Internal combustion engine control device and internal combustion engine control method

Abstract

Internal combustion engine control device (1A) comprising the following: a combustion parameter calculation unit (410) that calculates a combustion parameter of a combustion cycle in an internal combustion engine (100), a trend calculation unit (430) that calculates a trend of the change of the combustion parameter calculated by the combustion parameter calculation unit (410) over several combustion cycles, a variance calculation unit (440) which calculates a variance of the combustion parameter based on the combustion parameter in the multiple combustion cycles, and a combustion stability assessment unit (470) which assesses combustion stability on the basis of the trend of change of the combustion parameter in the multiple combustion cycles calculated by the trend calculation unit (430) and the variance of the combustion parameter calculated by the variance calculation unit (440), where the variance calculation unit (440) calculates the variance of the combustion parameter in the multiple combustion cycles from a mean value of the combustion parameters in the multiple combustion cycles, characterized by an effect removal unit (460) which removes an effect of a trend of change of a combustion parameter by subtracting a distribution resulting from the trend of change of the combustion parameter calculated by the trend calculation unit (430) from the variance of the combustion parameter calculated by the variance calculation unit (440) from the mean.

Description

Technical area

[0001] The present invention relates to an internal combustion engine control device and an internal combustion engine control method. Technical background

[0002] PTL 1 discloses an internal combustion engine combustion state detecting device that detects the combustion state of each cylinder group by combustion state detecting means in an internal combustion engine, which comprises combustion state control means for adjusting the state of the air-fuel mixture to achieve a target combustion state of the air-fuel mixture introduced into each of a plurality of cylinder groups in which a plurality of cylinders are divided, and for adjusting the state of the air-fuel mixture so that the combustion state between the cylinder groups converges to the same state. wherein the combustion state detecting device comprises combustion state detection prohibiting means for prohibiting detection of a combustion state by the combustion state detecting means before elapse of a reference convergence period in which the combustion state between the cylinder groups is expected to converge to the same state by adjusting the state of the air-fuel mixture. Citation listPatent literature

[0003] PTL 1: JP 2011 - 106 403 AEP 1 348 856 A1 discloses a control device according to the preamble of present claim 1. Summary of the inventionTechnical problem

[0004] The operating state of an internal combustion engine (engine) can be divided into an equilibrium state and a transient state. In the equilibrium state, the engine speed and torque are constant, and in the transient state, the engine speed and torque change. In engine design, engine characteristics are often evaluated in the equilibrium state. On the other hand, when a vehicle is traveling on a road, there are very few regions where the vehicle operates in the equilibrium state; in most regions, it operates in the transient state.

[0005] It is believed that many of the conventional inventions disclosed regarding the combustion state detection method were based on knowledge obtained through performance evaluation during the engine development stage. Therefore, most detection methods are either a detection method applicable only to the steady state, or a detection method that determines the steady state and the transient state, detects the combustion state in the case of the steady state, and prohibits detection of the combustion state in the case of the transient state (see PTL 1).

[0006] However, as described above, in actual operation, some regions operate in the steady state, and most regions operate in the transient state. It is also difficult to clearly state the criteria for distinguishing between the normal state and the transient state. Therefore, an object of the present invention is to provide a combustion state detection method applicable even during the transient state. Solution to the problem

[0007] To solve the above problem, the present invention comprises an internal combustion engine control unit according to claim 1 and an internal combustion engine control method according to claim 6. Advantageous effects of the invention

[0008] According to the present invention, the combustion stability can be accurately judged taking into account the effect of the trend even during the transient operation. Brief description of the drawings

[0009] They show: Fig. 1 a diagram showing a schematic representation of an internal combustion engine, Fig. 2 is a schematic diagram showing four cylinders of the internal combustion engine arranged in series, Fig. 3 is a graph showing a relationship between the crank angle and the cylinder pressure in a cylinder of the internal combustion engine, Fig. 4 a diagram showing four strokes of the cylinder of the internal combustion engine, Fig. 5 a graph showing the change in IMEP per combustion cycle in the cylinder of the internal combustion engine, Fig. 6 a graph showing the change of cPi per combustion cycle in the cylinder of the internal combustion engine, Fig. 7 is a diagram showing a distribution and mean value of combustion parameters during steady-state operation, Fig. 8 is a diagram showing a distribution and mean value of combustion parameters during transient operation, Fig. 9 is a diagram showing a distribution of combustion parameters and a change trend during transient operation, Fig. 10 a graph showing a change of Neu_cPi per combustion cycle in a given cylinder, Fig. 11 is a diagram showing a configuration of a control device according to a first embodiment, Fig. 12 is a flowchart of a method for judging a combustion state by the control device according to the first embodiment, Fig. 13 is a diagram showing a configuration of a control device according to a second embodiment, Fig. 14 is a flowchart of a combustion state judgment method by the control device according to the second embodiment, Fig. 15 is a diagram showing a distribution of combustion parameters, a change trend, and a sudden change in combustion during transient operation according to a third embodiment; Fig. 16 is a diagram showing a configuration of a control device according to the third embodiment, Fig. 17 is a flowchart of a combustion state judgment process by the control device according to the third embodiment, Fig. 18 is a graph showing the relationship between the crank angle and the cylinder pressure in the cylinder of the internal combustion engine, and Fig. 19 is a graph showing the relationship between the crank angle and the amount of heat generated in the cylinder of the internal combustion engine. Description of embodiments

[0010] Embodiments of the present invention will be described in detail below with reference to the drawings. Only the second embodiment falls within the scope of independent claims 1 and 6. [First embodiment]

[0011] First, according to an embodiment of the present invention, an engine control unit (ECU) 1 that controls an internal combustion engine will be described. Hereinafter, the ECU 1 will be referred to as a control device 1.

[0012] According to the present embodiment, a case where the internal combustion engine control device 1 is applied to an internal combustion engine 100 for a vehicle will be described as an example.

[0013] The Fig. 1 and Fig. 2 are schematic diagrams for describing the internal combustion engine 100 according to the present embodiment.

[0014] According to the present embodiment, a four-cylinder four-stroke gasoline engine is described as an example of the internal combustion engine 100, but the number of cylinders and cycles of the internal combustion engine 100 are not limited thereto.

