JP5103600B2 - Measuring method of instantaneous flow rate of gaseous fuel injector - Google Patents

Description

  The present invention relates to a method for measuring an instantaneous flow rate of a gaseous fuel injector that accurately measures an injection rate (also referred to as “mass flow rate”) of gaseous fuel ejected from the gaseous fuel injector at each time.

In recent years, as fuel for automobiles, in addition to liquid fuels such as gasoline and light oil, hydrogen and gaseous fuels typified by compressed natural gas (CNG) have attracted attention. In particular, CNG is used as a fuel for automobiles because it has the advantage of (i) low CO ₂ emission per unit calorific value and (ii) no sulfur oxide (SOX) emission because it contains almost no sulfur. The movement to use as is progressing. When gaseous fuel such as hydrogen or CNG is used in an automobile engine, hydrogen and CNG must be injected from a fuel injector (also referred to as a fuel injection valve) into an intake port or a cylinder, as in conventional gasoline / light oil. .
The application range of hydrogen, CNG, etc. is extremely wide, and various studies are being made such as an intake pipe injection spark ignition engine, an in-cylinder direct injection spark ignition engine, and an in-cylinder direct injection compression ignition engine.

Since the injection rate of the fuel injector (fuel injection amount per unit time: m ³ / sec or g / sec) has a great influence on the engine performance, accurate measurement of this injection rate is extremely important in the design of fuel injectors and engines. It becomes an important factor. There are many fuel injection rate measurement methods for conventional liquid fuels such as gasoline and light oil. In particular, a “Bosch injection rate meter” that guides injected fuel into a pipe having a constant cross-sectional area and measures the injection rate on the assumption of a one-dimensional flow is widely used (for example, see Non-Patent Document 1). .)

This Bosch injection rate meter is a method in which injected fuel is injected into a pipe having a constant cross-sectional area, and the flow rate of the fuel is obtained from an increase in pressure in the pipe. This is based on the principle that when the fuel is injected from the fuel nozzle into the measuring tube, the injection rate (m ³ / sec) is obtained by multiplying the cross-sectional area (m ² ) of the tube by the flow velocity (m / sec). We are using. That is, when the fuel injected from the injection nozzle flows through a thin pipe, a pressure increase in the pipe occurs, and this pressure increase is measured to determine the instantaneous flow rate of the fuel.

  However, although the Bosch injection rate meter can be used for measurement of liquid fuel, it cannot be used for gaseous fuel such as hydrogen and compressed natural gas (CNG). In other words, when this Bosch injection rate meter is applied to a gaseous fuel injector, the flow rate is measured due to the compressibility unique to gas, that is, the property that density changes with pressure change, and the increase in tube friction due to the increase in Reynolds number. There was a problem that would be impossible.

  For this reason, the inventors have proposed a method in which a one-dimensional flow of injected gaseous fuel is created in the same manner as the Bosch type fuel injection rate meter, and the injection rate from the gaseous fuel injector is obtained from the static pressure fluctuation. (See Patent Document 1).

  The gaseous fuel injection rate meter described in Patent Document 1 has a configuration in which a measurement pipe is connected to a gaseous fuel injector whose electromagnetic valve is controlled to be opened and closed, and an extension pipe for removing reflected waves is provided in the measurement pipe. ing. Then, the pressure in the measuring tube is measured by a pressure meter closely arranged in a small hole provided in the measuring tube, and this pressure is converted into the flow rate of the gaseous fuel by a predetermined conversion formula to obtain the injection rate of the gaseous fuel. Is.

Hiroshi Hayashi "Bosch injection rate meter" (Internal combustion engine Vol. 7, No. 12, pp. 58-64) International publication pamphlet WO2006 / 104176

  However, the gaseous fuel injection rate meter disclosed in Patent Document 1 is not sufficient in terms of measurement accuracy. In other words, the already-proposed device described in Patent Document 1 has a problem that a measurement error of about 10% is unavoidable, and a device for reducing this error as much as possible has been demanded.

  The present invention has been made in view of the above problems, and as a device for reducing measurement errors, (1) an improved relational expression for calculating an injection rate from a static pressure fluctuation in a measurement pipe into which gaseous fuel is injected. And (2) correcting the static pressure measured based on the pressure gradient due to pipe friction, thereby improving the accuracy of the gaseous fuel injection rate meter.

  In order to solve the above-mentioned problems and achieve the object of the present invention, the invention according to claim 1 is a pressure measurement method comprising a step of injecting gaseous fuel from a gaseous fuel injector into a measuring tube and a small hole provided in the measuring tube. The process of measuring the static pressure in the measuring tube with a measuring instrument, the process of calculating the density, velocity and sound speed of gaseous fuel from the static pressure measured with the pressure measuring instrument, and the static pressure measured with the pressure measuring instrument by pipe friction A gaseous fuel injector comprising: a step of correcting with a pressure gradient; and a step of calculating an injection rate (mass flow rate) of the gaseous fuel by substituting the static pressure corrected with the pressure gradient caused by the pipe friction into a predetermined calculation formula It is a measuring method of the gaseous fuel injection rate from.

（ただし、Ａは計測管の断面積（ｍ^２）、ρ（ｔ）は気体燃料の密度（ｋｇ／ｍ^３）、ｕ（ｔ）は気体燃料の流速（ｍ／ｓｅｃ）、γは比熱比（無次元の断熱係数）、Ｐ（ｔ）は計測管内の圧力（Ｐａ）、Ｐ_１＝Ｐ（０）、ρ_１＝ρ（０）である。） The invention described in claim 2 limits the predetermined arithmetic expression in the invention described in claim 1 to the following expression.

(Where A is the cross-sectional area of the measurement tube (m ² ), ρ (t) is the density of the gaseous fuel (kg / m ³ ), u (t) is the flow velocity of the gaseous fuel (m / sec), and γ is the specific heat ratio. (Dimensionless adiabatic coefficient), P (t) is the pressure (Pa) in the measuring tube, P ₁ = P (0), ρ ₁ = ρ (0).)

