WO2020104173A1

WO2020104173A1 - Method to determine misfire in a cylinder of an internal combustion engine

Info

Publication number: WO2020104173A1
Application number: PCT/EP2019/079960
Authority: WO
Inventors: Eduardo MAIELLO; Luca TASCEDDA; Ivan Alejandro BAHENA LEDEZMA
Original assignee: Delphi Automotive Systems Luxembourg Sa
Priority date: 2018-11-19
Filing date: 2019-11-01
Publication date: 2020-05-28
Also published as: GB201818804D0; GB2579073A

Abstract

A method of determining if there is a misfire in one or more cylinders of an internal combustion engine during operation, said method comprising: a) determining the exhaust manifold pressure over a respective time window for a particular cylinder; b) determining the maximum and minimum exhaust manifold pressures during said time window; c) determining engine rotational speed; d) determining at least one of the parameters of engine load, air consumption and fuel consumption; e) calculating a misfire index based on said maximum and minimum pressure determined in step b) and said parameters of steps c) and d); f) determining whether there is a misfire based on the value determined in step e).

Description

METHOD TO DETERMINE MISFIRE IN A CYLINDER OF AN

INTERNAL COMBUSTION ENGINE

TECHNICAL FIELD

This invention relates to internal combustion engines and in particular to a method of detecting misfire events.

BACKGROUND OF THE INVENTION

In an internal combustion engine, a misfire event is commonly defined as an unintentional lack of combustion in the cylinder.

When a misfire event occurs, the fuel injected in the cylinder does not bum, and this unbumed amount of fuel then passes out through the exhaust system of the engine. Identifying a misfire event is important in order to meet legal requirements for on-board diagnostics, and to avoid damage to catalytic converter.

Conventional methods for detecting misfire events are based on measuring the fluctuations in the rotationalspeed of the crankshaft, these methods are known as CSM (crankshaft speed method). These“prior art” methods are very susceptible to external influences, such as the use of an electric motor attached to the drivetrain in Hybrid and Mild-Hybrid (48Volts) applications, and/or vibration of the drivetrain caused by driving on irregular surfaces or“rough roads”. These external influences can introduce a quite high error probability and reduce the misfire detection capability under certain conditions.

Patent document US 2015/0136080 A1 describes methods and systems for determining engine misfire. US 9,316,565 B2 describes exhaust manifold pressure based misfire detection for internal combustion engines and US 6,651,490 B1 describes a process for detecting a misfire in an internal combustion engine and system for carrying out said process. The methods of detection of misfire events that rely only on crank speed oscillations (CSM) are susceptible to a series of interferences that could lead to a false detection or to the incapacity of detection, such as: drive train oscillations cause by resonance of drivetrain components, such as a dual mass flywheel; influence of road surface (“rough road”) causing fluctuations on rotational speed or vibration of the crankshaft; influence of electric motors connected to the drivetrain on hybrid and mild-hybrid (48V) applications. The methods that used exhaust manifold pressure analyzes where not always capable of detecting the occurrence of misfire on a reliable way due to physical limitations on the hardware or apparatus used, or due to the design of their algorithms.

It is an object of the invention to provide improved methodology to detect misfire events, and also to provide methods for this under conditions that prior art methods are not capable of. SUMMARY OF THE INVENTION

In one aspect is provided A method of determining if there is a misfire in one or more cylinders of an internal combustion engine during operation, said method comprising:

a) determining the exhaust manifold pressure over a respective time window for a particular cylinder;

b) determining the maximum and minimum exhaust manifold pressures during said time window;

c) determining engine rotational speed;

d) determining at least one of the parameters of engine load, air consumption and fuel consumption;

e) calculating a misfire index based on said maximum and minimum pressure determined in step b) and said parameters of steps c) and d); f) determining whether there is a misfire based on the value determined in step e).

The misfire index may be determined in step e) from the following equation:

where the values of A, B and C are determined according to the parameters in steps c) and d) . The time window in respect of the cylinder generally may span the whole or a substantial portion of the exhaust stroke.

Said index may be compared with a threshold and a misfire determined if said index exceeds said threshold.