[0015] As in Fig. As shown in Figure 1, the internal combustion engine 100 draws air into a cylinder 102 through an intake passage 101. In the cylinder 102, a piston 104 coupled to a crankshaft 103 moves vertically in synchronization with the rotation of the crankshaft 103, and an intake valve 105 and an exhaust valve 106 open and close in synchronization with this movement. Air is drawn into the cylinder 102 through the synchronization between the vertical movement of the piston 104 and the opening and closing of the intake valve 105 and the exhaust valve 106.

[0016] The intake air quantity taken into the cylinder 102 is adjusted by adjusting the opening of a throttle valve 107 provided in the intake pipe 101 based on the driver's accelerator pedal operation. The intake air quantity is measured by an air flow sensor 108 provided in the intake pipe 101. The measured intake air quantity is divided by a target air-fuel ratio determined by engine speed, intake pipe pressure, and the like, thereby calculating a target fuel injection quantity, and fuel is injected from an injector 109 according to the target fuel injection quantity.

[0017] The air-fuel mixture composed of air taken into the cylinder 102 and fuel injected from the injector 109 is ignited by a spark plug 110, causing the air-fuel mixture to explode. The expanded air-fuel mixture pushes the piston 104 down, and the downward movement of the piston 104 is converted into rotation of the crankshaft 103, which becomes the driving force of the vehicle. An EGR pipe 112 is provided from an exhaust passage 111 to the intake passage 101. Pumping loss can be reduced by returning the burned air-fuel mixture to the intake passage 101. The throttle valve 107, the injector 109, and the spark plug 110 are controlled by the control device 1 connected to the internal combustion engine 100. The control device 1 controls the air-fuel ratio and the ignition timing by controlling the operating state and the ambient state of the internal combustion engine 100.

[0018] As in Fig. 2, the internal combustion engine 100 is provided with four cylinders 102 connected in series.

[0019] According to the present embodiment, a first cylinder 1021, a second cylinder 1022, a third cylinder 1023, and a fourth cylinder 1024 are provided in this order from the side close to the throttle valve 107. Here, in the internal combustion engine 100, a difference occurs between the amount of air taken in from the intake pipe 101 and the amount of exhaust gas taken in from the EGR pipe 112 between a cylinder close to the throttle valve 107 (for example, the cylinder 1021) and a cylinder far from the throttle valve 107 (for example, the cylinder 1024).

[0020] Therefore, in the internal combustion engine 100, the combustion stability differs between the cylinders 1021 to 1024 even when the same amount of fuel is injected from a fuel injector 109 provided for each of the cylinders 1021 to 1024. Conventionally, the fuel consumption and exhaust emission performance of an internal combustion engine are within an acceptable range even if the difference in combustion stability between cylinders is ignored. However, there is an increasing demand for correcting the difference in combustion stability between cylinders in conjunction with a demand for further improving the fuel consumption and exhaust emission performance in lean combustion, EGR combustion, and the like of the internal combustion engine.

[0021] Therefore, the internal combustion engine 100 according to the present embodiment is provided with a cylinder pressure sensor 113 (see Fig. 1) for each of the cylinders 1021 to 1024 to detect the combustion state of the respective cylinders 1021 to 1024. Fig. 3 shows the relationship between the cylinder pressure Pcyl of each of the cylinders 1021 to 1024 measured by the cylinder pressure sensor 113 and the rotation angle (crank angle θ) of the crankshaft 103 detected by a crank angle sensor 1031. Fig. Figure 4 shows the relationship between the cylinder pressure Pcyl and the volume V in the cylinder 102.

[0022] In Fig. 3, the horizontal axis represents the crank angle θ, and the vertical axis represents the cylinder pressure Pcyl. In the internal combustion engine 100, the piston 104 reciprocates between a top dead center (TDC) and a bottom dead center (BDC) twice in one combustion cycle (the crankshaft 103 rotates 720 degrees), and the four strokes of an intake stroke, a compression stroke, a combustion (explosion) stroke, and an exhaust stroke are executed during this period.

[0023] In Fig. 4, the horizontal axis represents the volume V of the cylinder 102 and the vertical axis represents the cylinder pressure Pcyl. In the internal combustion engine 100, the amount of work W performed by a cylinder 102 in one combustion cycle can be calculated as follows according to the area (hatched section in Fig. 4), which is formed by the four strokes executed in one combustion cycle, can be expressed by equation 1. W=∫InletTDCInletTDC+720DegreePcyldV

[0024] The amount of work W / V per unit volume, obtained by dividing the amount of work W for one combustion cycle of a cylinder by the volume V of the cylinder, is called the indicated mean effective pressure (IMEP). The IMEP is widely used as a value representing the combustion energy of the internal combustion engine 100.

[0025] Fig. Figure 5 is a graph showing the change in IMEP (combustion energy) calculated per combustion cycle in a cylinder. For the purposes of explanation, Fig. 5 Changes of IMEP1 (solid line in the figure) of the first cylinder 1021 and IMEP2 (dashed line in the figure) of the second cylinder 1022 from cylinders 1021 to 1024. In Fig. 5, the IMEP is large in a period from 0 to 50 cycles. It is indicated that the fluctuation of the IMEP is small in this period because the load of the internal combustion engine 100 is high. Furthermore, it is indicated that the load of the internal combustion engine 100 gradually decreases in a period from 80 to 180 cycles, and the fluctuation of the IMEP per combustion cycle is small. In a period from 180 to 300 cycles, the IMEP is small. That is, in this period, the load of the internal combustion engine 100 decreases, and the fluctuation of IMEP1 of the first cylinder 1021 per combustion cycle increases. Accordingly, it is seen that the combustion in the first cylinder 1021 becomes unstable in the period from 180 to 300 cycles.

