JP2830012B2 - Combustion state measurement method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION <Industrial Application Field> The present invention relates to a method for measuring a combustion state, and is applied to a combustion control map creation device, a power test device, and the like, and aims at automation and improvement of test accuracy. Things.

<Related Art> In recent years, many spark ignition internal combustion engines such as gasoline engines have been electronically controlled in response to demands for fuel saving and low pollution. Control forms include ignition timing control, knocking control, and fuel injection control.
ECU (Electronic C)
The mainstream is the one that performs intensive operations on an ontrol unit, and high reliability and low cost have been realized in conjunction with advances in semiconductor devices.

By the way, in the various controls described above, in many cases, a method is adopted in which information from the measuring means is processed by a combustion control map (hereinafter, a map) in a ROM or a RAM to determine a control amount. The map includes a two-dimensional map in which input values and output values correspond one-to-one, such as a temperature correction map,
There are three-dimensional maps corresponding to two types of parameters, such as a map for obtaining a fuel injection amount from an engine speed and an intake pressure.

These maps are generally obtained by performing experiments. For example, when an ignition timing map is created, the ignition timing is gradually advanced with the air-fuel ratio, the engine speed, the intake pressure, etc. fixed, and the MBT (Minimum
um Spark Advance for Best Torque) and find the optimal ignition timing that does not lead to knocking. Next, parameters such as the air-fuel ratio and the engine speed are changed, and the optimum ignition timing under the conditions is determined. Hereinafter, by repeating this experiment, maps corresponding to various driving states can be obtained.

On the other hand, of course, electronic fuel injection devices,
Installation of an EGR device or the like generally has a large effect on engine output and operability. Therefore, in the research and development, it is indispensable to accurately grasp the power generation state of the engine, and power test devices for measuring engine output, torque, and the like have been conventionally manufactured and used.

A conventional power test apparatus is configured to operate an engine under various conditions using an operation controller or the like, and to detect output and generated torque by a dynamometer or a torque meter. The test is performed in a wide range of operating ranges from idling to near the maximum revolution, and various tests are performed with various ignition timings and air-fuel ratios.

<Problems to be Solved by the Invention> Incidentally, knocking has always been a problem in tests using the above-described map creation device and power test device.

Knocking is a phenomenon in which an air-fuel mixture in an unburned part during combustion is self-ignited without waiting for the propagation of a flame due to a temperature rise due to adiabatic compression or the like, and is temporarily burned. Since the combustion at that time occurs rapidly, the pressure and the temperature rise rapidly in the combustion chamber and a strong shock wave is generated. As a result, the ignition plug, the piston and the like are overheated to cause erosion, and mechanical vibrations and distortions are caused in various parts of the engine.

However, in the above test, while it is necessary to operate under conditions that generate the maximum torque at the same engine speed, the ignition timing MB that can extract the maximum torque from the engine is required.
T is near the ignition timing that causes knocking. Therefore, in the conventional power test, a skilled measurer monitors the output of an acceleration detection type knock sensor (hereinafter referred to as G sensor) while changing operating conditions, and judges knocking by listening to the impact sound of a shock wave. Then, knocking is avoided by delaying the ignition timing or the like.

However, when such a method relying on visual or auditory sense is employed, the test cannot be automated, and a considerable amount of time is required, resulting in a delay in the test. In addition, while the measurement staff had to work in a power test room where the environment was poor, there were hygiene and safety problems, but there were many erroneous measurements due to the unskilledness of the measurement staff. In addition, due to knocking at each measurement, mechanical vibrations and distortions occur in each part of the test engine, and ignition plugs and pistons are easily overheated and melted. It took.

The present invention has been made in view of the above circumstances, and provides a method of performing a map creation test, a power test, and the like without generating knocking, thereby enabling automation and speeding up of a test, and solving the above-described disadvantages. Aim.

