JP4299460B2 - Calculation method of indicated mean pressure of internal combustion engine - Google Patents

Description

[0001]
The present invention relates to an internal combustion engine in the field of internal combustion engines (i.e., ignition control engine) indicated mean pressure calculation method (i.e., the torque estimation method) of. In this field, along with improvements and developments in engine control and adjustment means, qualitative requirements regarding the operation of the engine have increased, and there is an increasing demand for accurate control of the overall operation of the engine.
[0002]
The control and adjustment means for the internal combustion engine include a richness probe, a motor-equipped throttle valve, a fully electronic ignition device, and the like. In particular, it is equipped in automobiles by knowing the total amount of engine torque as a function of the applied operating parameters such as richness, the number of revolutions, air charge amount, ignition timing advance, etc., which represents the ratio of the stoichiometric air-fuel ratio to the operating air-fuel ratio. It is possible to accurately control the operation of the engine, taking into account accessories such as air conditioners, power steering, automatic transmissions, anti-slip and track control devices that require energy supply directly or indirectly from the engine It is expected to be.
[0003]
It has been proposed to directly measure the engine torque with a special sensor while the engine is running. For example, the torque of the engine can be measured by measuring with a fixed sensor the period through which the tooth rows of the measurement rim fixed to the engine flywheel pass.
[0004]
It has also been proposed to calculate engine torque using various sensors such as air flow meters, richness probes, ignition and speed sensors, collector pressure and temperature sensors attached to the engine. According to the French publication No. 91 11 919, the effective torque of the engine at the optimum ignition timing advance is determined by a plotting method as a function of the speed and the collector pressure. Next, the indicated torque is obtained by adding the friction torque to this torque using a numerical table that is a function of the rotational speed. Next, this indicated gas torque is corrected by a specific ignition timing advance efficiency. The actual torque corresponding to the applied ignition timing is finally obtained by newly subtracting the loss due to friction from this gas torque.
[0005]
Further, in the publication, WO-9635874, a torque whose richness is 1 and the ignition timing advance is the optimum ignition timing advance is obtained by a drawing method, and the torque decreases as the ignition timing advance deviates from the optimum ignition timing advance. It is described that the correction is made by the ignition timing progress efficiency according to the law.
[0006]
However, the methods described in these documents do not provide a sufficiently accurate estimate of the torque value, and therefore the engine maximum, whether for reducing fuel consumption or for reducing combustion emissions. There is no optimal control that can achieve efficiency.
[0007]
The present invention uses the ignition timing advance efficiency, which is a variable that represents the effect on the indicated mean pressure that a particularly precisely defined ignition timing advance departs from the optimal ignition timing advance. The purpose is to solve the problem.
[0008]
The illustrated average pressure calculation method for an internal combustion engine according to the present invention (i.e., torque estimation method for an ignition control engine ) is an engine in which the intake valve and the exhaust valve are closed during a pressure cycle in the cylinder that varies with the volume of the cylinder. This is performed by calculating the indicated mean pressure of the positive loop corresponding to the above operation. This estimation is performed by calculating as a function of the air flow into the engine, the pressure in the collector of the engine, the operating speed N, the ignition timing advance AV, and the apparent richness Ri of combustion. . The above indicated mean pressure PMI + for the positive loop is
PMI ₊ = PMI + _opt, is _{Ri = 1 × π Ri × π} AV,
Here, PMI + _{opt, Ri = 1} is the indicated mean pressure when the ignition timing advance is the optimum ignition timing advance and the richness Ri is 1,
π _Ri is a richness coefficient that is a variable that represents the influence of the richness Ri away from 1 on the indicated mean pressure,
π _AV is the ignition timing advance efficiency, which is a variable representing the influence of the ignition timing advance away from the optimal ignition timing advance on the indicated mean pressure,
In the case where the ignition timing advance is the optimum ignition timing advance and the richness Ri is 1, the indicated average pressure PMI + _{opt, Ri = 1} is
_{PMI + opt, Ri = 1 =} K (P col, N) is a × Q _r,
Where K is a coefficient that depends on the pressure P _col in the collector and the operating speed N at the same time,
Q _r is Ri airflow der flowing into the engine per (1/2 rotation of the crankshaft 4 cylinder) stroke,
Ignition timing advance efficiency π _AV
Is a regression polynomial of a quadratic function for the distance between the optimum ignition timing advance AV _opt at the operating point and the applied ignition timing advance AV, and the curvature of the regression polynomial is a function of the optimum ignition timing advance and the richness Ri. Depending on the variable P
π _AV = 1-P × a _{(AV-AV opt)} ^2.