[0026] To quantify this instability, a method is carried out in which combustion stability is assessed using a parameter cPi calculated from the mean µ and standard deviation σ of the IMEP of several previous combustion cycles. This parameter cPi can be expressed by Equation 2 below. In this method, it is assumed that averaging over several tens to hundreds of cycles is used to assess combustion stability. That is, the cPi per cycle is calculated using the mean µ and standard deviation σ of the IMEP in several tens to hundreds of previous cycles. It is assumed that combustion is stable if the value of cPi is less than or equal to a threshold (a set threshold), and that combustion is unstable if the value of cPi exceeds the set threshold. cPi=σ(IMEP)μ(IMEP)

[0027] Fig. 6 shows the time series of the IMEP from Fig. 5 using equation 2 calculated cPi. In Fig. 6, the horizontal axis represents the combustion cycle and the vertical axis represents the parameter cPi described above. In Fig. 6, the cPi of the first cylinder 1021 is represented by cPi1 (solid line in the figure), and the cPi of the second cylinder 1022 is represented by cPi2 (dashed line in the figure). Fig. 6 is described below. The specified threshold value of cPi for assessing combustion stability is described in the following description 2 (1) First, in the period from 0 to 50 cycles in Fig. 6, the value of cPi for the first cylinder 1021 and the second cylinder 1022 is less than or equal to 2, respectively. Accordingly, it can be determined that the combustion states of the first cylinder 1021 and the second cylinder 1022 are both stable in this cycle period. Waveforms (see Fig. 5) of IMEP1 and IMEP2 before calculating cPi indicate that the fluctuation of IMEP1 and IMEP2 per combustion cycle is small in this period. Accordingly, the result of the determination based on cPi1 and cPi2 as described above, that the combustion in the first cylinder 1021 and the second cylinder 1022 is stable, is considered reasonable. (2) Next, in the period from 180 to 300 cycles, the value of cPi2 of the second cylinder 1022 is less than or equal to 2. This means that it can be determined that the combustion state of the second cylinder 1022 is stable in this period. On the other hand, the value of cPi1 of the first cylinder 1021 exceeds 2, which is judged to be unstable. Waveforms (see Fig. 5) of IMEP1 and IMEP2 before cPi is calculated indicate that the fluctuation of IMEP2 per combustion cycle in this period is small, while the fluctuation of IMEP1 per combustion cycle in this period is large. Accordingly, the result of the determination made based on cPi1 and cPi2 as described above, that the combustion of the second cylinder 1022 is stable and the combustion of the first cylinder 1021 is unstable, is considered reasonable. (3) The problem discussed here concerns the transient state (transient operation) that occurs in the period specified by cycles 80 to 180. In the period of this transient state, IMEP1 and IMEP2 have decreased as a result of a transition from a state where, for example, the engine speed and torque are high to a state where they are small. The original waveform of the IMEP (see Fig. 5) indicates that the combustion state is stable because the fluctuation is small, although both the IMEP1 of the first cylinder 1021 and the IMEP2 of the second cylinder 1022 fluctuate because they are in the transient state. In Fig. 6, however, both cPi1 of the first cylinder 1021 and cPi2 of the second cylinder 1022 exceed 2, which is the set threshold. Accordingly, by the method in which combustion is judged to be unstable if cPi is greater than the set threshold (2 in this case), combustion in the first cylinder 1021 and the second cylinder 1022 is judged to be unstable, although it is actually stable, as described above.

[0028] As described above, according to the present embodiment, attention is paid to a problem in a method for judging combustion stability based on a comparison between cPi and a predetermined threshold in a transient state occurring between 80 and 180 cycles as described above. In other words, an object of the present embodiment is to prevent combustion from being judged as unstable even though it is stable in a transient state, and to accurately determine combustion stability even in the transient state.

[0029] Next, the reason why the combustion stability is determined to be unstable even if it is stable in a transient state when using a combustion stability judgment method based on a comparison between cPi and the set threshold value will be explained in detail with reference to the Fig. 7 to 9.

[0030] Fig. 7 shows a distribution of IMEP (combustion energy) of multiple combustion cycles in a steady state (equilibrium operation), the mean value μ of IMEP of the multiple combustion cycles, and the standard deviation σ of the value of each IMEP from the mean value μ. As indicated in the above equation 2, the parameter cPi during the steady state operation of the internal combustion engine 100 is calculated by dividing the standard deviation σ of the value of each IMEP from the mean value μ of the IMEP at the set number of combustion cycles from a few tens to hundreds of cycles in the past by the mean value μ.

[0031] Fig. Figure 8 shows the distribution of IMEP in multiple combustion cycles in the transient state, its mean value µ, and the standard deviation σ from the mean value µ. As described above, cPi is calculated by dividing the standard deviation σ of the value of each IMEP from the mean value µ of the IMEP at the specified number of combustion cycles of several tens to hundreds of cycles in the past by the mean value µ. As described above, the transient state indicates a case where a transition occurs from a state where, for example, the engine speed and torque are high to a state where they are small. That is, in this transient state, a slight change occurs so that the IMEP, which is the combustion energy, transitions from a high value to a low value, or vice versa, due to a fluctuation in the engine speed and torque, for example.

[0032] Despite this slight change, the mean value µ of the IMEP remains constant in the multiple combustion cycles, so the standard deviation σ of each IMEP from the constant mean value µ exhibits the effect of a slight change and is larger than the actual combustion fluctuation. This means that cPi in the transient state is calculated to be larger by the amount of the slight change. Accordingly, the method of judging combustion stability based on a comparison between cPi and the set threshold value as described above always judges combustion in the transient state to be unstable. In other words, this method has the problem that combustion stability cannot be accurately judged.

[0033] Therefore, as in Fig. 9, according to the present embodiment, the combustion energy change trend (IMEP) is evaluated multiple times. This combustion energy trend can be referred to as a combustion energy trend. That is, according to the present embodiment, the distribution of the combustion energy difference in each cycle is evaluated not from the average value µ of the combustion energy during the transient operation, but from a straight line (approximate line) indicating the combustion energy change trend in multiple combustion cycles. The inventors of the present invention have made efforts and found that combustion stability can be accurately evaluated by using the combustion energy change trend.

[0034] Fig. 10 is a graph in which Neu_cPi is plotted. Neu_cPi is obtained by calculating an index value ρ of the distribution of the difference from a straight line, which indicates the trend of the change in combustion energy (IMEP) in several combustion cycles, instead of the mean value µ of IMEP from the time series of IMEP. Fig. 5 and dividing the index value ρ by the mean value µ. Accordingly, according to the present embodiment, an evaluation index of combustion stability is obtained by using the index value ρ of the distribution of the difference from a straight line, thereby indicating the trend of the change in combustion energy in a plurality of combustion cycles.

[0035] As described above, because the combustion energy (IMEP) in the period from 80 to 180 cycles in the transient state decreases in Fig. 5 changes slightly, the standard deviation σ of the value of each combustion energy from the mean value µ of the combustion energy increases as described above, and then it is determined that the combustion is unstable.

[0036] On the other hand, in the method of calculating the index value ρ of the distribution of the difference from the trend of the change in the combustion energy in the plurality of combustion cycles according to the present embodiment, it can be correctly determined that the combustion is stable without being influenced by the change due to the transition in such a transient state.

