JPH0629822B2 - Cylinder pressure detection type engine control device - Google Patents

Description

Description: FIELD OF THE INVENTION The present invention relates to engine control in which the pressure in a cylinder of an internal combustion engine such as a gasoline engine or a diesel engine is constantly detected, and the fuel supply amount or the like is controlled based on the pressure. The present invention relates to an apparatus, and more particularly to an engine control apparatus using a novel cylinder pressure sensor for that purpose.

[Background of the Invention]

A cylinder in an internal combustion engine that is in operation (hereinafter simply referred to as a cylinder)
When the internal pressure is constantly detected and the change state is known, information such as the air-fuel ratio, ignition timing (or fuel injection timing), and EGR (exhaust gas recirculation) amount can be accurately and quickly obtained. Therefore, by performing feedback control in the direction in which the deviation from the control target is reduced based on these pieces of information, it is possible to obtain a control system that operates accurately and with good responsiveness.

Therefore, various control systems based on such in-cylinder pressure detection have been proposed.

For example, in JP-A-57-153966, a combustion engine is obtained based on the outputs of a cylinder pressure sensor and a crank angle sensor,
A configuration for controlling the EGR amount so as to eliminate the deviation from the target value is disclosed.

Further, Japanese Patent Application Laid-Open No. 57-163128 discloses a configuration in which the air-fuel ratio is calculated and obtained from the outputs of the in-cylinder pressure sensor and the crank angle sensor, and the fuel amount is feedback-controlled so that there is no deviation from the target air-fuel ratio. is doing.

By the way, a key component in the construction of such a system is an in-cylinder pressure sensor, and a conventional strain gauge sensor is a strain gauge type sensor.
It is based on the piezoelectric type.

However, internal combustion engines, especially spark ignition engines such as gasoline engines, require the supply of high voltage energy in the system to discharge at the spark plug, and
The above-described conventional in-cylinder pressure sensor has a problem that radio noise is likely to be superimposed on the pressure signal. Furthermore, the strain gauge type has drawbacks such as low output and short life, and the piezoelectric type has insufficient accuracy.
They have drawbacks such as static verification being difficult, large zero point movement, necessity of moisture-proof storage, and easy picking up of mechanical disturbance vibrations. There is a drawback that the advantage of the detection method cannot be fully utilized.

[Object of the Invention]

The object of the present invention is to eliminate the disadvantages of the conventional in-cylinder pressure detecting means described above, and to provide a new in-cylinder with high accuracy and high reliability that can withstand mounting in an actual machine. It is to provide a system configuration for presenting a pressure detection means and suitably performing feedback control of various amounts that influence combustion by using the pressure detection means.

[Outline of Invention]

In order to achieve this object, in the present invention, in a sensor means of a method of detecting the bending amount of a flexible member which is deformed by the pressure in a cylinder by an optical fiber, the flexible member forming this sensor means is a thin plate circle. The optical fiber member is formed of a plate-shaped member, and the optical fiber member is formed of fiber element wires that are concentrically arranged in a triple annular shape and are arranged in the middle of the fiber elements wires arranged in the triple annular shape. The existing fiber element wire is used for illumination, and the fiber element wire on the inner circumference side and the fiber element wire on the outer circumference side are used for collecting reflected light, thereby eliminating the disadvantages of the conventional in-cylinder pressure sensor. There is a point.

Further, in the case of a multi-cylinder engine, the signal processing configuration is suitable for detecting the in-cylinder pressure of each cylinder by the sensor using the optical fiber and processing the signal.

Furthermore, the overall configuration is such that feedback control of various amounts that influence combustion is performed using the processed signal.