<Means for Solving the Problems> The chemical reaction of normal combustion in a spark ignition internal combustion engine is
The reaction is performed through a first stage peroxide reaction, a second stage cool flame reaction (or formaldehyde reaction), and a third stage hot flame reaction. The cause explosive reaction in this step is the third step, first, second stage the precursor reaction hydrocarbon in the fuel formaldehyde or OH, is decomposed to high energy free radicals such as HO ₂ It is. In the vicinity of the MBT, that is, the knocking occurrence condition, the first and second stage precursor reactions are progressing in the unburned region in the combustion chamber at the pressure and temperature just before the self-ignition, and there are many free radicals with high energy, It is in a more chemically activated state than usual. Therefore, when the flame surface reaches there, the third stage of the flame reaction occurs immediately without delay required for the precursor reaction, and the flame speed and, consequently, the heat generation rate increase.

On the other hand, in the state where the heat generation rate is high, the combustion reaction is rapidly progressing, so that the physical quantity related to combustion, that is, the physical quantity of combustion changes naturally as compared with the normal state. For example, the internal pressure of a combustion chamber (hereinafter, cylinder pressure), which is one of the physical quantities of combustion, increases with combustion, but the rate of increase increases as the heat generation rate increases. Therefore, it is possible to calculate the heat release rate or the change state of the physical quantity correlated with the heat release rate from the change state of the in-cylinder pressure, and if the change state is observed, it is determined whether the engine is in the vicinity of the knocking occurrence condition. You can know whether or not.

Based on the above findings, the present invention, in order to solve the above-mentioned problem, is a method for measuring the combustion state of a spark ignition internal combustion engine, which comprises detecting an output signal of an acceleration type knock sensor when knocking occurs and simultaneously Calculate the heat release rate and the physical quantity change status that is correlated with the heat release rate from the in-cylinder pressure that changes with the combustion in the combustion chamber of the internal combustion engine, and then knock based on the heat release rate and the physical quantity change status. The present invention proposes a combustion state measuring method characterized by detecting a state immediately before.

<Operation> The test engine was operated while appropriately changing the operating conditions.
Simultaneous detection of knocking by the G sensor and detection of cylinder pressure by a cylinder pressure sensor and the like are performed. Then, the detection signal of the latter sensor at the time of knocking occurrence is sent to the arithmetic unit, and the heat generation rate at that time or the change state of the physical quantity correlated therewith is calculated and stored. Thereafter, regardless of the G sensor, knocking avoidance operation is performed when the in-cylinder pressure is on the verge of knocking occurrence.

<Example> An example of the present invention will be specifically described with reference to the drawings.

FIG. 1 shows a schematic configuration of an ignition timing map creating apparatus to which a combustion state measuring method according to the present invention is applied. FIG. 2 is a graph showing the relationship between the crank angle and the heat generation rate, and FIG. 3 is a graph showing the relationship between the engine torque and the ignition timing at a predetermined air-fuel ratio. And
FIG. 4 shows a control flowchart in the present embodiment, and FIG. 5 shows an example of an ignition timing map created by the ignition timing map creating device of the embodiment.

As shown in FIG. 1, an ignition timing map creation device according to the present embodiment includes a test four-cycle four-cylinder spark ignition internal combustion engine (hereinafter, referred to as an engine) 1 and an operation controller 2 for controlling an operation state of the engine 1. And an arithmetic unit 3 for performing arithmetic processing on the data obtained by the operation to create a map.

In the combustion chamber 4 of each cylinder of the engine 1, an in-cylinder pressure sensor 6 as an in-cylinder pressure detecting means is attached in addition to the ignition plug 5.
A G sensor 24 is attached to the cylinder block 23. Both the in-cylinder pressure sensor 6 and the G sensor 24 incorporate a piezoelectric element, and convert the pressure in the cylinder and the vibration acceleration of the cylinder block 23 into electric charges and output them. In addition, a crank angle sensor 8 is provided adjacent to the flywheel 7,
The intake pipe 9 and the exhaust pipe 10 have an intake pressure sensor 11 for detecting intake pressure and an oxygen pressure sensor for detecting oxygen concentration in the exhaust, respectively.
The O ₂ sensor 12 or an exhaust analyzer (not shown) is attached. The crankshaft 13 is connected to a dynamometer 14 for applying a load to the engine 1 and measuring output, shaft torque, and the like.