[0037] That is, according to the present embodiment, the combustion stability can be accurately judged. [Control device configuration]

[0038] Fig. 11 shows the configuration of the control device 1 for implementing the combustion stability judgment according to the present embodiment. Each block in the diagram of Fig. 11 is a functional block of the control device 1 according to the present embodiment.

[0039] The control device 1 according to the present embodiment includes a combustion energy calculation unit 210 that calculates the combustion energy of each combustion cycle of the internal combustion engine 100. The cylinder pressure Pcyl detected by the cylinder pressure sensor 113 and the crank angle θ (also referred to as the rotation angle) of the crankshaft 103 detected by the crank angle sensor 1031 are input to the combustion energy calculation unit 210 per combustion cycle.

[0040] The control device 1 according to the present embodiment includes a trend calculation unit 230 that calculates a trend of change in the combustion energy calculated by the combustion energy calculation unit 210 in a plurality of combustion cycles, and a combustion stability judgment unit 250 that determines the combustion stability based on the combustion energy in the plurality of combustion cycles and the trend of change calculated by the trend calculation unit 230.

[0041] The control device 1 according to the present embodiment includes a difference calculation unit 240 that calculates the difference ε between the trend of change (Equation 5) of the combustion energy in the plurality of combustion cycles calculated by the trend calculation unit 230 and the combustion energy per combustion cycle calculated by the combustion energy calculation unit 210 (combustion parameter calculation unit), and the combustion stability judgment unit 250 determines the combustion stability based on the difference ε. The combustion energy calculated by the combustion energy calculation unit 210 is stored in a storage unit 220 (memory), and the trend calculation unit 230 and the combustion stability judgment unit 250 implement the above content using the respective combustion energy in the plurality of combustion cycles stored in the storage unit 220. [Determination method performed by the control device]

[0042] Next, a method for judging the combustion state according to the present embodiment will be described based on the above-described configuration of the control device 1.

[0043] Fig. 12 is a flowchart of the combustion state determination process executed by the control device 1. First, in step S301, the combustion energy calculation unit 210 starts calculating the combustion energy in a case where the piston 104 is at the top dead center (TDC) in the intake stroke based on the crank angle θ (rotation angle) of the crankshaft 103 detected by the crank angle sensor 1031. Then, the combustion energy calculation unit 210 initializes the combustion energy in the case of the TDC in the intake stroke as indicated by Equation 3 below. W=0

[0044] The combustion energy is calculated based on Equation 1 described above. If Equation 1 is expressed in discrete time, the combustion energy can be expressed by Equation 4 below. In Equation 1, IMEP, which is the combustion energy, is used as one of the combustion parameters. W=W_old+Pcyl×ΔV

[0045] The combustion energy calculation unit 210 detects the cylinder pressure Pcyl per cylinder 102 through the cylinder pressure sensor 113 per falling timing of an output signal from the crank angle sensor 1031, and calculates the amount ΔV of increase in the volume V in the cylinder 102 based on the change in the crank angle θ. Then, the combustion energy calculation unit 210 calculates the combustion energy per falling time of the output signal from the crank angle sensor 1031 by adding the product of the cylinder pressure Pcyl and the amount ΔV of increase in the volume V in the cylinder 102 to the work amount W_old calculated at the falling time of the previous output signal from the crank angle sensor 1031.

[0046] In step S302, the combustion energy calculation unit 210 calculates the combustion energy for one combustion cycle while the crank angle θ changes from the position of the TDC in the intake stroke, in which the calculation of the combustion energy is initiated, to the position of the TDC in the intake stroke after 720 degrees (two rotations of the crankshaft).

[0047] When this is detected by an output signal from the crank angle sensor 1031, the calculation of the combustion energy is completed and the calculated combustion energy for one combustion cycle is stored in the storage unit 220.

[0048] The storage unit 220 stores the combustion energy W_t of several to several tens of previous combustion cycles. W_t indicates the IMEP (combustion energy) in the t-th combustion cycle obtained by the above method.

[0049] In step S303, the trend calculation unit 230 calculates the trend of the change in combustion energy based on the distribution of combustion energy W_t of several to several tens of previous combustion cycles stored in the storage unit 220. Assuming that the combustion energy is plotted in the order of combustion cycles as shown in Fig. 9, the trend Tr of the change in combustion energy is given by the following equation 5. Tr=at+b

[0050] Then, the trend calculation unit 230 calculates the combustion energy change trend Tr (IMEP) by obtaining a and b in Equation 5. That is, the combustion energy change trend Tr is an index indicating how the combustion energy changes in the Fig. 5, and in other words, it is an approximate expression in a case where the Fig. 5 is approximately a straight line or the like. It can be said that the trend calculation unit 230 calculates the trend of the change in combustion energy by approximating the distribution of combustion energy in a plurality of combustion cycles by a linear function. The coefficients a and b of the trend Tr of the change in combustion energy can be calculated using the least squares method for the distribution of combustion energy W_t in the several to several tens of combustion cycles, for example, as shown in Fig. 5. That is, Equation 5 expresses the distribution of combustion energy in the multiple combustion cycles as an approximate expression of a linear function, which can be said to indicate the trend of change in combustion energy.

[0051] Next, in step S304, the difference calculation unit 240 calculates the difference ε_t from the trend Tr of the change in the combustion energy W_t calculated per combustion cycle in the several to several tens of combustion cycles based on the following equation 6. The difference ε_t is obtained for each of the plurality of combustion cycles, and the distribution of the combustion energy W_t can be judged by considering the trend Tr of the change in the combustion energy. ε_t=W_t−at−b

[0052] Then, in step S305, the difference calculation unit 240 calculates the sum of the squares of the differences ε in the several to several tens of combustion cycles with the following Equation 7 using the difference ε_t of the combustion energy W_t calculated in step S304. In Equation 7, T represents the number of combustion cycles for judging combustion stability. That is, Equation 7 can be referred to as the sum of the squares of the differences ε from the trend Tr of the change in the combustion energy W_t, taking into account the trend Tr of the change in the combustion energy in the plurality of combustion cycles. ∑t=1Tε_t2

[0053] The combustion stability judging unit 250 performs a stability judgment of the combustion state of the internal combustion engine 100 based on the sum of the squares of the differences ε from the trend Tr of the change in combustion energy W_t in the plurality of combustion cycles calculated in the above equation 7, and ends the processing. That is, the combustion stability judging unit 250 judges the combustion stability based on the difference ε calculated by the difference calculating unit 240, as described above. Specifically, the combustion stability judging unit 250 compares the sum of the squares of the differences ε from the trend Tr of the change in combustion energy W_t in the plurality of combustion cycles calculated using the equation 7, or the quotient obtained by dividing the sum by the number T of combustion cycles, with a predetermined threshold value.If the sum of the squares of the differences ε or the quotient obtained by dividing the sum by the number T of combustion cycles is less than or equal to the set threshold, the combustion stability judging unit 250 determines that the combustion in the plurality of combustion cycles is stable. Conversely, if the sum of the squares of the differences ε exceeds the set threshold, the combustion stability judging unit 250 determines that the combustion is unstable. The sum of the squares of the differences ε of the trend Tr of the change in combustion energy W_t in the plurality of combustion cycles calculated by Equation 7 or the quotient obtained by dividing the sum by the number T of combustion cycles varies depending on the engine load, so here the set threshold needs to be changed depending on the engine load.