Example of Invention

 Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

FIG. 1 is a principle view of an in-cylinder pressure sensor used in the present invention. First, FIG. 1 (a) is a cross-sectional view thereof, which is for illumination composed of a plurality of fiber element wires arranged annularly in the circumferential direction. Fiber 1, inner collecting fiber 2 consisting of a plurality of fiber strands arranged along the inner circumference thereof, outer collecting fiber 3 similarly arranged on the outer circumference, and facing at right angles to the axis of these fiber bundles It is shown that this pressure sensor is constituted by the diaphragm 4 whose circumferential end face is supported by the supporting member 5 over the entire circumference. In such a structure, when the pressure P is atmospheric pressure, the diaphragm 4 does not bend but maintains a flat state as shown in FIG. 1 (c).
Therefore, the light i ₀ guided from the illumination fiber 1, the inner collection fiber 2, evenly respectively outside collection fiber 3 i _1, i ₂ of the optical Charles Shirubebi reflected. Therefore in this case i ₁ = i ₂ . Next, when P> 0 and the pressure is larger than the atmospheric pressure as shown in FIG. 1 (b), the diaphragm 4 bends downward in the drawing. Therefore, in this case, less light i ₁ is reflected and guided from the diaphragm 4 to the inner collecting fiber 2, and conversely, light i ₂ is guided to the outer collecting fiber 3.
Will increase. On the other hand, on the contrary, when the pressure is smaller than the atmospheric pressure (P <0) as shown in FIG. 1 (d), the diaphragm 4 bends upward in the figure. Therefore, in this case
Contrary to the case of (b), i ₁ > i ₂ . That is, i ₁ and i ₂ show an inversely proportional relationship depending on the magnitude of the pressure P.

FIG. 2 shows the relationship between the ratio i ₂ / i ₁ of these two and the pressure, and it can be seen that there is an almost linear proportional relationship. When a diaphragm having a diameter of about 2 mm is used, the pressure range to be measured can be set arbitrarily by arbitrarily selecting the thickness of the diaphragm 4. Also, the diaphragm 4
This range can also be changed by selecting the material of.

FIG. 3 shows an embodiment of a cylinder pressure sensor embodied by applying this principle. A thin metal diaphragm 4, which is manufactured by controlling the thickness, flatness, etc. with high accuracy, is fixed to one end surface of the sensor housing 6. Further, the end face of the fiber bundle 7 is located at a point where the distance from the inner end face of the metal diaphragm 4 is controlled with high accuracy. The fiber bundle 7 is composed of an illumination fiber 1, an inner collecting fiber 2, an outer collecting fiber 3 and a filling material 8 which fills the boundary between them (the filling material can withstand a high temperature of several hundreds of degrees, and Those with moderate elasticity are desirable,
For example, silicon-based or fluorine-based plastic, solder glass, ceramic containing metal powder, or the like may be used. ) And therefore, the optical fiber bundle 7 is elastically held by the heat-resistant packing material 8,
During operation, the temperature may rise to 100 ° C or higher, and
Even if the optical fiber bundle 7 is mounted on an engine that often vibrates violently, there is no possibility that an extra force is applied to the optical fiber bundle 7 or the mounting state changes, and the pressure can always be reliably detected.

The optical fiber bundle 8 is adhered to the adhesive portion 9 with a strong adhesive material that can withstand high temperatures. Further, the diaphragm chamber 10 which is a space formed by the diaphragm 4 and the end face of the fiber bundle 7 is configured to be electrically connected to the atmosphere by the pressure introducing hole 11. In this case, it functions as a relative pressure type in-cylinder pressure sensor that detects the relative pressure. On the other hand, when the air in the diaphragm chamber 10 is sucked through the pressure introducing hole 11 and the pressure introducing hole 11 is closed after the inside of the diaphragm chamber 10 is evacuated, the absolute pressure cylinder pressure for detecting the absolute pressure is obtained. Functions as a sensor. Therefore, both of them can be arbitrarily selected depending on the application. Fiber bundle 7
The other end face of the sensor housing 6 is set at a position deeper than the other end face 12 of the sensor housing 6 and is set at a right angle to the axis 0-0 so that the flatness of the cut surface is controlled with high accuracy. Further, the inner peripheral surface 13 of the sensor housing 6 is configured to maintain a highly precise parallelism with the shaft 0-0 and an inner diameter precision in order to be connected to the connector with high precision. The whole sensor is represented by 14.

FIG. 4 shows an embodiment in which the in-cylinder pressure sensor 14 shown in FIG. 3 is connected by a fiber connector 15.

The position configuration of the fiber bundle 20 in the fiber connector 15 is the same as that of the in-cylinder pressure sensor 14 shown in FIG. 3, but the filling material 21 may be polyethylene or the like which is highly deformable. However, since high-temperature heat on the combustion chamber side may still be transmitted in the vicinity of this, as in the case of FIG. 3, the fiber cable 22 of a predetermined length is made of heat-resistant packing material.
It is also recommended that the above-mentioned structure is provided, and then a fiber connector is further provided to form a filling material having high deformability in the subsequent fiber cable.