  According to the measurement method of the gaseous fuel injection rate of the present invention, by improving the calculation formula for obtaining the injection rate (mass flow rate) used in the conventional method, and by obtaining the static pressure corrected by the pressure gradient due to the pipe friction, The error between the actual injection rate and the measured injection rate can be extremely reduced. As a result, the measurement of the injection rate that has not hitherto reached the practical level can be greatly improved to the practical level.

Hereinafter, a method for measuring an instantaneous flow rate of gaseous fuel according to an embodiment of the present invention will be described with reference to the drawings.
Similar to the invention described in Patent Document 1, the basic configuration of the present invention is similar to the Bosch type fuel injection rate meter. That is, even when the method of the present invention is carried out, it has a common point with the Bosch fuel injection rate meter in that the injection rate is obtained assuming a one-dimensional flow.

<Description of gaseous fuel injection rate meter>
First, as a premise for explaining an example of an embodiment of the method of the present invention, an apparatus configuration for carrying out the method of the present invention (hereinafter sometimes simply referred to as “this example”) will be described.
FIG. 1 is a schematic diagram showing an outline of a gas fuel instantaneous flow meter used in carrying out the method of the present invention. The gas fuel flow meter of this example includes a gas cylinder 1, a buffer chamber 2 connected to the gas cylinder 1, a gaseous fuel injector 3 through which the fuel gas is guided through the buffer chamber 2, a pressure measuring instrument 4, and a gas It is mainly composed of an inflow measuring tube 5 and an extension tube 6. In order to supply gaseous fuel gas to the measuring tube 5, an injector driver 7 that controls opening and closing of an on-off valve (not shown) of the gaseous fuel injector 3 with a solenoid, and a function generator 8 that supplies a rectangular wave pulse to the injector driver 7 are provided. Further, a digital oscilloscope 9 for observing the waveform generated from the function generator 8 and observing the pressure in the measuring tube measured by the pressure measuring instrument 4 is provided. The oscilloscope 9 does not have to be for digital use, and may be for analog use.

  FIG. 2 is a diagram showing in more detail the portions of the gaseous fuel injector 3 and the pressure measuring instrument 4 used in this example. 2A is a schematic diagram showing the overall configuration of the gaseous fuel injector 3 and the pressure measuring unit 4, and FIG. 2B is a side sectional view of FIG. 2A (B in FIG. 2C). FIG. 2C is a cross-sectional view of a portion indicated by AA in FIG. 2B.

  The gaseous fuel injector 3 includes a fuel injection part 31, a solenoid valve opening / closing part 32, and a fuel injection part 33. Further, the measurement tube 5 is provided with a small hole (static pressure hole) 42, and the pressure transducer 41 of the pressure measuring unit 4 is disposed in close contact with the small hole 42. A piezoelectric element 43 is provided inside the pressure transducer 41. The gaseous fuel injector 3 is fixed to a base (not shown) by a flange 11 and the pressure transducer 41 is fixed by a flange 12a.

Next, based on FIG.1 and FIG.2, operation | movement of the instantaneous flow volume measuring apparatus of the gaseous fuel of this example is demonstrated.
In FIG. 1, the gaseous fuel from the gas cylinder 1 is guided to the buffer chamber (pressure vessel) 2 along the arrow A in the figure. The buffer chamber 2 is provided to remove the pulsation of the gaseous fuel supplied from the gas cylinder 1, and the pressure of the gaseous fuel gas supplied from the buffer chamber 2 is made substantially constant, so that the gaseous fuel injector 3 To be supplied.

  As shown in FIG. 2, an electromagnetic valve (not shown) is provided in the electromagnetic valve opening / closing part 32 of the gaseous fuel injector 3, and this electromagnetic valve is opened / closed based on a control signal from the injector driver 7. Here, the valve opening period of the electromagnetic valve in one injection is several msec to several tens msec. By opening and closing the electromagnetic valve, the gaseous fuel supplied from the buffer chamber 2 to the gaseous fuel injector 3 is injected into the measuring tube 5 from the fuel injection unit 33. Here, the injection port of the fuel injection unit 33 has a diameter of 1 to 2 mmφ, the inner diameter of the measuring tube 5 is 8 mmφ (the cross-sectional area is constant), and the length is about 100 mm.

  Normally, when fuel of the same mass is injected from the gaseous fuel injector 3 into the measuring tube 5, its volume is roughly 1000 times that of liquid fuel. For this reason, since the speed of the gaseous fuel in the measuring tube 5 is also 1000 times, in order to reduce this speed, the inner diameter of the measuring tube 5 is increased to 8 mφ with respect to the diameter of the injection port of the fuel injector. ing. Due to this sudden expansion of the diameter, the cross-sectional area becomes 16 times or more, so the flow in the measuring tube 5 is greatly disturbed. In other words, turbulence occurs. For this reason, in the invention described in Patent Document 1, a stepped nozzle 10 having a tapered shape or an opening that gradually increases is provided. By providing this nozzle 10, pressure vibration in the measuring tube 5 was eliminated.

  As described above, in this example, the tapered or stepped nozzle 10 is provided, but the provision of the nozzle 10 causes another problem. That is, it is necessary to replace the taper nozzle 10 each time in accordance with the nozzle diameter of the gaseous fuel injector, and the flow is accelerated in the taper nozzle 10, so that the pressure rises too much to cause “overshoot”. Is that there is.

  As described above, the tapered nozzle 10 has an advantage of eliminating the pressure vibration, but has the above-described drawbacks. Therefore, as another embodiment of the present invention, the pressure can be reduced without providing the nozzle 10. A method of reducing vibration has been devised. That is, when the taper nozzle 10 is not used, it is known that the pressure vibration due to the flow disturbance appears in a relatively high frequency band on the order of 50 kHz or more. For this reason, it is possible to remove pressure vibration by performing “low-pass filter processing” on the acquired data or “moving average processing”.