[0054] The combustion stability of the internal combustion engine 100 can be judged (evaluated) based on values ​​calculated based on Equations 8 to 10 below, instead of the sum of the differences ε calculated by Equation 7 as described above. Here, a deviation p of Equation 8 is obtained by dividing the sum (Equation 7) of the squares of the differences ε from the trend Tr of the change in combustion energy W_t in the plurality of combustion cycles by the number T of combustion cycles and calculating the root of the quotient. The mean value µ of Equation 9 is obtained by calculating the sum of combustion energy W_t in the plurality of combustion cycles as described above and dividing the calculated sum by the number T of combustion cycles.Neu_cPi in Equation 10 is obtained by dividing the index value p of the distribution of differences ε in Equation 8 by the mean value µ in Equation 9. The present invention can also be realized by setting a threshold value for judging combustion stability with respect to Neu_cPi in Equation 10.

[0055] In this case, the combustion stability judgment unit 250 compares Neu_cPi in Equation 10 with the predetermined threshold. The combustion stability judgment unit 250 determines that the combustion is stable in the plurality of combustion cycles if Neu_cPi is less than or equal to the predetermined threshold, and conversely, determines that the combustion is unstable if Neu_cPi exceeds the predetermined threshold. σ=∑t=1Tε_t2T μ=∑t=1TW_tT New_cPi=σμ

[0056] As described above, Fig. 10 is a graph in which Neu_cPi, as obtained by equations 8 to 10, corresponds to the time series of the IMEP from Fig. 5. Accordingly, according to the present embodiment, it can be correctly determined that the combustion state is stable because Neu_cPi of the combustion energy W_t is small in the transient state and in the period from 80 to 180 cycles in which the combustion state is stable. Therefore, according to the present embodiment, the combustion stability can be accurately judged even in the transient state. [Second embodiment]

[0057] A second embodiment of the present invention will be described below with reference to the drawings. Equation 2 described in the first embodiment or the equation in Fig. 6 represents the quotient obtained by dividing the index value ρ of the distribution of the difference from the mean µ by the mean µ, as in Fig. 7 or Fig. 8. This means that in equation 2, Fig. 6, Fig. 7 or Fig. 8, the combustion stability is evaluated based on the value calculated by Equation 11 below. This Equation 11 calculates the difference of the combustion energy W_t calculated per combustion cycle in the multiple combustion cycles from the average value µ of the combustion energy in the multiple combustion cycles and expresses the sum of the squares of the differences in the multiple combustion cycles T. That is, the trend Tr of the change in the combustion energy W_t is not taken into account because it is the difference from the average value µ. ∑t=1T(W_t−μ)2

[0058] The following discusses the relationship between the combustion stability based on Σ(ε_t)^2 (Equation 7), where the effect of the trend Tr of the change in the combustion energy W_t is removed as proposed in the first embodiment, and the combustion stability based on Σ(W_t-µ)^2, where the effect of the trend Tr of the change achieved with cPi is removed, as expressed by Equations 2 and 11. In Equation 6 described above, ε_t on the left side is set to 0 and the mean value of the right side is obtained, thereby obtaining Equation 12 below. 0=μ−aT+12−b

[0059] By subtracting both sides of Equation 12 from both sides of Equation 6 described above, Equation 13 below is obtained. W_t−μ=ε_t+a(t−T+12)

[0060] Therefore, the total sum of combustion cycles t = 1 to T in Equation 13 can be expressed by Equation 14 below.

[0061] T indicates the number of combustion cycles to assess combustion stability. ∑t=1T(W_t−μ)2=∑t=1T{ε_t+a(t−T+12)}2

[0062] Since ε_t and a × {t - (T + 1) / 2} are independent of each other in Equation 14 described above, Equation 15 below is derived. ∑t=1T(W_t−μ)2=∑t=1Tε_t2+a2∑t=1T(t−T+12)2

[0063] In Equation 15, the sum of the squares of the differences ε from the trend Tr of the change in combustion energy W_t in the plurality of combustion cycles as expressed in Equation 7 is expressed by the first term on the right side of Equation 15. The left side of Equation 15 is the sum of the squares of the difference in combustion energy W_t from the mean µ of the combustion energy W_t (IMEP) in the plurality of combustion cycles as expressed in Equation 11, which is an index of the distribution of the mean µ. Furthermore, the second term on the right side of Equation 15 expresses an equation obtained by multiplying the square of the slope a of the trend Tr of the change in combustion energy W_t by a constant, using a period (number of combustion cycles for combustion stability judgment) T to calculate the trend of the change in combustion energy.

[0064] From the above, the sum (first term on the right side of Equation 15) of the square of the difference ε from the trend Tr of the change in combustion energy W_t in the several combustion cycles, as expressed by Equation 7, can be obtained by subtracting the value (second term on the right side of Fig. 15) based on the constant obtained from the slope a of the trend Tr of the change in the combustion energy W_t and the number T of combustion cycles from the index (left side of Equation 15, Equation 11) of the distribution of the combustion energy W_t, the mean value µ of the combustion energy W_t (IMEP) in the plurality of combustion cycles.

[0065] In view of this, the means for solving the problem according to the present embodiment will be described below. [Control device configuration]

[0066] Fig. 13 shows the configuration of the control device 1A for implementing the combustion stability judgment according to the present embodiment. Each block in the diagram of Fig. 13 is a functional block of the control device 1A according to the present embodiment.