[0009]
Thus, a sufficiently simple mathematical formula is used to make it actually usable by the engine control computer.
[0010]
In one embodiment, the richness coefficient π _Ri is a variable that depends on the richness Ri, and is equal to 1 when the richness Ri is equal to 1.
[0011]
The change in PMI + that changes with the ignition timing progress of the positive loop is modeled by a quadratic curve close to the actual measurement point.
[0012]
Preferably, at a measurement point that is closer to the value of the optimum ignition timing advance,
A (i) = A _min + (PMI + (i) -PMI + _min ) × (A _max -A _min ) / (PMI + _max -PMI + _min )
Here, A _min is the minimum weight associated with the measured minimum value of the indicated mean pressure PMI +, A _max is the weight of the weight A according to the maximum weight associated with the measured maximum value of the indicated mean pressure PMI +,
Minimization criteria:
min [Σ _i A (i) ² × (PMI + (i) mes-PMI + (i) calc) ² ]
Where i is the number of the measurement point on one curve, PMI + (i) mes is the indicated mean pressure PMI + measured at point i, and PMI + (i) calc is calculated by mathematical approximation at point i Illustrated mean pressure PMI +,
Run using
[0013]
In fact, to represent the high performance of an ignition control engine, an estimate of the indicated mean pressure of the positive loop of the pressure cycle in the cylinder, which varies with the volume of the cylinder, is used. This estimate does not consider the recoverable potential work during combustion in the cylinders, but is unique to each engine, the friction and distribution of moving equipment characterized by the friction mean pressure PMF, and the intake air as well. Or it is possible to release from some constraints, such as the negative loop of the pressure cycle (= f (volume)) corresponding to the operation of the engine when the exhaust valve is open.
[0014]
The invention will be better understood by reading the detailed description of the embodiments. Embodiment shown here is clarified with an accompanying drawing as a non-limiting example.
[0015]
FIG. 1 shows a graph showing pressure varying with cylinder volume.
[0016]
FIG. 2 is a graph showing a change in the indicated average pressure of the positive loop depending on the ignition timing progress.
[0017]
FIG. 3 is a graph similar to FIG. 2, which is obtained with and without weighting.
[0018]
FIG. 4 is a diagram illustrating the operation of obtaining the torque value as a function of the engine speed, the air flow rate, the collector pressure, and the ignition timing progress.
[0019]
FIG. 5 is a diagram showing the operation of the torque block shown in FIG.
[0020]
FIG. 6 is a diagram illustrating an operation for obtaining the ignition timing progress efficiency.
[0021]
As shown in FIG. 1, the change in the pressure governing the internal combustion engine cylinder of ignition control as a function of the cylinder volume is positive, corresponding to the operation of the engine when the intake and exhaust valves are closed. It is divided into two main loops: a loop and a negative loop corresponding to the operation of the engine in the exhaust stroke and the intake stroke where the intake or exhaust valve is open.
[0022]
Accordingly, the work recovered in the form of mechanical energy is equal to the indicated mean pressure PMI + for the positive loop minus the indicated mean pressure PMI- for the negative loop and the friction mean pressure PMF.
[0023]
The indicated mean pressure PMI + of the positive loop is the indicated mean pressure PMI + when the ignition timing advance AV is the optimal ignition timing advance AV _opt and the richness Ri representing the ratio of the stoichiometric air-fuel ratio to the air-fuel ratio during operation is 1. _{opt, Ri = 1} , richness
Ri is a variable representing the effect on the indicated mean pressure to richness coefficient [pi _Ri and the ignition timing progress AV is a variable representing the effect that there is indicated mean pressure away from 1 away from the optimal ignition timing progress AV _o ignition It can be estimated by multiplying the timing advance efficiency π _AV . That is,
PMI ₊ = PMI + _opt, is _{Ri = 1 × π Ri × π} AV.