  Here, “moving average processing” means an average method using adjacent data. That is, as will be described later, the pressure vibration in the measurement tube is detected by a pressure sensor (see FIG. 2), sampled by a digital oscilloscope, for example, every 2 μsec, and acquired as time series data. When the vibration of the data obtained in this way is intense, fine fluctuations can be removed by averaging the data before and after a certain time, for example, 10 μs (5 points before and after). By performing such processing, the same effect as the low-pass filter can be obtained.

  In FIG. 2, the measurement tube 5 is provided with a small hole (static pressure hole) 42 having a diameter of about 1 mmφ at a position 100 mm downstream from the injection port. Sensor) 41 is arranged in close contact. As the pressure transducer 41, for example, a piezo-type pressure sensor (Piezotronics, H112A22) is used, and the static pressure in the measuring tube 5 is measured. That is, when the gaseous fuel is injected into the measuring tube 5, a pressure wave is generated in the measuring tube 5, and the pressure in the measuring tube 5 fluctuates. This pressure fluctuation is detected by the pressure transducer 41. The The signal from the pressure transducer 41 is recorded by a digital oscilloscope 9 (Recroy, WR6030A).

  Furthermore, in this example, an extension pipe 6 is attached downstream of the measurement pipe 5. The diameter of the extension pipe 6 is 8 mmφ which is the same as that of the measuring pipe 5 and the length is as long as 4 m. Thus, the reason why the length of the extension pipe 6 is long is to eliminate the influence of the reflected wave in the pressure measurement section of the measurement pipe 5. That is, the pressure wave accompanying the fuel injection is reflected at the downstream end of the extension pipe 6 and returns to the pressure measurement portion of the measurement pipe 5 as a reflected wave. When this reflected wave is superimposed on a normal pressure wave accompanying fuel injection, measurement becomes impossible, and therefore an extension pipe 6 is provided to gain time until the reflected wave arrives. The downstream end of the extension pipe 6 may be opened to the atmosphere, or the back pressure valve 13 is installed at the downstream end of the extension pipe 6 to reduce the flow rate, thereby reducing the static in the measurement pipe 5 and the extension pipe 6. The pressure rises uniformly and can simulate fuel injection under high pressure conditions in the engine cylinder.

<Description of formulation>
In the measurement of the instantaneous flow rate of the gaseous fuel used in the embodiment of the present invention, there is a feature in the mathematical formula for deriving the injection rate (instantaneous flow rate per unit time) of the gaseous fuel from the pressure detected by the pressure converter 41. Hereinafter, a method for deriving the mathematical expression will be described with reference to FIG.

In order to explain the method of deriving the instantaneous fuel in the present invention, the following theory assumes that the flow is one-dimensional, compressible, non-viscous, and adiabatic.
FIG. 3 is a schematic diagram showing the propagation of the pressure wave in the measuring tube 5 and the change of each gas variable. The X axis is the direction of flow and indicates the travel distance of the gaseous fuel injected from the injector. t is time. The cross-sectional area of the measuring tube 5 is A (m ² ), and the gas velocity (flow velocity), sound velocity, static pressure, density, static temperature, and total temperature are u (m / sec), a (m / sec), P (kPa), ρ (kg / m ³ ), T (K), and T _t (K). Here, the total temperature is the sum of thermal energy and kinetic energy, and the static temperature is the temperature obtained by subtracting the kinetic energy from the total temperature. Therefore, when the stationary gaseous fuel is injected and starts to move, the thermal energy is converted into kinetic energy, so that the static temperature is lower than the total temperature.

First, it is assumed that gas flows from the injector into the measuring tube 5 at a minute speed du. As a result, the gas (region 3) in front of the substance boundary (the boundary between region 2 and region 3 in FIG. 3) is compressed, and a compression wave propagates at the speed of sound a. The velocity of the gas behind the compression wave is du. After a minute time dt, the substance boundary reaches a distance dudt. This is region 2. At this time, the compression wave reaches the distance dt. Here, the flow field is divided into the following regions 1 to 4.
Region 1: A stagnation-point gas region upstream from the injector nozzle and an inner region of the injector 3.
Region 2: A region of gas injected from the injector and guided into the measurement tube 5.
Region 3: The gas originally in the measuring tube 5 is compressed by the gaseous fuel injected from the injector.
Region 4: This is a region that is originally in the measurement tube 5 but is not affected by the gaseous fuel injected from the injector.

  In the invention disclosed in Patent Document 1, the calculation formula is derived on the assumption that all the variables are continuous between the region 2 and the region 3, but in reality, there is a substance between the region 2 and the region 3. Since there is a boundary, the sound speed a, density ρ, and static temperature T are discontinuous. In the embodiment of the present invention, a new formulation was attempted in consideration of the discontinuity of these parameters.

  Hereinafter, the process of deriving the formula used in this example will be described. In this example, when gas is injected from the injector into the measuring tube 5 between the region 1 and the region 2, the flow velocity u, the sound velocity a, the static pressure P, the density ρ, and the static temperature T are as shown in FIG. Change. Further, between the region 2 and the region 3, the velocity u and the static pressure P are equal values from the mechanical equilibrium condition and the continuous condition. On the other hand, the static temperature T does not match between the region 2 and the region 3. This is because the gas in the region 2 is a gas expanded from the injector, and thus accelerates while the static temperature is lowered. On the other hand, the gas in the region 3 is originally in the measuring tube 5 and is compressed by the injected gaseous fuel to increase the pressure, so that the static temperature T rises. The relationship between the pressure, density, and temperature of the gas satisfies the gas equation of state (P = ρRT: R is a gas constant). For this reason, the sound speed a and the density ρ are also discontinuous between the region 2 and the region 3. Therefore, the speed of sound, density, and static temperature in region 2 are set as a ′, ρ ′, and T ′, respectively, and the following formulas are derived.