[0067] The control device 1A according to the present embodiment includes a combustion energy calculation unit 410 that calculates the combustion energy of each combustion cycle of the internal combustion engine 100, and a trend calculation unit 430 that calculates the trend of the change in the combustion energy calculated by the combustion energy calculation unit 410 in a plurality of combustion cycles.The control device 1A according to the present embodiment includes a variance calculation unit 440 that calculates the variance of the combustion energy based on the combustion energy (IMEP) in the plurality of combustion cycles, and a combustion stability judgment unit 470 that judges the combustion stability based on the trend of the change in the combustion energy in the plurality of combustion cycles calculated by the trend calculation unit 430 and the variance of the combustion energy (IMEP) calculated by the variance calculation unit 440.

[0068] Hereinafter, a method for judging the combustion state by the control device 1A will be described with reference to the flowchart of Fig. 14 described. [Judgment method performed by the control device]

[0069] First, the combustion energy calculation unit 410 derives in step S501 Fig. 14, the calculation of the combustion energy in the case where the piston 104 is at the TDC position in the intake stroke based on the crank angle θ (rotation angle) of the crankshaft 103 detected by the crank angle sensor 1031. Because the combustion energy calculation process executed at the combustion energy calculation unit 410 is similar to the combustion energy calculation process executed by the combustion energy calculation unit 210 according to the first embodiment (see step S301 of Fig. 12), a detailed description is omitted.

[0070] In step S502, when it is determined that the crank angle θ is the TDC of the intake stroke that is 720 degrees behind the TDC of the intake stroke in which the calculation of the combustion energy was initiated, the combustion energy calculation unit 410 calculates the combustion energy for one combustion cycle and ends the calculation of the combustion energy.

[0071] The combustion energy calculated by the combustion energy calculation unit 410 is stored in a storage unit 420 (a memory). The storage unit 420 stores the combustion energy of several previous cycles to several tens of previous cycles.

[0072] In step S503, the variance calculation unit 440 calculates the variance (the quotient obtained by dividing the left side of Equations 11 or 15 by T) of the mean value µ of the combustion energy of the previous plurality of combustion cycles stored in the storage unit 420. That is, the variance calculation unit 440 obtains the difference of the combustion energy W_t calculated per combustion cycle in the plurality of combustion cycles from the mean value µ of the combustion energy W_t in the plurality of combustion cycles and obtains the mean value of the square of the difference in the plurality of combustion cycles. Equation 16 expresses the root (standard deviation σ) of the quotient obtained by dividing the sum of the values ​​on the left sides of Equations 11 and 15 by the number T of times of multiple combustion cycles. By dividing the value of Equation 16 by the mean value µ, cPi is obtained as described in the first embodiment. σ=∑t=1T(W_t−μ)2T

[0073] In step S504, the trend calculation unit 430 calculates the trend Tr of the change in the combustion energy W_t in the plurality of combustion cycles based on the distribution of the combustion energy W_t stored in the storage unit 420 in the plurality of previous combustion cycles. The method for calculating the trend Tr of the change in the combustion energy W_t by the trend calculation unit 430 is similar to the method for calculating the trend Tr of the change in the combustion energy W_t by the trend calculation unit 230 according to the first embodiment (see step S303). Fig. 12). This means that the trend calculation unit 430 calculates the Fig. 10 is represented as an approximate expression (at + b) of the primary straight line using the least squares method and the coefficients a and b are calculated, thereby obtaining the trend Tr of the change of the combustion energy W_t in the plurality of combustion cycles.

[0074] In step S505, an effect calculation unit 450 calculates the effect of the trend Tr of the change in combustion energy W_t on the variance of combustion energy W_t based on the slope a of the trend Tr of the change in combustion energy W_t calculated by the trend calculation unit 430 and the number T of combustion cycles in one period to calculate the trend Tr of the change in combustion energy W_t. Specifically, the effect calculation unit 450 can calculate the effect of the trend Tr of the change in combustion energy W_t on the variance of combustion energy W_t by obtaining the second term on the right side of Equation 15.

[0075] In step S506, an effect removal unit 460 removes the effect of the trend Tr of the change in combustion energy W_t on the variance of combustion energy W_t calculated by the effect calculation unit 450 in step S505, based on the quotient (variance of combustion energy W_t) obtained by dividing the sum of the squares of the differences from the mean value µ of combustion energy W_t in the plurality of combustion cycles, calculated by the variance calculation unit 440 in step S503, by T. Specifically, the effect removal unit 460 calculates the quotient obtained by dividing the sum of the squares of the differences ε from the trend Tr of the change in combustion energy W_t in the plurality of combustion cycles by T based on Equation 17 below.

[0076] In other words, the effect removal unit 460 removes the effect of the trend Tr of the change in the combustion energy W_t calculated by the effect calculation unit 450 by subtracting a contribution (a^2*(Σ(t-(T+1) / 2)^2) / T) due to the trend Tr of the change in the combustion energy W_t from the variance ((Σ(W_t-µ)^2) / T) calculated by the variance calculation unit 440 from the mean value µ of the combustion energy W_t in the plurality of combustion cycles. With this configuration, the effect removal unit 460 can calculate an index value (Σ(ε_t^2) / T) of the distribution of the combustion energy W_t from which the effect of the trend Tr of the change in the combustion energy W_t has been removed. ∑t=1Tε_t2T=∑t=1T(W_t−μ)T−a2∑t=1T(t−T+12)2T

[0077] Equation 17 corresponds to Equation 7 described in the first embodiment. Therefore, Equation 17 realizes the calculation corresponding to Equation 7 as described in the first embodiment from a different point of view. Accordingly, in step S507, the combustion stability judging unit 470 judges the combustion stability based on the index value Σ(ε_t^2) of the distribution of combustion energy W_t, from which the effect of the trend Tr of the change in combustion energy W_t has been removed by the effect removal unit 460, or the quotient (Σ(ε_t^2) / T) obtained by dividing the index value Σ(ε_t^2) by the combustion cycle T. Since this method is similar to that according to the first embodiment, a detailed description thereof will be omitted.As described above, in the transient operation of the internal combustion engine 100, the combustion stability judging unit 470 can appropriately judge the combustion stability based on the combustion energy distribution index value W_t after removing the effect of the combustion energy change trend (refer to FIG. Fig. 10). Furthermore, according to the present embodiment, the combustion stability can be evaluated by calculating Equation 17, so the computational effort is lower than that of the calculations of Equations 6 and 7 according to the first embodiment. Therefore, the present invention can be implemented even when the performance of a microcomputer of the control device is not high.