The method may include determining one or more parameters of i)) ambient pressure; ii) one of cam-phasers (VVT) position, intake and/or exhaust valves opening and closing positions; iii)) one of cylinder mixture or“Air Fuel Ratio” or Lambda, and additionally using the parameter(s) in step e).

The method may include determining one or more parameters of iv) particulate filter loading or saturation; v) turbocharger control actuator position; vi) spark angle or efficiency vii) fuel injection timing; and additionally using the parameter(s) in step e).

The values of A B C are determined form look up tables or MAPs relating to respective parameters

BRIEF DESCRIPTION OF THE DRAWINGS

The present invention is now described by way of example with reference to the accompanying drawings in which: Figure 1 shows a chart of cylinder pressure against time in terms of engine rotations for three cylinders of an internal combustion engine;

Figure 2 shows a block diagram of the methodology of determining P_max and

P min:

- Figure 3 shows a chart similar to figure 1 with respect to an engine cycle with respect to three cylinders;

Figure 4 shows the plots from figures 1 and 3 on their own for clarity chart. Fig 5 shows the same graph as figure 3 with identical reference numerals indicating the same, but also with the respective time windows for each cylinder;

Figure 6 shows the figure 5 further reduced for clarity and to illustrate the determination of the method with respect to cylinder# 3.

Figures 7a 7b and 7c respectively show how the factors A B and C are determined

Background

Figure 1 shows a chart of cylinder pressure against time in terms of engine rotations for three cylinders of an internal combustion engine, designated P#1 P#2 and P#3 in respect of the three cylinders cyl#l,cyl#2, and cy#3, where the engine runs without misfires. The Top Dead Centre (TDC) subsequent to the combustion stroke (TDC F) in the region of spark/firing and subsequent Bottom Dead Centre (BDC) time (with respect to crankshaft angle is shown) for each cylinder cyl#l,cyl#2, and cy#3 for cylinder 1 ,2, and 3, as indicated. Plot Pm3 shows a plot of cylinder pressure P#3 against time /crankshaft angle where there is a misfire in cylinder 3.

Invention In one aspect is provided a method for detecting the occurrence of a misfire event in one or more cylinders based on the sampling and analyzes of the exhaust manifold pressure values, also known as“P3”, and determining misfire from this which may include calculating a misfire index, and comparing the calculated index with one or more threshold values for identifying a misfire event. So the problem of lack of robustness or incapacity of detecting the occurrence of a misfire event in a combustion engine is solved by using a process and a method for detecting the occurrence of a misfire event in one or more cylinders based on a method based on the sampling and analyzes of the exhaust manifold pressure values, also known as “P3”, calculating a misfire index, and comparing the calculated index with one or more threshold values for identifying if a misfire event has occurred in one or more cylinders. The above mentioned method can also be used on Hybrid and Mild Hybrid (48V) applications.

Example

In one example a method for detecting a misfire event in one or more cylinders of an internal combustion engine, includes the following steps:

a) sampling the exhaust manifold pressure during e.g. at least one engine cycle at an appropriate sampling rate, e.g. proportional to rotational speed of the engine; b) analyzing the sampled signal, and identifying the minimum and maximum pressures Pmax and Pmin over a certain crankshaft angle window correspondent to (i.e. for each) each cylinder;

c) calculating a misfire intensity index from the signal analysis, according to the equation below:

Determination of the factors A, B and C are factors what will be described later.

The next step comprises comparing the misfire intensity index with at least one threshold value; and - identifying the occurrence of a misfire event based on comparison of misfire intensity index with at least one threshold value. If the misfire intensity index exceeds at least one threshold value, then the occurrence of a misfire event can be confirmed. The factors A, B, C and the thresholds mentioned above, can be calculated as functions of: a number of factors or parameters. These parameters may be measured inferred or provided by a model. Determination of Pmin and Pmax

Figure 2 shows a block diagram of the methodology of determining P_max and P_min.