[0024]
When Ritchine scan Ri is equal to 1, a richness coefficient [pi _{Ri = 1.} When the richness Ri is 1 and the ignition timing advance AV is the optimum ignition timing advance AV _o , the indicated average pressure of the engine is the flow rate of air flowing into the engine per stroke (1/2 rotation of the 4-cylinder crankshaft). and Q _r, equal to the product of the coefficient K which depends simultaneously on the pressure P _col the rotational speed N of the collector. That is,
PMI + _{opt, Ri = 1} = K (P _col , N) × Q _r .
[0025]
The coefficient K is a drawing method value representing the combustion efficiency of the engine when the richness Ri is 1. The ignition timing advance efficiency π _AV characterizes the change in the indicated mean pressure PMI + of the positive loop as a function of the ignition timing progress. This change follows a regression function of a quadratic function with respect to the distance between the optimum ignition timing advance AV _opt of the operating point and the applied ignition timing advance AV, and the curvature of the regression polynomial is a function of the optimum ignition timing advance AV _opt and Richness Ri. . That is,
_{π AV = 1-f (AV} opt) × h (Ri) × - is (AV AV _opt) ^2.
[0026]
The term AV _opt is a coefficient determined by the plotting method of the optimum ignition timing advance AV _{opt that} varies with the collector pressure P _col and the operating speed N. This plotting method refers to the optimal ignition timing progress as a function of the collector pressure P _col and the operating speed N, and plots when Richness Ri is equal to 1 and when Richness Ri is less than 1. Richness between two construction methods, with construction methods
Obtained from the result of nonlinear interpolation that varies with Ri. That is,
AV _opt (P _col , N, Ri) = interp (AV _opt (Ri = 1), AV _opt (Ri <1), Ri) .
[0027]
As shown in FIGS. 2 and 3, the mathematical expression used to estimate the indicated mean pressure PMI + of the positive loop is related to the determination of the optimal ignition timing advance AV _opt for a wide range of engine operating points. It is desirable to be able to identify all these optimal ignition timing advancements in an accurate manner. These determinations are performed primarily in the flatter areas of the curve, which, from its shape, are called hat-like ignition timing advancements. Using a quadratic curve, the change in the indicated mean pressure PMI + of the positive loop is modeled as a function of the ignition timing progress. The modeling equation is obtained by minimizing the distance between the measurement point and the mathematical curve. This type of minimization results in an inaccurate estimation of the optimal ignition timing advance point, even if a properly calculated PMI + value is kept close to the measured value.
[0028]
In the example of FIG. 2, the difference between the measuring point and the curve is evident, can not be used to seek the value of the optimum ignition timing progress AV _o from this curve. Due to the accuracy requirements required to measure the optimal ignition timing advance AV _opt, the operating point close to the experimental optimal ignition timing advance needs to be weighted to give greater importance than the operating point far from the optimal ignition timing advance Become.
[0029]
Therefore, an approximation is performed for the weighted measurement values that change according to the distance that the ignition timing progress is away from the vicinity of the optimal ignition timing progress. The curve thus obtained (FIG. 3) passes closer to the set of measurement points of the cap ignition timing progress according to the weighting minimization criterion. The criteria for this weighting are as follows. That is, A _min as the minimum weight associated with the measured minimum value PMI + _min of the indicated average pressure of the positive loop and A _max as the maximum weight associated with the measured maximum value PMI + _max of the illustrated average pressure of the positive loop. Define. The weight varies linearly as a function of the value of the measured value PMI + measure. That is,
_{A (i) = A min +} (PMI + (i) -PMI + min) × (A max -A min) / (PMI + max -PMI + min).
[0030]
The distance of the weighted measurement value is minimized by the following mathematical approximation.
[0031]
PMI + = PMI + _opt × [1-C × (AV- _AVopt ) ² ]
Here, C is the curvature of the quadratic curve.