In addition, the boundary part of the area | region 3 and the area | region 4 is a wave front of a pressure wave. For this reason, all the variables slightly change between the region 3 and the region 4. In the region 1 and the region 4, since the gas velocity u is both “0”, the static temperature T (the static temperature T is the same as the total temperature T _t since u = 0) is the same.

すなわち、ρ（ｋｇ／ｍ３）、ａ（ｍ／ｓｅｃ）、Ａ（ｍ２）、ｄｔ（ｓｅｃ）であるから、ρａＡｄｔの次元は（ｋｇ）であり、質量の次元となっている。 Under the above preconditions, gas flows from the gaseous fuel injector 3 in the region 1 into the region 2 in the measuring tube 5. For this reason, the gas originally in the measurement tube changes from the length adt (the length obtained by adding the region 2 and the region 3) to the length (a-du) dt of the region 3. Thus, although the volume of the gas that was originally in the measurement tube is compressed by the injected gaseous fuel, the mass does not change before and after the compression, and therefore equation (1) is established.

That is, since ρ (kg / m3), a (m / sec), A (m2), and dt (sec), the dimension of ρaAdt is (kg), which is the dimension of mass.

この（１）式と（２）式から、二次の微小項を省略して演算することにより、（３）式が成立する。

Subsequently, since the momentum changes before and after the pressure wavefront (the boundary between the region 3 and the region 4) have the same impulse, the equation (2) is established. That is, since the pressure on the region 3 side as viewed from the wavefront of the pressure wave is (P + dP) and the pressure in the region 4 is P, the left side of the equation (2) is an impulse (N) that pushes the wavefront per unit time.・ Sec). On the other hand, the right side of equation (2) shows the product of mass per unit time ρaA ((kg / m ³ ) * (m / sec) * (m ² ) * sec = kg) and velocity change before and after the wavefront (momentum change). ). That is, the momentum in region 4 is a value obtained by subtracting ρaA * (a-du) from ρaA * a (ρaA is the moving mass per unit time, and a is the speed of sound).

From the equations (1) and (2), the calculation is performed by omitting the secondary minute term, so that the equation (3) is established.

  As can be seen from the equation (3), the speed change du is a value obtained by dividing the static pressure change dP by the product of the sound speed a and the density ρ. The product of the sound speed a and the density ρ is a value called acoustic impedance. In the Bosch liquid fuel injection rate meter, this value is formulated as a constant, but in a compressible fluid such as this example, both the speed of sound a and the density ρ are variables, and values that change with time. It becomes.

Hereinafter, all functions are expressed as functions of time. Therefore, equation (3) can be written as equation (4).

但し、γは比熱比、Ｐ_０、ρ_０は、計測管５内の初期静圧と初期密度である。（５）式から、ある時刻ｔにおける密度ρ（ｔ）は、（６）式で求められる。

By integrating both sides of the equation (4), the relational expression between the static pressure and the velocity in the measuring tube 5 can be finally obtained. In the embodiment of the present invention, only the pressure in the measuring tube 5 is measured. For this reason, it is necessary to express the sound speed a (t) and the density ρ (t) in the equation (4) as a function of the static pressure P (t) in the pipe.
Since the adiabatic change can occur in the measuring tube 5, the equation (5) is established.

Where γ is the specific heat ratio, and P ₀ and ρ ₀ are the initial static pressure and the initial density in the measuring tube 5. From the equation (5), the density ρ (t) at a certain time t can be obtained by the equation (6).

On the other hand, from equation (2), the sound speed a (t) can be expressed as equation (7).

（６）式と（８）式より、密度ρ（ｔ）と音速ａ（ｔ）は、計測管５内の圧力Ｐ（ｔ）として表すことが可能となる。 Substituting equation (6) into equation (7) establishes equation (8).

From the equations (6) and (8), the density ρ (t) and the sound velocity a (t) can be expressed as the pressure P (t) in the measuring tube 5.

この（９）式を積分すると、（１０）式が成立する。

ここで、Ｃは積分定数であるが、（１０）式で、時刻ｔ＝０のとき、速度ｕ（０）＝０、計測管５内の静圧Ｐ（０）＝Ｐ_０であるから、この初期値を代入して、（１０）式を書き直すと、（１１）式になる。
すなわち、各時刻ｔにおける速度ｕ（ｔ）は、計測管５内の圧力Ｐ（ｔ）のみの関数として表すことができる。

Substituting equations (6) and (8) into equation (4) establishes equation (9).

When this equation (9) is integrated, equation (10) is established.

Here, C is an integral constant. In the equation (10), when time t = 0, the speed u (0) = 0 and the static pressure P (0) = P ₀ in the measurement tube 5 Substituting this initial value and rewriting equation (10) yields equation (11).
That is, the speed u (t) at each time t can be expressed as a function of only the pressure P (t) in the measurement tube 5.

  Therefore, the gas velocity u (t) at each time t can be obtained by substituting the pressure P (t) measured in the measuring tube 5 into the equation (11).

ここで、ρ´（ｔ）は領域２における気体燃料の密度であるが、このρ´（ｔ）を、計測管５内の圧力Ｐ（ｔ）の関数として表す必要がある。
そこで、領域１と領域２の間でのエネルギー保存側を考慮すると、（１３）式が成立する。

但し、Ｔ_０は初期静温（＝全温）、Ｒは気体定数である。 On the other hand, the gaseous fuel injection rate m (t) is an instantaneous flow rate of the gas that has passed between the region 1 and the region 2 in FIG. 3 and can be expressed by equation (12). Here, in the equation (12), a dot (•) is added on m in the sense that m (t) represents an instantaneous flow rate.

Here, ρ ′ (t) is the density of the gaseous fuel in the region 2, and this ρ ′ (t) needs to be expressed as a function of the pressure P (t) in the measuring tube 5.
Therefore, considering the energy conservation side between region 1 and region 2, equation (13) is established.