A pressure sensor 1 which measure exhaust manifold pressure P3 is shown which sends an analogue single to a A/D converter 2. The resultant signal is passed through a low pass filter 3 and then during an engine cycle the pressure is monitored of the time windows for each cylinder, represented by blocks 41 ,42,43. With respect to each cylinder (via the respective time window) the Pmax and Pmin value of P3 is determined shown by blocks 6 and 5 respectively . The Pmin and Pmax are then fed into an array for Pmin and P max (ref numerals 9 and 8 respectively) .

Figure 3 shows a chart similar to figure 1 with respect to an engine cycle with respect to three cylinders cyl#l,cyl#2, and cy#3 for cylinder 1 ,2, ad 3. Again the chart shows pressure plots in respect to three cylinders during normal combustion. Again is shown a pressure plot of cylinder 3 with misfire. Identical reference numerals of the figure 3 relate to the same as in figure 1. Thus again the figure shows results of where there is normal combustion and also a misfire in cylinder 3. In addition the chart shows a plot of exhaust manifold pressure for the case where there is no misfire Pex and where there is misfire in cylinder 3, (Pexm3). As can be seen there is a marked drop in exhaust manifold pressure (in the figure generally between -70 deg and +100 deg crankshaft angle (which is generally over the period of the exhaust stroke of cylinder 3.) when there is a misfire compared to when there is no misfire. Figure 4 shows the plots from figures 1 and 3 on their own for clarity chart.

In aspects of the invention the exhaust pressure is monitored over time windows (crankshaft intervals) and analyzed in order to determine misfires. In particular the maximum and minimum exhaust pressures are determined in the respective window.

Fig 5 shows the same graph as figure 3 with identical reference numerals indicating the same, but also with the respective time windows for each cylinder. The position of the TDC and BDC with respect to each numbered cylinder (#1,#2, #3) cylinder is as shown. In the figure TDC F again denotes the TDC at the end of the compression stroke and TDC NF indicates the TDC at the end of the exhaust stoke.

For cylinder 3 the respective window is referenced as“window Cyl 3”. The time windows for collecting data for determining misfire in cylinder 3 is shown as TW#3. This generally spans the time from the BDC after the nominal combustion stroke to the subsequent TDC at the end of the exhaust stroke. When there is no misfire one would expect higher pressure in the exhaust manifold during this exhaust stroke / time window compared to when there is a misfire. The skilled person would readily be aware how to select the appropriate time window in respect of a or each cylinder. The actual exhaust strokes/potential time windows in respect of cylinders 1 and 2 are show by reference numerals TW#1 and TW#2.

Figure 6 shows the figure 5 further reduced for clarity and to illustrate the determination of the method with respect to cylinder# 3. The time window for determination of Pmin and Pmax with respect to cylinder #3 is shown with reference numeral. TW#3. The chart shows plots of exhaust manifold pressure where there is misfire in cylinder 3 and no misfire. Within the time window in order to determine whether there is a misfire, the values of Pmax and Pin are determined i.e. the points of maximum and minimum exhaust manifold pressure during this time window. These are then extracted and used in the later steps of the methodology. As can be seen the chart shows the values of Pmax and Pmin for both cases of misfire and no misfire in cylinder 3. Where there are misfire and their respect values of Pmax and Pmin. Determination of parameters A, B and C

In one basic example the parameters used in the calculation of A, B and C are a) engine rotational speed; b) one of engine load, or air and fuel consumption;

In a preferred embodiment additional parameters are used e.g. c) ambient pressure; d) one of cam-phasers (VVT) position, intake and/or exhaust valves opening and closing positions; e) one of cylinder mixture or“Air Fuel Ratio” or Lambda In further refined more complex embodiment other parameters may be used such as f) particulate filter loading or saturation; g) turbocharger control actuator position (wastegate or VGT); h) spark angle/efficiency (in case of spark ignition engines) i) fuel injection timing (in case of compression ignition and spark ignition engines). The skilled person would be aware how to determined factors A, B and C from the above parameters.

Detailed Examples of Determination of Factors A B and C The factors A, B and C are calculated according to the diagram blocks in figures 7a 7b and 7c which respectively show how the factors A B and C are determined

Each figure is similar and the inputs/parameters which may be used in the determination are shown on the left.