As described above, by using the fiber connector 15, first, only the in-cylinder pressure sensor 14 is attached to, for example, the cylinder of the engine, and then the fiber connector 15 is connected to the sensor housing 6 to the sensor housing 6. Therefore, there is no possibility that the fiber bundle 20 on the connector side will be subjected to an extra force when the sensor is attached and the fiber will be broken, and therefore high detection accuracy can be easily obtained.

By the way, in the embodiment shown in FIG. 4, in order to secure the coaxiality of the fiber bundle 7 on the sensor side and the fiber bundle 20 on the connector side with high accuracy, the circumferential surfaces 23 and 24 are engaged with each other to mate them. ing. And, in this case, both are fixed by the screw portion 25. However, in this case, it is difficult to completely match the respective fiber strands (see FIG. 1 (a)) in the circumferential direction, and transmission loss of fiber light occurs at this connection portion. Therefore, in order to suppress this loss as much as possible, a limiter is provided in the circumferential direction and the sensor,
It is necessary to connect both of the connectors only by the movement in the thrust direction to prevent the displacement in the circumferential direction. Since the structure thereof can be sufficiently achieved by the ordinary technique, it is omitted in this specification.

By the way, the above cylinder pressure sensor has a diaphragm diameter of φ2
It is possible to select an arbitrary size of about φ15 mm and freely select the length direction. Therefore, the third
A sensor with the configuration shown in the figure is used in the combustion chamber of the engine (in the cylinder).
It is also possible to adopt a configuration in which it is directly arranged in. As is well known, quartz is generally used as a fiber material, and multi-component glass is also frequently used. For quartz, the softening temperature is 17
It is extremely high at 00 ℃, and even if it is installed in the immediate vicinity of the combustion chamber, it is not affected. Also, because quartz is not an electrical conductor,
Since there are no problems such as external induction and radio wave interference, there is no fear of receiving electrical noise due to ignition, so it has the characteristics that it can withstand even under adverse conditions near the combustion chamber.

However, if it is not desirable to newly provide a hole for mounting the in-cylinder pressure sensor in the engine, as shown in FIG. 5, the sensor may be attached to a part of the ignition plug. As described above, since the in-cylinder pressure sensor according to the present invention can be constructed with a very small diameter of about φ2 mm, it can be mounted on the conventional spark plug without significantly changing the structure thereof.

In FIG. 5, 30 is a central electrode and 31 is an external electrode, both of which are electrically insulated by a porcelain insulator 32. The porcelain insulator 32 is fixed by a metal body 33, and a screw portion 34 for attaching to the engine combustion chamber is provided at the lower end portion thereof. Further, the center electrode 30 can be connected to a high voltage cord by a connection terminal 35.

Although the above-mentioned structure is the structure itself of a normal spark plug, in this spark plug, the pressure hole 36 penetrates the metal body 33 as shown by the fracture surface of FIG. A rectangular groove is formed as described above, and the opening end of the groove is configured to connect the combustion chamber (in the cylinder) and the pressure hole 36. Further, an in-cylinder pressure sensor 14 is attached to the other open end of the pressure hole 36.

In the above configuration, the diameter of the pressure hole 36 may be about φ1 mm, but it is desirable to increase it to about φ2 to φ3 mm or more when there is a possibility that problems such as clogging due to Devogit and deterioration of pressure response sensitivity may occur. . Further, the diaphragm surface 38, which is the pressure-sensitive portion of the in-cylinder pressure sensor 14, is more advantageous from the viewpoint of responsiveness as it is closer to the combustion chamber (in-cylinder), so that the pressure hole 36 is shortened as long as the structure allows. It is desirable to install in When the engine is equipped with the spark plug, as shown in FIG.
It is desirable that the fiber connector 15 shown in the figure is mounted without being connected, and the connector is connected after the mounting is completed.

Next, using the cylinder pressure sensor described above, a fiber for detecting the cylinder pressure of each cylinder of a multi-cylinder engine,
And the method of coupling the optical system will be described. A four-cycle four-cylinder engine will be described in detail as an example.

Various methods are conceivable as a method of coupling the fiber and the optical system, but only typical ones will be shown here.

First, the method of FIG. 6 will be explained.
Reference numerals 40, 41, 42 and 43 are cylinders, and cylinder pressure sensors 44, 45, 46 and 47 are provided respectively. Light from the light source 48 is equally branched by the optical branching device 49 and introduced into these in-cylinder pressure sensors. This is the illumination light i ₀ introduced by the illumination fiber 1 described in FIG. Also,
From the in-cylinder pressure sensor, the light from the inner collection fiber 2 i ₁
And the light i ₂ from the outer collection fiber 3 is output. i ₁
Light is guided to the optical coupler 50 in parallel for four cylinders, and the combined light is photoelectrically converted by the light receiver 51. Similarly, the light of i ₂ is also photoelectrically converted by the light receiver 53 via the optical coupler 52.