[0078] It is possible to obtain Neu_cPi by Equations 8 to 10 based on the index value (Σ(ε_t^2) / T) of the combustion energy distribution without the contribution of the combustion energy change trend obtained by Equation 17. However, the description thereof will be omitted because this method is similar to that according to the first embodiment. According to the first and second embodiments, the case where the combustion energy change trend Tr is calculated during the transient operation of the internal combustion engine 100 to perform the judgment of the stability of the combustion state during the transient operation was described as an example. However, the control device 1, 1A may perform the judgment of the stability of the combustion state after calculating the combustion energy change trend even during the steady state. [Third Embodiment]

[0079] A third embodiment of the present invention will be described below with reference to the drawings. The object of the first and second embodiments is to correctly judge the distribution of the change in combustion energy W_t at multiple specified times of combustion cycles. However, there is also a need for detecting a sudden change in combustion energy W_t in a combustion cycle.

[0080] Fig. 15 is a diagram for describing a state where a sudden change in combustion energy occurs during transient operation according to the present embodiment. The combustion energy change trend Tr in the multiple combustion cycles is obtained from the combustion energy distribution according to the first and second embodiments. Therefore, according to the present embodiment, a method for detecting a sudden combustion change using the change trend Tr will be described. [Control device configuration]

[0081] Fig. 16 shows the configuration of a sudden combustion change detection control device 1B according to the present embodiment. Each block in the diagram of Fig. 16 is a functional block of the control device 1B according to the present embodiment.

[0082] Similar to the first and second embodiments, the control device 1B according to the present embodiment includes a combustion energy calculation unit 610 that calculates the combustion energy of each combustion cycle of the internal combustion engine 100, a storage unit 620 that stores the combustion energy of a plurality of previous combustion cycles, and a trend calculation unit 630 that calculates a trend of the change in the combustion energy calculated by the combustion energy calculation unit 610 in the plurality of combustion cycles. [Judgment method performed by the control device]

[0083] Next, a method for judging a sudden combustion change by the control device 1B according to the present embodiment will be described. Fig. 17 is a flowchart of a combustion state judgment method by the control device 1B. First, in step S701, the combustion energy calculation unit 610 initiates the calculation of combustion energy in the case where the piston 104 is located at the TDC position in the intake stroke based on the crank angle θ (rotation angle) of the crankshaft 103 detected by the crank angle sensor 1031. The method of calculating combustion energy by the combustion energy calculation unit 610 is similar to that according to the first and second embodiments, and therefore, its description is omitted.

[0084] In step S702, when it is determined that the crank angle θ is the TDC of the intake stroke that is 720 degrees behind the TDC of the intake stroke in which the calculation of the combustion energy was initiated, the combustion energy calculation unit 610 calculates the combustion energy for one combustion cycle and ends the calculation of the combustion energy.

[0085] The calculated combustion energy for one combustion cycle is stored in the storage unit 620, and the storage unit 620 stores the combustion energy of several to several tens of previous cycles.

[0086] In step S703, the trend calculation unit 630 calculates the combustion energy change trend Tr based on the combustion energy distribution W_t in the previous several combustion cycles (several to several tens) stored in the storage unit 620. The method for calculating the combustion energy change trend Tr by the trend calculation unit 630 is similar to the method for calculating the combustion energy change trend Tr by the trend calculation unit described in the first and second embodiments, so its description is omitted.

[0087] In step S704, a difference calculation unit 640 calculates the difference ΔW_t (at + b - Wn) between the trend Tr (at + b) of the change in combustion energy calculated by the trend calculation unit 630 in step S703 and the combustion energy W_t calculated by the combustion energy calculation unit 610 in step S701.

[0088] Then, in step S705, a combustion sudden change judgment unit 650 (also referred to as a sudden fluctuation judgment unit) judges whether the difference ΔW_t (at + b - Wn) calculated by the difference calculation unit 640 in step S704 exceeds a set threshold ΔWh. If it is determined that the difference ΔW_t (at + b - Wn) exceeds the set threshold ΔWh (ΔW_t (at + b - Wn) > ΔWh), the combustion sudden change judgment unit 650 determines that the combustion energy has suddenly changed in the combustion cycle. On the other hand, when the sudden change in combustion judgment unit 650 determines that the difference ΔW_t is less than or equal to the set threshold ΔWh (ΔW_t (at + b - Wn) ≤ ΔWh), it determines that there is no sudden change in the combustion energy in the combustion cycle.This allows sudden fluctuations in combustion energy to be detected.

[0089] Although the above embodiments have been described based on variations in combustion energy for judging the stability of the combustion state by the control devices 1, 1A, 1B, the combustion parameters for judging the stability of the combustion state are not limited thereto. Fig. 18 is a graph showing a change in cylinder pressure in one combustion cycle. Here, the stability of the combustion state can be judged based on the width of the crank angle θPmax distribution, assuming that combustion becomes maximum at a crank angle θPmax at which the cylinder pressure becomes maximum. When the combustion state of the cylinder of the internal combustion engine 100 is stable, the width of the crank angle θPmax distribution at which combustion becomes maximum falls within a predetermined set range. On the other hand, when the combustion state of the internal combustion engine 100 is unstable, the width of the crank angle θPmax distribution at which combustion becomes maximum exceeds the predetermined set range.

[0090] Accordingly, the combustion stability judging unit (250, 470) of the control device (1, 1A, 1B) can judge (evaluate) the stability of the combustion state of the internal combustion engine 100 based on the width of the distribution of the crank angle θPmax at which combustion becomes maximum as a judgment parameter. θPmax is also referred to as the combustion timing. By paying attention to the combustion timing in this way, the stability of the combustion state can be correctly judged (evaluated) even in the transient state, as in the first and second embodiments.

[0091] Fig. Figure 19 shows the relationship between a heat quantity Q in a combustion cycle and a corresponding crank angle. CA10 is the crank angle at the time when the combustion ratio reaches 10% of the maximum, so that the heat quantity generated is 10% of the maximum value Qmax of the heat quantity generated.

[0092] CA50 is the crank angle at the time when the combustion ratio becomes 50% of the maximum, so that the generated heat amount is 50% of the maximum value Qmax of the generated heat amount. Here, the control device (1, 1A, 1B) includes a combustion rate calculation unit that calculates the combustion rate in one combustion cycle based on CA10 or CA50. For example, the combustion rate calculation unit can calculate the combustion rate by calculating a period (CA50 - CA10) ranging from the crank angle CA10 where Qmax × 0.1 holds to the crank angle CA50 where Qmax × 0.5 holds, which is 50% of the maximum value Qmax.