In figure 7a engine load and rotational speed are input into a 2D lookup table or map A1. The output of this is a raw value of A which may be optionally refined by one or more factors determined by other input parameters, the factor refinement being performed by multiplying the outputs A by one or more factors as indicated in boxes 20,21 ,22,23 : An optional further input may be exhaust cam phaser position where the above refinement factor for box 20 is determined from a look up table A2. A further optional input may be spark efficiency - this is input to look up table map A3 to determine the factor for box 21. A further optional input may be Air Fuel ratio (A/F) - this is input to look up table map A4 to determine the factor for box 22. A further optional input may be barometric pressure - this is input to look up table map A5 to determine the factor for box 23. The values on the look-up tables A1 , A2, A3, A4 and A5 may to be adjusted according to the engine in which this method is applied.

The factors B and C are calculated in a similar way, using the look-up tables Bl, B2, B3, B4, B5 and Cl, C2, C3, C4, C5 shown in figure 7b and 7c

Determination of Misfire.

A misfire index may be computed from the following equation

If this index is greater or higher than a threshold, then a misfire may be determined in the appropriate cylinder.

In the prior art, the detection of a misfire events are based on measuring the fluctuations in the rotational Speed of the crankshaft, these methods are known as CSM (crankshaft speed method). These“prior art” methods are very susceptible to external influences, such as the use of a electric motor attached to the drivetrain in Hybrid and Mild-Hybrid (48Volts) applications, and/or vibration of the drivetrain caused by driving on irregular surfaces or“rough roads”. These external influences can introduce a quite high error probability and reduce the misfire detection capability under certain conditions.

This invention uses an algorithm that evaluates the behavior of the exhaust manifold pressure signal to effectively determine if a misfire event occurred or not. Although some methods based in the“prior art” also uses the exhaust pressure for detecting a misfire event, the advantage of this invention compared to the prior art is the improvement of the detection capability by using a more precise and not so complex algorithm, independently of the number or arrangement of cylinders. This misfire detection method can also be applied on Hybrid and or Mild Hybrid (48V) applications, not mentioned by the prior art.

The method provides an improvement in the robustness of the misfire detection using an efficient algorithm to verify the behavior of the exhaust pressure during the engine operation, that allows a misfire event to be detected even under certain conditions that other conventional methods are not capable. This invention can be used on spark-ignited (commonly referred as Otto cycle) or compression- ignition (commonly referred as Diesel cycle) engines. This misfire detection method can also be used on Hybrid and or Mild Hybrid (48V) applications.

Claims

1. A method of determining if there is a misfire in one or more cylinders of an internal combustion engine during operation, said method comprising:

b) determining the maximum and minimum exhaust manifold pressures in said particular cylinder during said time window;

c) determining engine rotational speed;

e) calculating a misfire index based on said maximum and minimum pressure determined in step b) and said parameters of steps c) and d);

f) determining whether there is a misfire based on the value determined in step e).

2. A method as claimed in claim 1 where the misfire index is determined in step e) from the following equation:

where the values of A, B and C are determined according to the parameters in steps c) and d) .

3. A method as claimed in claims 1 or 2 where the time window in respect of the cylinder generally spans the whole or a substantial portion of the exhaust stroke.

4. A method as claimed in claims 1 to 3 wherein said index is compared with a threshold and a misfire determined if said index exceeds said threshold.

5. A method as claimed in claim 1 to 4 including determining one or more parameters of i)) ambient pressure; ii) one of cam-phasers (VVT) position, intake and/or exhaust valves opening and closing positions; iii)) one of cylinder mixture or“Air Fuel Ratio” or Lambda, and additionally using the parameter(s) in step e).

6. A method as claimed in claim 1 to 4 including determining one or more parameters of iv) particulate filter loading or saturation; v) turbocharger control actuator position; vi) spark angle or efficiency vii) fuel injection timing; and additionally using the parameter(s) in step e).

7. A method as claimed in claims 1 to 6 where the values of A B C are determined form look up tables or MAPs relating to respective parameters