In the case of this method, as shown in FIG. 7, the outputs of i ₁ and i ₂ have a waveform as if the pressure of each cylinder were smoothed. That is, the pressure waveform 54 of the first cylinder and the pressure waveform 5 of the second cylinder
5, the pressure waveform 56 of the third cylinder, and the pressure waveform 57 of the fourth cylinder are faithfully detected by the respective cylinder pressure sensors 44, 45, 46, 47, and the optical couplers 50, 51 are respectively provided with i ₁₁ , i ₁₂ , The light of i ₁₃ , i ₁₄ and i ₂₁ , i ₂₂ , i ₂₃ , i ₂₄ is introduced, but in the optical coupler, these lights are combined and output in the form of addition, so i ₁ is 58, i ₂ Is a waveform such as 59. Therefore, the output waveform 60 of i ₂ / i ₁ has a waveform oscillating above the zero point level. Therefore, the average value of the waveform of i ₂ / i ₁ has a characteristic like an average effective pressure for four cylinders.

Next, the method of FIG. 8 will be described.

The optical system for introducing the illumination light i ₀ is the same as in the case of FIG. 6, and the light from the light source 48 is equally branched by the optical branching device 49 and guided to the in-cylinder pressure sensors, respectively.

In this method, however, the method for coupling the light collecting and collecting optical elements i ₁ and i _{2 with} the optical system is different from that shown in FIG. That is, the inner collecting and collecting i ₁₁ and the outer collecting i ₂₁ of the first cylinder are separately received by the light receivers 61a and 61b, and similarly, the light receiving parts of the second cylinder are similarly received.
Light receivers 63a and 63b for the third cylinders 62a and 62b, and light receivers 64a and 64b for the fourth cylinder are separately received. Therefore, in the case of this method, the information obtained from the in-cylinder pressure sensor is the same as the waveforms 54, 55, 56 and 57 in FIG. 7 as the pressure waveform of each cylinder.

Finally, the method shown in FIG. 9 will be described.

In the case of this method, four types of light sources having different wavelengths are provided as 70, 71, 72 and 73, respectively, and the light sources are guided to the respective in-cylinder pressure sensors via introducing fibers. From each cylinder pressure sensor, the wavelengths λ ₁ , λ ₂ , λ ₃ , λ
₄ , i ₁₁ and i ₂₁ at wavelength λ ₁ from the first cylinder, i ₁₂ and i ₂₂ at wavelength λ ₂ from the second cylinder, and wavelength λ ₃ from the third cylinder. in i ₁₃ and i ₂₃ is further from the fourth cylinder are output i ₁₄ and i ₂₄ at a wavelength of lambda _4, optical group i ₁ to the optical coupler 73, the optical group i ₂ optical coupler 74. By these couplers 73 and 74, i ₁ and i ₂ of λ ₁ to λ ₄ become one optical path respectively, and are guided to the optical demultiplexer 77 and the optical demultiplexer 78 via the fiber cables 75 and 76, respectively. . Then, the two optical demultiplexers 77 and 78 branch the optical paths that have been one optical path for each wavelength component. That is, it branches into i ₁ and i ₂ for each cylinder. This branched light is 61 in FIG.
Light is received by each of the light receivers a to 64a and 61b to 64b. Therefore, the pressure waveform information finally obtained from the light receiver is the same as the method shown in FIG. However, in the case of this method, the number of fiber cables can be greatly reduced between the cables of 75 and 76 as compared with the method of FIG. 8, so that a simplified configuration can be obtained.

Next, as the light source used in FIGS. 6, 8 and 9,
Although any semiconductor laser, light emitting diode, or the like may be used, it is more preferable to use a semiconductor laser that can obtain coherent light.

As is apparent from the above description, the in-cylinder pressure information obtained from the photodetector is (1) one average and smooth signal for four cylinders (see FIG.
(2) The waveform for each cylinder is one of the four signals (corresponding to 54, 55, 56, 57 in Fig. 7).