The operation controller 2 drives the spark plug 5 via an ignition driver or the like, and also controls a fuel injection valve, a throttle valve, and the like (not shown) to control an injection amount, an intake amount, and the like.

The measured values measured by the sensors and the dynamometer 14 and the control values of the operation controller 2 are all input to the arithmetic unit 3. In the arithmetic unit 3, the signal from the in-cylinder pressure sensor 6 is input to the heat generation pattern arithmetic unit 15 together with the signal from the crank angle sensor 8, and after the processing is performed, RA
Input to M16. The signal from the crank angle sensor 8 is sent to the O ₂ sensor 12 via the engine speed calculation unit 17.
Alternatively, signals from an exhaust analyzer (not shown)
The signals are amplified by 18 and input to the RAM 16 respectively. The signals from other sensors and devices are stored in RAM16
To enter. The arithmetic unit 3 is connected to not only the above-mentioned sensors and devices but also detection means (not shown) for detecting atmospheric conditions such as atmospheric pressure and atmospheric temperature.
Enter 6

The arithmetic unit 3 further includes a data bank 19, an optimal ignition timing arithmetic unit 20, and a map creating unit 21. The data input from the heat generation pattern calculation unit 15 to the RAM 16 is input to the data bank 19 via the optimum ignition timing calculation unit 20. Further, various other signals (data) in the RAM 16 are also arranged and input to the data bank 19. The map creation unit 21 creates an ignition timing map using each data in the data bank 19.

Hereinafter, the operation of this embodiment will be described with reference to the flowchart of FIG.

Is activated on the operation controller 2, in the engine 1 starts to rotate, before the arithmetic unit 3 test, the minimum frequency of the engine rotational speed N _E at the start of the test (idling),
The load L is set to no load, the air-fuel ratio A / F is set to the maximum rich, and the ignition timing S is set to the maximum retard value.

Then, the actual engine speed N _E by the crank angle sensor 8, the actual load L by the throttle position sensor or the like, not the intake pressure sensor 11 or shown is O ₂ sensor
12 or actual air-fuel ratio using an exhaust analyzer not shown
The A / F is measured and compared with the preset value in the arithmetic unit. When there is a difference between the measured value and the set value, a correction amount for the set value is calculated so as to eliminate the difference, and is input to the operation controller 2.

When the set value and the measured value are equal, it is determined whether the region is a region where both the combustion determination using the in-cylinder pressure P and the combustion determination based on the output from the G sensor are performed. Since the criterion differs depending on the engine, it is set by a method such as inputting as a data sentence before executing the program according to this flowchart. The crank angle sensor 8 determines the crank angle θ, and the in-cylinder pressure sensor 6 determines the in-cylinder pressure P of each cylinder.
For further confirmation, the actual rotational speed N _E , the actual load L, and the actual air-fuel ratio A / F at that time are respectively detected. In addition, the output from the G sensor 24 is detected. Then, the heat generation pattern calculation unit 15 calculates the heat generation rate dQ / dθ by using the crank angle θ and the in-cylinder pressure P in the following procedure.

First, as shown below, the heat release amount dQ and the internal energy increment
An operation is performed using each operation expression for obtaining du and a state equation.

Here, G is the amount of combustion gas, A is the heat equivalent of work, R is the gas constant, Cv is the specific heat of constant volume, k is the ratio of the specific heat, and T is the absolute temperature.

From equations (1), (2) and (3) Therefore, the heat release rate (dQ / dθ) is as follows.

At this time, in the combustion stroke (top dead center to 50 ° after top dead center) Therefore, the above equation may be approximated as follows.