However, _{T 0} is the initial ShizuAtsushi (= total temperature), R is the gas constant.

したがって、各時刻における気体燃料噴射率ｍ（ｔ）は、（１１）式、（１２）式及び（１４）式より、（１５）式として求めることができる。

すなわち、気体燃料噴射率ｍ（ｔ）は、計測管５内の静圧Ｐ（ｔ）のみの関数となる。
この（１５）式を見ると分かるように、気体燃料噴射率ｍ（ｔ）の算出において、インジェクタの上流（図３の領域１）の情報は一切必要とされない。
もともと計測管５内にあった気体の変数初期値と、噴射後の計測管内の静圧変化のみで気体燃料の噴射率の計測が可能となることが分かる。 From this equation (13), the density ρ ′ (t) of the region 2 can be expressed by equation (14).

Therefore, the gaseous fuel injection rate m (t) at each time can be obtained as equation (15) from equations (11), (12), and (14).

That is, the gaseous fuel injection rate m (t) is a function of only the static pressure P (t) in the measuring tube 5.
As can be seen from the equation (15), no information on the upstream side of the injector (region 1 in FIG. 3) is required in calculating the gaseous fuel injection rate m (t).
It can be seen that the injection rate of the gaseous fuel can be measured only by the gas variable initial value originally in the measuring tube 5 and the static pressure change in the measuring tube after injection.

In general, in the case of liquid fuel, the injection rate is expressed as a volume flow rate. In the case of gas fuel, the volume changes with temperature and pressure, so the gas fuel injection rate is generally expressed as a mass flow rate. It is.
In the embodiment of the present invention, the gaseous fuel injection rate (mass flow rate) m (t) can be determined using the above-described calculation formula (15). Therefore, measurement of the static pressure in the measuring tube 5 is extremely important.

<Description of static pressure history in the measuring tube>
FIG. 4 shows the temporal change of the static pressure with respect to the input voltage. FIG. 4A shows the valve opening / closing signals of the injector, and FIG. 4B shows the static pressure in the measuring tube 5 at that time. The horizontal axis is time t (ms). Here, the input time t of the valve opening signal is set to zero. The injection time τ is 20 ms. The initial value of the static pressure of the measuring tube 5 is P (0) = P0 = 101 (kPa).
FIG. 4A shows a rectangular signal given from the function generator 8 to the injector driver 7. FIG. 4B is a waveform diagram of the change in pressure at the pressure measurement portion when the rectangular signal is supplied to the gaseous fuel injector 3 with the digital oscilloscope 9.

As shown in FIG. 4, when a valve opening signal is input at time t = 0 (ms), the static pressure P in the measuring tube rapidly increases with a time delay of about 2 ms. That, along with the injection start of the gaseous fuel from the injector, the gas in the measurement pipe 5 is compressed, the static pressure P increases by [Delta] P _1. The time delay until the static pressure P starts to rise is due to the time delay from the input of the valve opening signal to the start of gaseous fuel injection. The time required to increase the static pressure is about 0.5 ms. This period is a valve opening transition period of the injector, and is a time required until the solenoid valve inside the injector increases and becomes fully open.

Time t = 3 ms later, the static pressure within the measuring tube 5 is further increased by [Delta] P _2. During this period, the injector is in a steady injection with the injector fully open, and the static pressure P in the measuring tube should be constant. This increase in static pressure is thought to be due to pipe friction. This increase in static pressure requires correction, which will be described later.

Next, when a valve closing signal is input at time t = 20 ms, the static pressure P in the measuring pipe 5 rapidly decreases after a time delay of about 1 ms. This is because the supply of gaseous fuel from the injector stops and the gas in the measuring tube 5 expands, so that the static pressure P decreases. The time required for this static pressure P to decrease is about 1 ms.
This period is a transition period during which the injector is closed, and is a period of time from when the solenoid valve stroke inside the injector is reduced until it is fully closed.

Next, after time t = 22 ms, the static pressure P in the measurement tube 5 is slightly larger than the initial value P0. During this period, the injector is fully closed, and the injection rate is zero. Deviation from the initial value P ₀ of the static pressure P of the period is also considered to be due to pipe friction.

  At time t = 27 ms, the static pressure P in the measurement tube 5 starts to decrease. Of course, the injector is not operating during this period. This decrease in static pressure is due to the reflected wave generated at the downstream end of the extension pipe 6 (see FIG. 1) returning to the measurement unit. When the reflected waves are superposed in this way, pressure measurement no longer makes sense, so all measurements must be made before the reflected waves return.

<Explanation on tube friction correction>
Change [Delta] P ₂ static pressure of 4 has been described to be caused by pipe friction, the method of correcting the pipe friction will be described with reference to FIG. Fig.5 (a) is the schematic diagram which showed the pressure gradient by the pipe | tube friction in the measurement pipe | tube 5 (refer FIG. 2). The uppermost stream of the measurement tube 5 shown in FIG. 5A is the injection surface from the gaseous fuel injector 3, and corresponds to the boundary between the region 1 and the region 2 in FIG.

As shown in FIG. 4B, when the gaseous fuel flows into the measuring tube 5 by opening the injector, the compression wave propagates through the measuring tube 5 and the static pressure in the measuring tube 5 Increases by ΔP ₁ . Thereafter, the injector is fully opened, and the steady injection period is reached. By this fuel injection, the gas in the range where the compression wave has reached gradually transitions to the turbulent state.

As in the case of this example, when gaseous fuel is injected into the measuring tube 5, the Reynolds number Re of the gas in the measuring tube 5 reaches tens of thousands. As described above, when the Reynolds number Re is large, the flow is disturbed in the measuring tube 5 and thus the tube friction occurs.
FIG. 5B shows the time change of the static pressure at the measurement point. When the gas fuel from a gas fuel injector 3 has elapsed predetermined by injection time, the static pressure P increases by [Delta] P _1. Thereafter, the pressure continues to be listed gradually while the injection continues. Increase in the meantime the static pressure is [Delta] P _2. FIG. 5B corresponds to FIG. 4B and shows that the static pressure changes by dP ₂ during the minute time dt. This occurs depending on the pressure gradient dP ₂ / dx due to the pipe friction shown in FIG.