Therefore, the following describes the signal processing methods (1) and (2). First, when the inner light collecting and collecting i ₁ and the outer collecting and collecting i ₂ from the in-cylinder pressure sensor are received by the light receivers 51 and 52 as shown in FIG. 10 (a), the curves A and B in FIG. 10 (b) are received. Outputs of voltage V ₁ proportional to the light i ₁ and voltage V ₂ proportional to the light i ₂ are obtained as shown in. Therefore, this signal is input to the A / D converter 80, and both signals are alternately A / D converted for each predetermined angle Δθc of the crank angle signal θc, and the converted signal is converted until the next conversion. (Between 2Δθc) is sequentially held by the CPU
Entered in 81. In the CPU 81, this alternating input
The calculation of V _x = V ₂ / V ₁ is performed using the digital signals of V ₁ and V ₂ . That is, a voltage signal of V ₂ / V ₁ equivalent to i ₂ / i ₁ shown in FIG. 2 is obtained.

Next, the signal processing method (2) described above will be described.

In this case, it becomes as shown in FIG. First, Fig. 11
As shown in (a), the inner light collecting and collecting groups i _{11 to} i ₁₄ are converted into voltage waveforms V _{11 to} V ₁₄ by the inner light collecting and collecting light receiving element groups 61a to 64a, respectively, while the outer light collecting and collecting groups i _{21 to} i ₂₄ are converted. Are converted into voltage waveforms of V _{21 to} V ₂₄ by the outer light collecting and collecting groups 61b to 64b, respectively. Then, these voltage waveform groups are input to the first A / D converter 82 and the second A / D converter 83, respectively.
Next, as shown in FIG. 11 (b), for example, in the first A / D converter 82, V ₁₁ → V ₁₂ for each predetermined crank angle Δθc.
→ V ₁₃ → V ₁₄ → V ₁₁ ……… In order, convert to digital signal. Therefore, for the V ₁₁ signal, a ₁ → a ₂ → a ₃ → ……
…, V ₁₂ signal is b ₁ → b ₂ → b ₃ …, V ₁₃ signal is c ₁ →
c ₂ → c ₃ → ……, and V ₁₄ signal is d ₁ → d ₂ → d ₃ → ……. The digitized signals are input sequentially _{CPU84, V x1 = V 21/} 11, V x2 = V 22/12, V x3 = V 23/13, V x4 =
Performs operations in the order of V _24/14, the pressure signals of the cylinders
Obtain the waveforms of V _x1 , V _x2 , V _x3 , and V _x4 .

As is clear from the description of the data fetching in FIGS. 10 and 11, the smaller the data fetching interval Δθc, the higher the detection accuracy. Therefore, this Δθc needs to be reduced as long as the A / D conversion speed allows. is there. Judging from the capabilities of currently available A / D converters, it is thought that a crank angle of about 2 ° is possible. However, since this limit changes depending on the engine speed, Δθc _{1000 to 1999} = 1 °, Δθc for every predetermined number of revolutions, for example, every 1000 rpm.
_{2000 to 2999} = 2 °, Δθc _{3000 to 3999} = 4 °, Δθc
It is also effective to change the intake angle stepwise like _{4000 to 4999} = 8 °.

By the above signal processing, it is possible to detect (a) the average in-cylinder pressure signal of four cylinders, or (b) the in-cylinder pressure signal of each cylinder.

Next, an embodiment in which engine control is performed using this in-cylinder pressure sensor will be described.

First, FIG. 12 shows a system diagram of such an in-cylinder pressure detection type engine control device.

In FIG. 12, a signal from the in-cylinder pressure sensor 14 attached to the spark plug 200 is input to the microcomputer 202 via a signal processing circuit 201 such as a photoelectric circuit. Further, the signal of the crank angle sensor 203 for detecting the rotation speed and the crank angle and the signal of the air-fuel ratio sensor 204 are also input. Then, based on these signals, the optimum ignition timing, fuel injection timing, and correction amount of the injection amount are set to the ignition coil 20.
5, output to the injection valve 206, whereby the engine is controlled to an optimum operating state.

Here, the principle of the method of detecting the air-fuel ratio and the ignition timing by the output of the in-cylinder pressure sensor will be described.

In FIG. 13, (a) shows the change of the in-cylinder pressure P with respect to the crank angle θ, and the change rate dP / dθ of the in-cylinder pressure P with respect to the crank angle θ is shown in the same figure ( b). Needless to say, the in-cylinder pressure P is obtained from the output of the in-cylinder pressure sensor 14.

Therefore, when the point Q at which dP / dθ in FIG. 13 (b) shows the maximum value is obtained in FIG. 13 (a), this point Q is the ignition point. Therefore, it can be seen that from the beginning of the compression process to the end of the explosion stroke, before the point Q, the fuel is not ignited and the pressure P is increased only by the compression.