That is, the heat release rate can be approximated by the rate of change of the in-cylinder pressure.

When calculating the heat release rate as described above, it is desirable to cut high-frequency vibration components due to knocking or the like with a filter. That is, a high-frequency vibration component is always superimposed on the acupressure diagram, and by cutting this vibration component, the change state of the heat generation rate is simplified as shown in FIG. Therefore, in this embodiment,
Low-pass filter using FFT method or spline function method22
Is used.

Next, the heat generation pattern calculation unit 15 calculates the heat generation rate dQ / dθ from the obtained heat generation rate dQ / dθ (or the rate of change of the in-cylinder pressure dP / dθ... d
The time required for the fall of θ, that is, the transition time from the maximum value to the completion of combustion is calculated.

The transition time is not the actual time, but the crank angle θ _{100 at the} maximum value and the crank angle θ ₀ at the completion of combustion.
The difference | θ ₁₀₀ −θ ₀ | is used.

Next, the presence or absence of knock is determined based on the detected output from the G sensor 24. As a determination method, a conventionally known knock determination method using a G sensor output is used.

 Subsequently, the output torque T of the engine is detected.

Then, the aforementioned engine rotational N _E, load L, air-fuel ratio A / F, the ignition timing S, transition time | θ ₁₀₀ -θ ₀ |, knock presence by G sensor output, stores the torque T in RAM 16, i = 1
00 is determined.

If i is less than 100, i = i + 1 and the engine speed _NE , load L, air-fuel ratio A / F, ignition timing S, transition time |
θ ₁₀₀ −θ ₀ |, Knock due to G sensor output, torque T
Is calculated or detected and stored in the RAM 16.

When i = 100, 100 stored in the RAM 16
Delete data out of the allowable fluctuation range from the data of the times. Specifically, data of 100 times, rotation speed N _E , load L, air-fuel ratio A / F
The data in which one of the values is extremely different from the set value, the data of the cycle in which the value of the in-cylinder pressure P is abnormally low and a misfire is presumed, and the like are deleted.

It calculates the knock detection probability C ₁ using knock existence determination result by the G sensor 24 cycles which is the remaining valid data, is compared with the allowable probability value C ₀ which had been set in advance. The knock detection probability C ₁ by the G sensor 24 is an allowable probability value C ₀
If this is the case, based on the statistically processed value of the data of the transition time | θ ₁₀₀ −θ ₀ | of the cycle regarded as the valid data, the absolute value for the determination time of the transition time | θ ₁₀₀ −θ ₀ | Correct the set value. Specifically, when the valid data is compared with an absolute set value for a determination reference of a transition time | θ ₁₀₀ −θ ₀ | described later, the probability that the transition time becomes smaller than the absolute set value is an allowable value A described later. _The absolute setting value is corrected so that it becomes _zero . In this case, since the current set ignition timing S is ahead of the optimum ignition timing, each operation parameter is stored in the RAM 16 as MAP data with a delay of dX °. Next, G
If the knock detection probability by the sensor is equal to or less than the allowable probability value C, the above-described transition time | θ ₁₀₀ −θ ₀ | is compared with an absolute set value described later to determine whether or not the knocking state is present.

That is, as shown in FIG. 2, the heat release rate (shown by a solid line) in the state immediately before knocking changes greatly in the way of falling compared to the state before that (shown by a broken line), and the transition time | θ
₁₀₀ −θ ₀ | is shorter. Therefore, if the transition time | θ ₁₀₀ −θ ₀ | immediately before knocking is set as an absolute setting value,
It is possible to determine whether or not the vehicle is in a knocking state.

The determination method, the transition time | θ ₁₀₀ -θ ₀ | calculates the probability A ₁ becomes smaller than the absolute setting value is compared with this acceptable value A _0. Here, A ₀ is a reference value for determining the knock margin, and A ₁
If ≤A ₀ , it is determined that knocking is not sufficiently performed, and A ₁
If> _A0 , it is determined that there is no margin for knocking.

