wherein, in case a misfire is detected in the current expansion stroke, the engine control tries to achieve combustion in the next cycle by correcting the fuel injection amount.

8. Engine control method for detecting misfiring according to claim 7, wherein the engine control keeps the gas exchange valves shut until the next expansion stroke and appropriately amends the air-fuel ratio within the respective cylinder for which misfire has been detected.

9. Engine control method for detecting misfiring according to claim 7, wherein the engine control discharges the unburnt air fuel mixture from the cylinder and charges an adjusted air fuel mixture in the cylinder for the next combustion event.

10. Engine control device for detecting misfiring, comprising

an engine with at least one cylinder;

engine crank position determination means; a pressure sensor capable of sensing the in-cylinder pressure within the at least one cylinder of the engine; and

an electronic control unit;

wherein the electronic control unit is capable of determining a misfire within the current expansion stroke based on a temperature index calculated by the electronic control unit from the in-cylinder pressure and the engine crank position, such that the electronic control unit can prevent misfiring in the next cycle.