  As shown in FIG. 5B, the static pressure P varies depending on the position in the measurement tube 5 even during the steady injection period. That is, the static pressure P in the measuring tube 5 that measures the static pressure P is high in the upstream and low in the downstream. Thus, if the static pressure in the pipe increases due to the occurrence of pipe friction, the apparent mass flow rate will increase, making it difficult to measure the static pressure in the measurement pipe.

すなわち、静圧の時間変化は圧力勾配ｄＰ_２／ｄｘと速度ａ（ｔ）の積で表すことができる。 Here, the relationship between the pressure gradient dP ₂ / dx due to pipe friction and the static pressure time change dP ₂ / dt is considered. The pressure distribution in a certain minute section dx propagates in the pipe at the sound velocity a (t) at that position. The time required for this minute section dx to pass through the measurement point is dx / a (t), and the static pressure P changes by dP ₂ during this time.
Therefore, the time change dP ₂ / dt of the static pressure P can be expressed by equation (16).

That is, the time change of the static pressure can be expressed by the product of the pressure gradient dP ₂ / dx and the speed a (t).

On the other hand, the pressure gradient dP ₂ / dx due to pipe friction is determined by the Reynolds number Re of the flow in the pipe. In the invention described in Patent Document 1, the Reynolds number Re is decreased by increasing the inner diameter D of the measuring tube 5 and the pressure gradient due to tube friction is suppressed. However, this method has a problem that (i) it is necessary to change the inner diameter of the measuring tube 5 in accordance with the injection rate of the injector to be used, and (ii) the pressure gradient cannot be made completely zero. It was. Therefore, in this example, the magnitude of the pressure gradient dP ₂ / dx was theoretically estimated, and by correcting the amount, the static pressure was measured more accurately.

ここで、ｆは管摩擦係数であり、レイノルズ数Ｒｅが１０^５以下の滑らかな管内の乱流の場合には、（１８）式で表される。

以上、説明したように、管摩擦による圧力勾配ｄＰ_２／ｄｘは、レイノルズ数Ｒｅ、計測管内径Ｄ、密度ρ、速度ｕで与えられ、前述した諸式を用いて予測することができる。 That is, the pressure gradient dP ₂ / dx due to tube friction can be expressed by the equation (17) from the Darcy-Weissbach equation.

Here, f is a pipe friction coefficient, and is expressed by the equation (18) in the case of turbulent flow in a smooth pipe having a Reynolds number Re of 10 ⁵ or less.

As described above, the pressure gradient dP ₂ / dx due to tube friction is given by the Reynolds number Re, the measurement tube inner diameter D, the density ρ, and the velocity u, and can be predicted using the above-described equations.

FIG. 6 shows experimental values and theoretical values of pressure gradient due to pipe friction. FIG. 6A is an experimental value, and FIG. 6B is a theoretical value obtained by the Darcy-Weissbach equation. The vertical axis represents the pressure gradient dP ₂ / dx, and the horizontal axis represents the Reynolds number Re.
FIG. 6A shows an experimental value of the pressure gradient obtained from the static pressure time change dP ₂ / dt using the equation (16). The injection pressure was gradually changed, and the Reynolds number Re during the steady injection period was changed variously.
FIG. 6B is a theoretical value of the pressure gradient obtained using the equations (17) and (18).

  It should be noted that the experimental value (a) and the theoretical value (b) in FIG. 6 are similar in tendency but have different absolute values. The theoretical value calculated using the equations (17) and (18) always has a pressure gradient value that is four times greater than the experimental value calculated using the actual change in static pressure time. In this regard, it has been confirmed that similar results can be obtained even when injectors with different specifications are used. The difference between the experimental value and the theoretical value is assumed to be caused by the unsteady flow unique to the gaseous fuel injection rate meter.

FIG. 7 shows a schematic diagram of turbulence in a pipe of a steady flow and an unsteady flow, where (a) shows a steady flow and (b) shows an unsteady flow.
That is, FIG. 7 (a) shows a fully developed tube turbulence. Expressions (17) and (18) are expressions that hold for such a sufficiently developed turbulent flow. On the other hand, in the case of the unsteady flow shown in FIG. 7B, the gas in the tube that was originally stationary suddenly starts to move from the moment when the pressure wave arrives, and the boundary layer grows. Eventually, the run-up section ends and transitions to turbulent flow.

この（１９）式で求められるΔＰ_２（ｔ）を、図４（ｂ）に示される計測された静圧履歴Ｐ（ｔ）から順次差し引くことにより、管摩擦の影響を補正することができる。 For this reason, in the case of unsteady flow, the value of the formed pressure gradient is smaller than the theoretical value. In this example, the ratio between the theoretical value of the pressure gradient and the experimental value was obtained, and the correction coefficient was α.
That is, the static pressure increase ΔP ₂ (t) due to tube friction at a certain time t is obtained by integrating the pressure change due to tube friction every minute time from time zero to time t. From the equations (16) to (17), the static pressure increase ΔP ₂ (t) can be expressed as the equation (19).

By sequentially subtracting ΔP ₂ (t) obtained by the equation (19) from the measured static pressure history P (t) shown in FIG. 4B, the influence of pipe friction can be corrected.

<Description of procedure for determining correction coefficient α>
The procedure for calculating the gaseous fuel injection rate in the present example described above has been described. This procedure will be described in detail based on the flowcharts shown in FIGS.