If A ₁ > A ₀ , it is determined that the margin for knock is beyond the limit, so the current set ignition timing is delayed by dX °, and each operation parameter is stored in the RAM 16 as MAP data.

If A ₁ ≦ A ₀ , the MBT is determined.

That is, the average value T _n-1 of the torque T at the time of the previous set ignition timing data capture, compares the average value T _n of the torque T at the time of this setting ignition timing data capture, comparison result T _n ≧
The previous ignition timing when T _n <T _n-1 changes from T _n-1 to MB
T.

If T _n ≧ T _n-1 , the current transition time | θ ₁₀₀ −
The deviation between the average value B ₁ of θ ₀ | and the reference value B ₀ is determined. here,
The reference value B ₀ is used to determine whether or not the ignition timing S can be further advanced. The knock margin due to the probability A ₁ that the transition time | θ ₁₀₀ −θ ₀ | becomes smaller than the absolute set value is determined. This is to determine the advance angle limit with higher accuracy than the determination. Then, B ₁ and variance b ₁ of B ₀ is allowable value b ₀ or more than dX ° advances the ignition timing S if hidden again performs 100 times of data acquisition described above.

On the other hand, the deviation b ₁ of B ₁ and B ₀ is equal tolerance b ₀ or less, it is determined that the operating parameter is an optimal value at that time, and stores the respective values in the RAM as the MAP data. That is, when the advance angle limit comes before the MBT, the advance angle is interrupted at that time.

Also, if a T _n <T _n-1 is determined in the aforementioned maximum torque, determines that had been noted advancing past the MBT,
The ignition timing S is retarded by dX °, and each operation parameter is stored in the RAM 16 as MAP data.

Here, i represents the number of samplings for statistically processing the data. In this embodiment, 100 is used, but of course, another appropriate value may be used.

Also, in a region where the combustion determination based on the output from the G sensor 24 is not performed, the crank angle θ, the in-cylinder pressure P, the rotational speed N _E , the load L, and the air-fuel ratio A / F are detected by setting i = 1. The heat generation pattern calculation unit 15 calculates the heat generation rate dQ / dθ, and further calculates the transition time | θ ₁₀₀ −θ ₁ |. Subsequently, the engine output torque T is detected, and the above detected data, calculation data, and ignition timing data S are stored in the RAM 16, and are repeated until i = 100. Here, i may be another appropriate value as described above. When i = 100, the data outside the allowable fluctuation range is deleted in the same manner as described above, and the transition time of the remaining valid data | θ ₁₀₀ −θ ₀ |
And the transition time | θ ₁₀₀ −θ ₀ | are compared to determine whether or not the engine is in a knocking state, and MAP data is obtained.

As described above, when the optimum ignition timing S when the engine speed _NE is the minimum engine speed (idling), the load L is no load, and the air-fuel ratio A / F is the maximum rich is obtained, the rotational speed N _E to the minimum rotational speed (idling), while the load L was fixed to no load, to lean the air-fuel ratio a / F than the maximum rich by dY, optimum ignition timing in each of the a / F value S Ask for.

Then, when the optimum ignition timing S for each A / F value from the highest rich to the lean limit value is obtained, the value of the load L is changed and data is fetched. That is, the minimum rotation speed of the engine rotational speed N _E (idling)
Next, the load L is fixed to a value increased by dL from the no load, and the optimum ignition time S for all A / F values is obtained.

After that, the load L is increased by dL to the maximum load,
For each load value, the A / F value is changed from the highest rich to the lean limit value, and the optimum ignition timing S is obtained. Then, the maximum rotational speed of the engine rotational speed N _E,
When the optimal ignition timing S for all combinations of the load L and the A / F value is obtained, the engine speed is changed and the data is acquired. That, Yuki by increasing the engine speed from the minimum rotational speed by dN _E until the maximum rotational speed, for each of the rotational speed by changing the load L and A / F value, determining the optimum ignition timing with respect to all setting values.