First, in order to obtain the air-fuel ratio, in the characteristic shown in FIG. 13 (a), specific positions θ _a and θ _b are set at predetermined crank angle positions before and after the ignition point Q, respectively. Then, these positions theta _a, respectively in theta _b pressure P _a, detects P _b,
Determining the air-fuel ratio A / F by these ratios P _b / P _a. That is, the in-cylinder pressure P before the ignition point Q is due to compression and is therefore independent of the air-fuel ratio, but on the other hand, the ignition point Q
Since the in-cylinder pressure P after is changed by the air-fuel ratio, these ratios represent the air-fuel ratio. Actually, there is almost a perfect proportional relationship between them as shown in Fig. 14, and if the proportional constant is calculated in advance, it can be accurately calculated from the data P _a and P _{b with} almost no delay. Air-fuel ratio A / F
You can know. The characteristic indicated by the broken line in FIG. 13 (a) is intended when it becomes that different air-fuel ratio A / F in the case of the solid line, the cylinder pressure at this time is theta _b is theta _b '
It has shown that it has become.

Therefore, if the air-fuel ratio is obtained from the data P _a and P _b and the fuel supply amount by the injection valve 206 is controlled so that it converges to a predetermined value, the air-fuel ratio can be controlled with good responsiveness.

Next, the ignition control based on the ignition timing will be described.

With respect to the crank angle position θ ₁ where the ignition point Q appears, it is known in advance what value the crank angle position θ ₁ will have when the maximum engine efficiency is obtained. That is, this angular position θ
₁ has an optimal value. Therefore, if the ignition crank angle position θ ₁ can be detected by the in-cylinder pressure sensor 14, the ignition signal Tign for the ignition coil 205 is corrected so that the ignition crank angle position θ ₁ converges to an optimum value, whereby optimum ignition control can be obtained.

Therefore, a method of obtaining the data θ ₁ will be described with reference to FIG.

The data θ ₁ is represented by the delay time Tθ from TDC.
Therefore, if TDC is known from the signal of the crank angle sensor 203 and the time from the time when TDC is reached until the point Q appears on the in-cylinder pressure characteristic is measured, the data T _θ can be obtained and the data θ ₁ can be obtained. . Therefore, Fig. 15
For the dP / dθ characteristics shown in (a), the TD as shown in (b) of the same figure
At the time of rising at C and the maximum value of dP / dθ, that is, the 13th
A gate pulse g which falls at the time when the ignition point Q in FIG. 15 (a) appears is created, and by this, an appropriate clock pulse CK as shown in FIG. 15 (c) is gated to produce the same as shown in FIG. 15 (d). You just have to get the pulses and count them. In order to find the time when the characteristic dP / dθ shows the maximum value,
The signal that changes as shown in FIG. 15 (a) is differentiated, that is, the signal shown in FIG.
It suffices to detect the time at which it first shows zero after TDC after differentiating.

FIG. 16 shows an example of processing required for ignition timing correction control.

First, the crank angle θ and the in-cylinder pressure P are read, (dP / dθ) _max is detected from this, the data θ _{1 at} this time, that is, the time T _θ of the deviation from TDC is detected, and the correction amount T corresponding _{thereto is} detected. (T _θ ) is added to the ignition timing signal Tign to obtain a corrected signal Tign.

Here, a proportional constant in the A / F characteristic with respect to P _b / P _a shown in FIG. 14, that is, a method of calibrating this characteristic will be described.

In this embodiment, an air-fuel ratio sensor 204 is provided in the system as shown in FIG. Therefore, calibration is performed using the air-fuel ratio sensor 204. First, the sensor 204 has the original air-fuel ratio sensor characteristic as shown in FIG. 17, that is, the A / F is linearly detected. If it is possible, the A / F data obtained from the air-fuel ratio sensor 204 when the engine operating condition is in the steady state is used.
The value of A / F based on P _d / P _a , which is the data of, may be corrected.