  FIG. 8 is a flowchart for deriving the correction coefficient α from the result of the preliminary experiment. In this flowchart, as a preliminary experiment, when a gas is injected from the gaseous fuel injector 3 at a certain fixed injection rate, an operation procedure for confirming how much the theoretical value and the actually measured value of the pressure gradient differ from each other is shown. It is shown. The ratio between the theoretical value of the pressure gradient and the actually measured value is the correction coefficient α.

  First, gaseous fuel is injected from the gaseous fuel injector 3 into the measuring tube 5 at a constant injection rate (step S1). Here, it is important to inject at a “constant injection rate”. When the gaseous fuel injector 3 is fully opened, the injection rate is constant. If the gas fuel pressure supplied to the gas fuel injector is set high, the injection rate becomes constant at a large value, and if the pressure is set low, the injection rate becomes constant at a small value. The injection time at this constant injection rate is 20 ms, for example. Next, the static pressure P (t) in the measurement tube 5 is measured by the pressure transducer 41 (see FIG. 2) closely arranged in the small hole 42 provided in the measurement tube 5 at the most downstream side of the measurement tube 5. (Step S2). The static pressure P (t) is a pressure that changes every moment with time.

  Next, the gas density ρ (t) is calculated by substituting the static pressure P (t) into the equation (6) (step S3). Also, the sound speed a (t) is calculated by substituting the static pressure P (t) into the equation (8) (step S4), and the gas fuel flow velocity u (t) is calculated by substituting it into the equation (11). (Step S5).

Subsequently, when the measured static pressure P (t) is time-differentiated and substituted into the equation (16), the pressure gradient dP ₂ / dx in the measurement tube 5 is obtained (step S6). On the other hand, the theoretical value of the pressure gradient due to pipe friction is also calculated from the Darcy-Weissbach equation shown by equation (17) (step S7). Although the experimental value and the theoretical value of the pressure gradient dP ₂ / dx obtained in this way show a similar tendency with respect to the change in the Reynolds number, the value has an opening of about four times as shown in FIG. As explained in. The ratio between the experimental value measured in this way and the theoretical value calculated from the Darcy-Weissbach equation is obtained as the correction coefficient α (step S8). This correction coefficient α is a device-specific value determined by the measurement tube inner diameter D. Therefore, once determined, the value can be used as a constant thereafter.
In the preliminary test, it was confirmed that the value of the correction coefficient α did not change by executing the process according to the above procedure while changing the injection rate in various ways. That is, the correction coefficient α does not depend on the injection rate.

<Description of calculation procedure for gaseous fuel injection rate (mass flow rate)>
FIG. 9 is a flowchart showing a procedure for obtaining from the static pressure P obtained by measuring the gaseous fuel injection rate using the correction coefficient α. In FIG. 9, the same steps as those in FIG. 8 are denoted by the same step numbers. However, in the flowchart of the preliminary experiment shown in FIG. 8, in step S1, gaseous fuel is injected from the gaseous fuel injector 3 into the measuring tube 5 at a constant injection rate, but in step S1 in FIG. 9, the mass flow rate is calculated. Therefore, the injection rate is not constant. The injection rate in FIG. 9 is an unknown number, and this injection rate is measured. In particular, it is a great advantage of this measurement method that it is possible to measure an abrupt change in the injection rate during the valve opening transient period, the valve closing transient period, and the like.

The ratio between the theoretical value (equation 17) and the experimental value (equation 16) of the pressure gradient (dP ₂ / dx in FIG. 5A) obtained in the preliminary experiment of FIG. 8 is obtained in step S8 of FIG. ing. Further, as described above, the pressure gradient (theoretical value) in the measuring tube 5 is obtained by the Darcy-Weissbach equation (step S10). When the pressure gradient obtained in step S10 is multiplied by the correction coefficient α, a pressure gradient close to the experimental value obtained from equation (16) is obtained from the time derivative of the static pressure P, and by integrating this, the pressure rise at time t is obtained. ΔP ₂ (t) is obtained (step S11).

The pressure increase ΔP ₂ (t) obtained in step S11 is subtracted from the static pressure P obtained in step S2, thereby obtaining (P (t) −ΔP ₂ (t)) (step S12). As a result, the pressure gradient due to the pipe friction in the measurement pipe is corrected, and the static pressure history in the pipe is as shown in FIG.

  The static pressure P obtained in step S12 is an instantaneous static pressure P (t) that changes with time, and this static pressure P (t) is substituted into equation (15) for obtaining a mass flow rate (injection rate). Thus, the instantaneous mass flow rate (injection rate) of the gaseous fuel is obtained (step S13).

<Description of measurement accuracy of gaseous fuel injection rate meter>
FIG. 10A shows the valve opening of the injector. FIG. 10B is a diagram showing the time history of the valve closing signal, and FIG. 10B is a diagram showing the in-tube static pressure history corrected by the equation (19). Moreover, FIG.10 (c) has shown the gaseous fuel injection rate m computed using (15) Formula. In all the drawings, the horizontal axis is time t, and the input time of the valve opening signal is the zero point on the horizontal axis. The injection time τ is 20 ms.

FIG. 10B shows that the influence of the static pressure increase ΔP ₂ (t) due to the pipe friction is completely corrected. As a result, even if the injector is changed as in the prior art, there is no need to change the inner diameter of the measuring tube each time, which is extremely effective. Further, as shown in FIG. 10C, it is possible to obtain a gaseous fuel injection rate that changes from moment to moment from the in-tube static pressure history measured at one point in the measuring tube 5. This makes it easier to grasp the injector characteristics such as the transient characteristics of the valve opening / closing signals.