As described above, the optimum ignition timing S for all combinations from the initial value to the limit value for the engine speed _NE , the load L, and the air-fuel ratio A / F is obtained. It is input to bank 19.

Next, the data in the data bank 19 is compiled into a measurement data group shown in FIG. Finally, the map generation unit 21 extracts and arranges the optimum ignition timing data, and generates an ignition timing map as shown in FIG. 5 from the optimum ignition timing data.

Although the description of the embodiment has been finished above, the present invention is not limited to this embodiment. For example, in determining whether or not knocking conditions are near, a heat generation rate of 50% with respect to the maximum value of the heat generation rate is determined. it may be other regions such as in the range of up to crank angle theta ₁₀ showing a 10% heat generation rate from the crank angle theta ₅₀ showing the incidence. Further, the present invention is not used only for creating an ignition timing and an air-fuel ratio map, but is based on measurement data groups such as an EGR rate and a supercharging pressure shown in FIGS. 7 and 8, respectively.
You may apply to the combustion control map preparation apparatus which produces each map. Further, the present invention may be applied to a power test device or the like.

<Effect of the Invention> According to the combustion state measurement method of the present invention, the state immediately before knocking is detected from the change in the in-cylinder pressure. Prevention of knocking during power test can be automated. In addition, since strong knocking does not occur during these tests, the life of the test engine is improved.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a schematic configuration diagram showing an ignition timing map creating apparatus to which a combustion state measuring method of the present invention is applied. Also, the second
FIG. 3 and FIG. 3 are graphs respectively showing the relationship between the crank angle and the heat release rate, and the relationship between the engine shaft torque and the ignition timing at a predetermined air-fuel ratio. FIG. 4 is a control flowchart of the ignition timing map creating device, and FIG. 5 is an ignition timing map created by the ignition timing map creating device. 6 to 8 show data groups for creating an ignition timing and an air-fuel ratio map, an ignition timing and an EGR rate map, and an ignition timing and a supercharging pressure map, respectively. In the figure, 1 is an engine, 2 is an operation controller, 3 is an arithmetic unit,
6-cylinder pressure sensor, a crank angle sensor 8, the intake pressure sensor 11, the O ₂ sensor 12, 14 is dynamometer, 15 heat generation pattern calculation unit, 16 RAM, 17 is an engine rotational speed calculation unit, 19 Is a data bank, 20 is an optimal ignition timing calculation unit, 21 is a map creation unit, 24 is a G sensor, 25
Is an operation parameter setting and correction unit.

──────────────────────────────────────────────────続 き Continued on the front page (72) Hiromitsu Ando 5-33-8 Shiba, Minato-ku, Tokyo Inside Mitsubishi Motors Corporation (72) Daisuke Mibayashi 5-33-8 Shiba, Minato-ku, Tokyo No. Mitsubishi Motors Corporation (56) References JP-A-64-32070 (JP, A) JP-A-63-212768 (JP, A) JP-A-61-234275 (JP, A) JP-A-63 -248972 (JP, A) JP-A-62-130330 (JP, A) JP-A-1-301946 (JP, A) JP-A-59-136543 (JP, A) (58) Fields investigated (Int. . ⁶ , DB name) F02D 45/00

Claims (2)

(57) [Claims] 1. A method for measuring a combustion state of a spark ignition internal combustion engine, comprising: detecting an output signal of an acceleration type knock sensor at the time of occurrence of knocking; A combustion state measuring method, wherein a change state of a heat release rate is calculated from an internal pressure, and thereafter, a state immediately before knocking is detected based on the change state of the heat release rate. 2. A method for measuring a combustion state of a spark ignition internal combustion engine, comprising: detecting an output signal of an acceleration type knock sensor at the time of knocking; A combustion state measuring method, wherein a physical quantity having a correlation with a heat release rate is calculated from an internal pressure, and thereafter, a state immediately before knocking is detected based on the change state.