Further, as shown in FIG. 18 (a), the air-fuel ratio sensor 204 has λ =
When the output is inverted at 1.0, which is a so-called O ₂ sensor, etc., the output of the sensor changes to level 1 and level 0 every time it crosses the slice level V _s, as shown in FIG. Is obtained, the number N of level changes at this time is measured during a predetermined time width Tj, and the number N
When λ exceeds a predetermined value M ₂ , it is assumed that λ = 1.0, and the A / F can be corrected based on this. Therefore, the processing at this time is, for example, as shown in FIG. 19. First, the change amount dQa / dt of the accelerator operation angle Qa exceeds a predetermined value M ₁ to determine whether the operating state of the engine is in a steady state. Whether or not the number of times is N, which is N
When expressed by (Tj), exceeds a predetermined value M ₂ , λ = 1.0
And the value of P _b / P _a when λ = 1.0 stored at this time, which is expressed as (P _b / P _a ) λ = 1.0, and P
_b / P _a is compared, and if there is a difference, correction is performed by adding _a correction amount (P _b / P _a ) ′ that makes them equal.

Next, another embodiment for obtaining the air-fuel ratio from the in-cylinder pressure characteristic shown in FIG. 13 (a) will be described with reference to FIG.

From the compression stroke to the explosion stroke, the in-cylinder pressure P is taken in for each minute crank angle Δθ, and the integrated value ΣP /
Calculate Σ (Δθ). At this time, if the integral value until reaching the ignition point Q is A ₁ and the integral value after that is A ₂ , the air-fuel ratio A / F is proportional to A ₂ / A ₁ . In addition, these
As is clear from FIG. 20, A ₁ and A ₂ represent the area of the portion surrounded by the pressure characteristics before and after the ignition point Q. The relationship between the A / F for the A ₂ / A ₁ at this time is shown in Figure 21.

Therefore, A / F may be detected based on the data A ₂ / A ₁ to control the engine.

Next, FIG. 22 shows in-cylinder pressure characteristics in a diesel engine. (A) in FIG. 22 (a) shows a case where an in-cylinder pressure sensor is provided in the pre-combustion chamber of a pre-combustion chamber type engine. In the figure
(B) shows the case where the in-cylinder pressure sensor is provided in the combustion chamber of the direct injection engine.

Even in this case, the air-fuel ratio A / F is the ratio of the in-cylinder pressures P _a and P _b at the predetermined crank angles θ _a and θ _b set before and after the ignition point Q.
It can be detected by P _b / P _a or the ratio A ₂ / A ₁ of the integrated values A ₁ , A ₂ .

Further, the ignition point Q can be detected at the point where (dP / dθ) _max is reached. Therefore, similar to the process of FIG. 16, the ignition timing Ti
Instead of gn, the fuel injection timing Ting can be corrected to control the ignition timing of the diesel engine.

By the way, according to the cylinder pressure sensor of this embodiment,
Knocking can be detected, and FIG. 23 shows the result of detecting the in-cylinder pressure when knocking occurs. In FIG. 23, (a) is the case of a gasoline engine and (b) is the case of a diesel engine. In both cases, high frequency components appear near the maximum pressure. Similarly, FIG. 23 (c) is an enlarged view of only this high-frequency component, and the knock degree indicating the degree of knocking can be obtained by detecting ΔP of this high-frequency component.

FIG. 24 shows an example of a simple circuit for detecting ΔP,
In the figure, the part (a) is a filter part that passes only high frequency components and cuts low frequency components, and the part (b) is an integrating circuit that outputs an analog output according to the magnitude of ΔP. By using the circuit, a force corresponding to the high frequency ΔP can be obtained. In FIG. 25, this output is shown with the degree of knock on the horizontal axis, and it can be seen that the output and the degree of knock have a one-to-one correspondence.

Therefore, depending on this knock degree, knock control can be performed by correcting the ignition timing in the case of a gasoline engine and the injection timing in the case of a diesel engine. Fig. 24 is a flowchart showing an example of various processes, in which ΔP is detected by the circuit of Fig. 24 and Ip corresponding thereto is read, and ignition timing Tign in the case of a gasoline engine, injection timing Tinj in the case of a diesel engine. Then, T (Ip) is calculated by calculating the correction amount T (Ip) or by using the map shown as the parameters of the rotation speed N and the load Q. After this, this T
(Ip) corrects Tign and Tinj and outputs them.

〔The invention's effect〕

As described above, according to the cylinder internal pressure detecting means used in the present invention, the flexible member is formed of a thin disk-shaped member, and the optical fiber member is formed into a concentric circular shape.
The fiber element wires formed in a double annular shape are used for illumination, and the fiber element wires arranged in the middle of the fiber elements arranged in a triple annular shape are used for illumination. Since the fiber strand of is used for collecting the reflected light, the reflected light having a linear relationship with the pressure inside the cylinder can be obtained, and the pressure inside the cylinder can be accurately detected. Therefore, according to the present invention, without being affected by noise, etc.,
Moreover, since the in-cylinder pressure can be accurately detected and applied to the engine control, it is possible to perform control with excellent responsiveness by taking full advantage of the in-cylinder pressure detection method, excluding the drawbacks of the conventional technology. A cylinder pressure detection type engine control device can be easily provided.