<Explanation of calibration test>
FIG. 11 is a schematic diagram of an experimental apparatus used for the calibration test. This experimental device is for confirming the accuracy of the injection rate measured by the gaseous fuel injection rate meter. The same components as those in the first experimental apparatus are denoted by the same reference numerals.
In the calibration test using the experimental apparatus of FIG. 11, instead of injecting fuel from the fuel injector 3 toward the pressure measuring device 4 as in the experimental apparatus of FIG. 1, the gaseous fuel injector 3 is connected to the vacuum vessel 14. Fuel is injected into the vacuum vessel 14.
Here, since the pressure pulsation in the pipe upstream from the gaseous fuel injector 3 may affect the injection rate, the pipe from the gas cylinder 1 to the gaseous fuel injector 3 is the same as the experimental apparatus shown in FIG. The same thing is used. Note that the pressure ratio between the in-container pressure P _V and the injection pressure P _IJ is always considered to be smaller than the critical pressure ratio.

  Initially, a test was performed in which the inside of the vacuum vessel 14 was evacuated and the function generator 8 and the injector driver 7 were operated to inject gaseous fuel from the gaseous fuel injector 3 into the vacuum vessel 14. And injection period (tau) was fixed and 1,000-2,000 repetition injections were performed. The vacuum vessel 14 is provided with a pressure gauge 15 (mercury manometer) formed of a U-shaped glass tube.

From this experiment, the total mass flow rate of the gas was determined from the pressure increase in the vacuum vessel 14, and the mass flow rate per injection was determined by dividing the total mass flow rate by the number of injections. On the other hand, the injection rate at each time obtained by the gaseous fuel injection rate meter in the experimental apparatus of FIG. 1 was integrated, the mass flow rate per injection was obtained, and the two were compared.
Thereby, it was verified that the measurement accuracy of the gaseous fuel injection rate meter of this example was considerably improved as compared with the invention described in Patent Document 1.

FIG. 12 is a diagram showing the results of the calibration test. It is the figure which compared and showed the measurement result of the gaseous fuel injection rate meter, and the calibration test result. The vertical axis | shaft of Fig.12 (a) is a mass flow rate (mg / st), and has shown the mass flow rate per injection. In addition, the vertical axis in FIG. 12B indicates an error. The horizontal axis is the injection time τ.
As can be seen from FIG. 12 (a), the measurement result (◯ mark) of the gaseous fuel injection rate meter and the calibration test result (● mark) almost coincide. Also, from FIG. 12B, it can be seen that the error rate slightly changes depending on the injection time τ, but the difference between them is only 2 to 3%.
In the invention described in Patent Document 1, the measurement error of the gaseous fuel injection rate meter was about 10%. In contrast, in the measurement method of this example, the measurement accuracy is greatly improved by (i) improving the relational expression for calculating the injection rate from the static pressure in the pipe, and (ii) correcting the pipe friction. I was able to.

  As described above, according to the present invention, it can be said that the instantaneous flow rate measurement of gaseous fuel is close to the practical level. Therefore, it is considered that the method for measuring a gaseous fuel injection rate according to the present invention brings about an effective effect for designing a fuel injector and an engine of an automobile internal combustion engine or other engines using gaseous fuel.

It is a figure which shows an example of embodiment of the flow measuring device of the gaseous fuel injector of this invention. It is a figure which shows the principal part of the embodiment of the flow measuring device of the gaseous fuel injector of this invention. It is the schematic diagram which showed the propagation of the pressure wave in the measurement pipe | tube 5 in the gaseous fuel injector of this invention, and the change of each gas variable. It is a figure which shows the drive signal (a) which opens and closes the solenoid valve of a gaseous fuel injector, and the time change (before correction | amendment) (b) of the static pressure of a measurement point. Illustrates static圧履history of (b) in the measurement point pipe friction due to the pressure gradient _dP 2 / dx (a). Is a diagram showing the experimental values and the theoretical value of the pressure gradient dP 2 _/ dx by pipe friction, which is (a) experimental value, the theoretical value obtained by the formula (b) is Darcy Wise Bach. It is the schematic diagram which showed the disturbance of (a) steady flow and (b) unsteady flow in a pipe. It is a flowchart which shows the procedure which determines the correction coefficient (alpha) based on a preliminary test. It is a flowchart which shows the procedure which calculates the injection rate (mass flow rate) of gaseous fuel. (A) is the time history of the valve opening / closing signal of the injector, (b) is the static pressure history in the pipe corrected by the equation (18), and (c) is the gaseous fuel calculated using the equation (14). It is a figure which shows the injection rate m. It is the schematic which shows the experimental apparatus for performing the calibration test (calibration test) of a gaseous fuel injector. It is the figure which compared and showed the measurement result of the gaseous fuel injection rate meter, and the calibration test result.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 ... Gas cylinder of gaseous fuel, 2 ... Buffer chamber (pressure vessel), 3 ... Gas fuel injector, 4 ... Pressure measuring instrument, 5 ... Measuring pipe, 6 ... Extension pipe, 7 ... Injector driver, 8 ... Function generator, 9 ... Digital oscilloscope, 14 ... Vacuum container, 15 ... Pressure gauge

Claims (2)

Injecting gaseous fuel into the measuring tube from the gaseous fuel injector;
A step of measuring the static pressure in the measuring tube with a pressure measuring instrument through a small hole provided in the measuring tube;
Calculating the density and velocity of the gaseous fuel and the speed of sound from the static pressure measured by the pressure meter;
Correcting the static pressure measured by the pressure gauge with a pressure gradient caused by pipe friction;
Calculating the flow mass (injection rate) of the gaseous fuel by substituting the static pressure corrected by the pressure gradient due to the pipe friction into a predetermined calculation formula;
Measuring method of gaseous fuel injection rate from gaseous fuel injector including The method for measuring a gaseous fuel injection rate from a gaseous fuel injector according to claim 1, wherein the predetermined arithmetic expression is as follows.

(Where A is the cross-sectional area of the measurement tube (m ² ), ρ (t) is the density of the gaseous fuel (kg / m ³ ), u (t) is the flow velocity of the gaseous fuel (m / sec), and γ is the specific heat ratio. (Dimensionless adiabatic coefficient), P (t) is the pressure (Pa) in the measuring tube, P ₁ = P (0), ρ ₁ = ρ (0).)

Images