[0032]
The relevant minimization criteria are thus written as follows:
[0033]
min [ΣA (i) ² × (PMI + (i) mes-PMI + (i) calc) ² ]
Where i is the number of the measurement point during one cap-like ignition timing advancement, PMI + (i) mes is the PMI + measured at point i, and PMI + (i) calc is the mathematical at the point i PMI + calculated by approximation.
[0034]
For each cap-like ignition timing progression thus calculated, a unique curve that meets the minimization criteria can be passed. This curve shows each operating point (richness
Ri , number of revolutions, collector pressure).
[0035]
PMI + = K × _Qr × [1-C × (AV-AV _opt ) ² ]
here,
PMI + _opt = K × Q _r .
[0036]
As shown in FIG. 3, by using a weighted method, a better approximation of the measurement points is possible.
[0037]
Calibration of the overall parameters of the model based on the test results on the engine platform is performed by minimizing the distance between the model and the measurement. The following are calibrated for optimization:
[0038]
-G (Ri): Richness Ri efficiency table,
−f (AV _opt ): a table of modeled cap timing curve and a function of optimal timing timing,
-H (Ri): Effect of richness Ri on the modeled cap ignition timing curve.
[0039]
As an overall criterion for minimization, we introduce a subordinate criterion that is applied simultaneously to the deviation between the curve, the optimal ignition timing advance and the value of the indicated mean pressure PMI + of the positive loop.
[0040]
For each sub-minimization criterion, a weight and accuracy target is entered that goes into the overall determination of the deviation between the model value and the measured value. The standard for minimizing the absolute value of the difference between the measured value and the model value is written as follows. That is,
CRIT (PMI +) = min [a × Σabs (PMI + (i) mes-PMI + (i) calc)]
The accuracy target can be set at 0.3 bar, for example.
[0041]
The standard for minimizing the difference between the model curve and the measurement is written as follows: That is,
CRIT (courbe) = min [b × Σabs (1-courbre (k) calc / courbre (k) mes)]
Here, k is the number of the cap-shaped portion of the curve.
The accuracy goal can be taken, for example, for an error of 20% of the curve.
[0042]
The standard for minimizing the difference in the optimal ignition timing progress between the model and the measured value is written as follows. That is,
CRIT (AV _opt ) = min [c × Σabs (AV _opt (k) mes-AV _opt (k) calc)]
For example, the accuracy target can be an error of 1 ° of the estimated value of the optimum ignition timing progress. The resulting overall criterion for minimization is the algebraic sum of the subordinate criteria defined above. Each minimization criterion is assigned a target multiplication weight (coefficients a, b, c) with a desired accuracy. In consideration of the accuracy target given in the present embodiment, the following proportional conversion criterion is imposed. That is,
a × 0.3 = b × 0.2 = c × 1.
[0043]
As shown in FIG. 4, the torque value is obtained from the average working pressure PME in the engine. The value of the average working pressure PME is obtained from the exhaust pressure P _echpp , the output of the torque block whose function is shown in FIG. The pressure in the exhaust is calculated from the air flow rate Q _r entering the engine to be displayed in kg / sec.
[0044]
As shown in FIG. 5, the torque block causes the pressure P _col in the collector, the indicated average pressure PMI + of the positive loop, the ignition timing advance efficiency π _AV , the richness coefficient π _Ri, and the average operating pressure PMF of the engine. A value can be obtained. As described above, the value of PMI + takes into account weighting, and the air flow rate Q _r flowing into the engine _expressed in grams per revolution, the pressure P _col in the collector, and the operating rotation at richness Ri 1 It is obtained from the drawing method of the coefficient K as a function of the number N, the ignition timing advance efficiency π _AV, and the richness coefficient π _Ri . The method for obtaining the ignition timing advance efficiency π _AV is shown in FIG. A curve representing the richness coefficient π _Ri is obtained from the value of the richness Ri. The average operating pressure PMF is calculated from the operating speed N. The amount of air Q _r displayed in grams per revolution is calculated from the amount of air flowing into the engine _expressed in kg / sec and the operating speed N.