[Brief description of drawings]

FIG. 1 is a diagram for explaining the principle of the in-cylinder pressure sensor according to the present invention, in which FIG. 1 (a) is a cross-sectional view, (b), (c) and (d) are operation explanatory diagrams,
FIG. 2 is a characteristic diagram thereof, FIG. 3 is a vertical sectional view showing an embodiment of the in-cylinder pressure sensor according to the present invention, FIG. 4 is a vertical sectional view including a connector, and FIG. FIG. 6 is an explanatory view showing an example, FIG. 6 is an explanatory view showing an embodiment of an optical system for the in-cylinder pressure sensor, FIG. 7 is a waveform diagram for explaining its operation, and FIGS. 8 and 9 are other ones. 10 (a) and 10 (b) are block diagrams showing an embodiment of a method of capturing a signal from the in-cylinder pressure sensor, a waveform diagram for explaining the operation thereof, and FIG. Similarly, (a) and (b) are block diagrams showing another embodiment and waveform diagrams for explaining the operation thereof, FIG. 12 is a block diagram showing one embodiment of a control system to which the present invention is applied, and FIG. 13 is FIG. 14 is a characteristic diagram showing the change in cylinder pressure, FIG. 14 is a characteristic diagram showing the relationship between cylinder pressure and A / F, and FIG. 15 is an explanatory diagram of the ignition point detection operation. Flowchart showing one embodiment of an ignition timing correction processing, FIG. 17 is a characteristic diagram of the A / F sensor, FIG. 18 are views for explaining the operation of the O ₂ sensor, Fig. 19 is A
20 is a flow chart showing an embodiment of the A / F correction processing, FIG. 20 is an explanatory view of another embodiment for obtaining A / F from the cylinder pressure characteristic, and FIG. 21 is a characteristic diagram showing the relationship between pressure and A / F. 22 is an explanatory view of cylinder pressure characteristics of a diesel engine, FIG. 23 is an explanatory view of knock detection, FIG. 24 is a circuit diagram showing an embodiment of a circuit for knock detection, and FIG. 25 is its operation. FIG. 26 is a characteristic diagram for explanation, and FIG. 26 is a flowchart showing an embodiment of a correction process necessary for knock control. 1 ... Illumination fiber, 2 ... Inner collection fiber, 3
…… Outside collection fiber, 4 …… Diaphragm, 6 ……
Sensor housing, 7 ... Fiber bundle, 8 ... Filling material, 10 ... Diaphragm chamber, 14 ... In-cylinder pressure sensor, 20
0 ...... Spark plug, 201 ...... Signal processing circuit, 202 ...... Microcomputer, 203 ...... Crank angle sensor, 204
...... Air-fuel ratio sensor, 205 ...... Ignition coil, 20
6 ... Injection valve.

 ─────────────────────────────────────────────────── --- Continuation of the front page (56) References JP-A-57-163128 (JP, A) JP-A-57-29912 (JP, A) JP-A-58-93958 (JP, A) Actual development Sho-58- 28395 (JP, U)

Claims (2)

[Claims] 1. A flexible member that bends in accordance with an internal pressure of a cylinder of an engine, and an optical fiber that takes out a change in a light reflection direction on the surface of the flexible member due to the bending as a light amount change. In the engine control device, which includes an in-cylinder pressure detecting means having a member, and controls the operation of the engine in accordance with the in-cylinder pressure detected by the in-cylinder pressure detecting means, the flexible member is a thin disk-shaped member. And the optical fiber member is formed of fiber strands which are concentrically arranged in a triple annular shape, and which are arranged in the middle of the fiber elements arranged in the triple annular shape. An internal cylinder pressure detection engine control device characterized in that the wire is used for illumination, and the fiber wire on the inner circumference side and the fiber wire on the outer circumference side are used for collecting reflected light. 2. The cylinder pressure detecting means according to claim 1, wherein the cylinder pressure detecting means is attached to a metal body of an ignition plug of an engine, and is inserted into the cylinder through a pressure hole formed in the metal body. An in-cylinder pressure detection type engine control device, which is configured to communicate with each other.