[0093] The combustion stability judging unit (250, 470) of the control device (1, 1A, 1B) determines that the combustion state is stable if the period (combustion rate: CA50 - CA10) is within a predetermined set range, and determines that the combustion state is unstable if the period is beyond the predetermined set range. This allows the stability of the combustion state to be correctly judged (evaluated) in the transient state, as in the first and second embodiments.

[0094] As described above, according to the first and second embodiments, combustion stability is evaluated by evaluating the distribution of combustion energy. However, in addition to the combustion energy in each combustion cycle, the combustion parameter for evaluating combustion stability may be the peak position θPmax of combustion (ie, combustion time) or the length of the period in which a certain amount of heat is generated (ie, combustion rate).

[0095] In addition, the distribution of the combustion parameters described above is detected, and if the distribution width exceeds an allowable value, errors such as excessive vibration of the internal combustion engine 100 and misfire occur. Therefore, in the above embodiments, when it is determined that combustion has become unstable due to a wide distribution width of the combustion parameters, or when it is determined that the combustion parameters have suddenly changed, it is desirable to control the injector to increase fuel injection and thereby increase the air-fuel ratio. In this way, combustion can be stabilized.

[0096] Additionally, to quickly warm up an exhaust catalyst at startup time, the conversion of heat generated in the cylinder 102 can be controlled by retarding (postponing) the ignition timing so that the heat generated in the cylinder 102 is converted to exhaust heat to a greater extent than to work applied to the piston 104. The more the ignition timing is retarded, the faster the catalyst heats up, but the more unstable the combustion becomes. Therefore, in the above embodiments, it is desirable to control the spark plug so that the retardation of the ignition timing is reduced when it is determined that combustion has become unstable due to a large variation in the combustion parameters or when it is determined that the combustion parameters have suddenly changed. This makes it possible to stabilize the combustion.

[0097] As described above, the internal combustion engine control device 1, 1A, 1B described in the above embodiments includes a control unit (a microcomputer) that controls either the air-fuel ratio of the internal combustion engine or the ignition timing based on the combustion stability calculated by the combustion stability judging unit (250, 470, 650).

[0098] The above embodiments have been described using an example in which the internal combustion engine control device 1, 1A, 1B is applied to the internal combustion engine 100 for a vehicle. However, the present invention is not limited thereto and can also be applied to internal combustion engines for ships, aircraft, and various other types of equipment. Furthermore, the present invention can be realized by combining all of the above-described embodiments or by combining any two embodiments. Furthermore, the present invention is not limited to having all of the configurations of the above embodiments, and part of the configuration of one embodiment may be replaced with the configuration of another embodiment.Moreover, a portion of the configuration of one embodiment may be added to, removed from, or replaced by the configuration of another embodiment. List of reference symbols 1 control device 100 combustion engine 101 Inlet line 102 cylinders 1021 first cylinder 1022 second cylinder 1023 third cylinder 1024 fourth cylinder 103 Crankshaft 1031 Crank angle sensor 1032 memory board 104 pistons 105 Inlet valve 105A intake manifold 106 Exhaust valve 106A Outlet nozzle 107 throttle valve 108 Air flow sensor 109 Fuel injection device 110 Spark plug 111 Outlet line 112 EGR pipe 113 Cylinder pressure sensor 210 Combustion energy calculation unit 220 storage unit 230 Trend calculation unit 240 Difference calculation unit 250 Combustion Stability Assessment Unit

Claims

[1] Internal combustion engine control device (1A) comprising: a combustion parameter calculation unit (410) that calculates a combustion parameter of a combustion cycle in an internal combustion engine (100), a trend calculation unit (430) that calculates a trend of the change of the combustion parameter calculated by the combustion parameter calculation unit (410) in a plurality of combustion cycles, a variance calculation unit (440) which calculates a variance of the combustion parameter based on the combustion parameter in the plurality of combustion cycles, and a combustion stability judging unit (470) which judges combustion stability based on the trend of the change in the combustion parameter in the plurality of combustion cycles calculated by the trend calculating unit (430) and the variance of the combustion parameter calculated by the variance calculating unit (440), wherein the variance calculation unit (440) calculates the variance of the combustion parameter in the plurality of combustion cycles from an average value of the combustion parameters in the plurality of combustion cycles, characterized by an effect removal unit (460) that removes an effect of a trend of change in a combustion parameter by subtracting a distribution due to the trend of change in the combustion parameter calculated by the trend calculation unit (430) from the variance of the combustion parameter from the mean value calculated by the variance calculation unit (440). [2] The internal combustion engine control device (1A) according to claim 1, wherein the trend calculation unit (430) calculates the trend of the change of the combustion parameter by approximating a distribution of the combustion parameter in the plurality of combustion cycles by a linear function. [3] The internal combustion engine control device (1A) according to claim 1, wherein the combustion stability judging unit (470) judges the combustion stability based on an index value of a distribution of a combustion parameter from which the effect of the trend of change of the combustion parameter has been removed by the effect removing unit (460). [4] The internal combustion engine control device (1A) according to claim 1, wherein the combustion parameter is the combustion energy, the combustion timing or the combustion speed in the combustion cycle. [5] An internal combustion engine control device (1A) according to claim 1, comprising a control unit that controls either the air-fuel ratio of the internal combustion engine (100) or the ignition timing based on the combustion stability judged by the combustion stability judging unit (470). [6] Internal combustion engine control method comprising: a combustion parameter calculation step (S501) in which a combustion parameter of a combustion cycle in an internal combustion engine (100) is calculated, a variance calculation step (S503) in which a variance of the combustion parameter is calculated based on the combustion parameter in the plurality of combustion cycles, wherein the variance calculation step (S503) calculates the variance of the combustion parameter in the plurality of combustion cycles from an average value of the combustion parameters in the plurality of combustion cycles, a trend calculation step (S504) in which a trend of the change of the combustion parameter in a plurality of combustion cycles is calculated, an effect removal step (S506) in which an effect of a trend of change in a combustion parameter is removed from the mean value by subtracting a distribution due to the trend of change in the combustion parameter calculated by the trend calculation step (S504) from the variance of the combustion parameter calculated by the variance calculation step (S503), and a combustion stability judging step (S507) wherein combustion stability is judged based on the variance of the combustion parameter in the plurality of combustion cycles calculated by the variance calculating step (S503) and the trend of change of the combustion parameter calculated by the trend calculating step (S504).

Images