[0045]
As shown in FIG. 6, the ignition timing advance efficiency π _AV is determined by the influence h (Ri) of the richness Ri on the modeled cap-shaped ignition timing progress curve, the ignition timing progress AV, and the modeled cap-like shape. It is obtained from the table f (AV _opt ) of the ignition timing advance curve and the optimum ignition timing advance AV _opt (Ri). The function h (Ri) is obtained from the value of the richness Ri. And modeled galea ignition timing progress curve table f (AV _opt), the optimum ignition timing progress AV _{opt (Ri)} is the case richness Ri is 1, richness Ri is 1 or less, for example 0.68 Starting from the plotting method set for the case, it is calculated as a function of the operating speed N and the pressure _Pcol in the collector and the variable P depending on the richness Ri.
[0046]
The present invention allows the use of a highly accurate estimation method of engine torque that allows improved control of the engine as expressed by improved efficiency and reduced energy consumption.

Claims (4)

The indicated mean pressure of the positive loop corresponding to the operation of the engine when the intake valve and the exhaust valve of the pressure cycle in the cylinder changing with the volume of the cylinder are closed, and the air flow rate flowing into the engine, Richness representing the pressure in the collector of the engine, the operating speed N, the ignition timing advance, and the ratio of the stoichiometric air-fuel ratio to the operating air-fuel ratio
An illustrated mean pressure calculation method for an internal combustion engine by calculating as a function of Ri ,
The indicated mean pressure PMI + of the positive loop is
PMI + = PMI + _{opt, Ri = 1} × π _Ri × π _AV ,
Here, PMI + _{opt, Ri = 1} is the indicated mean pressure when the ignition timing advance is the optimum ignition timing advance and the richness Ri is 1,
π _Ri is a richness coefficient that is a variable that represents the influence of the richness Ri away from 1 on the indicated mean pressure,
π _AV is the ignition timing advance efficiency, which is a variable representing the influence of the ignition timing advance away from the optimal ignition timing advance on the indicated mean pressure,
In the case where the ignition timing advance is the optimum ignition timing advance and the richness Ri is 1, the indicated average pressure PMI + _{opt, Ri = 1} is
_{PMI + opt, Ri = 1 =} K (P col, N) is a × _{Q r,}
Where K is a coefficient that depends on the pressure P _col in the collector and the operating speed N at the same time,
Q _r is the flow rate of air flowing into the engine per stroke,
Der is,
The ignition timing advance efficiency π _AV follows a regression function of a quadratic function with respect to the distance between the optimum ignition timing advance AV _opt of the operating point and the applied ignition timing advance AV, and the curvature of the regression polynomial is the optimal ignition timing advance and the richness. Depends on the variable P, which is a function with Ri,
π _AV = 1−P × (AV-AV _opt ) ²
Indicated mean pressure calculation method for an internal combustion engine, characterized in that it. The method for calculating an indicated mean pressure of an internal combustion engine according to claim 1, wherein the richness coefficient π _Ri is a variable depending on the richness Ri, and is equal to 1 when the richness Ri is equal to 1. 2. The indicated average pressure of the internal combustion engine according to claim 1, wherein the change in the indicated mean pressure PMI + that changes with the ignition timing progress of the positive loop is modeled by a quadratic curve close to the actual measurement point. Method of calculation. At the measurement point that is recognized as more important at the measurement point closer to the optimal ignition timing progress of the experiment, the following formula:
  A (i) = A _min + (PMI + (i) -PMI + _min ) × (A _max -A _min ) / (PMI + _max -PMI + _min )
  Where A _min Is the minimum weight associated with the measured minimum value of PMI +, A _max Is the maximum weight associated with the measured maximum value of PMI +,
  Weight A according to
  Minimization criteria:
  min [Σ _i A (i) ² × (PMI + (i) mes-PMI + (i) calc) ² ]
  Where i is the number of the measurement point on one curve, PMI + (i) mes is the PMI + measured at point i, and PMI + (i) calc is the PMI + calculated by mathematical approximation at point i. ,
  The method for calculating the indicated mean pressure of the internal combustion engine according to claim 3, wherein

Images