WO1991013248A1

WO1991013248A1 - Engine condition determining and operating method

Info

Publication number: WO1991013248A1
Application number: PCT/US1991/001392
Authority: WO
Inventors: Donald J. Remboski; Steven L. Plee; Jialin Yang; Robert W. Law; Michael T. Vincent
Original assignee: Barrack Technology Limited
Priority date: 1990-02-26
Filing date: 1991-02-25
Publication date: 1991-09-05
Also published as: AU7464991A

Abstract

An internal combustion engine (11) having a luminosity detector (18) in a combustion chamber (16) and an arrangement for measuring certain parameters and running conditions of the engine (11) such as peak heat release rate in the combustion chamber (16), NOx emissions and air/fuel ratio based on particular gain independent parameters of the luminosity signal. In addition, these gain independent luminosity parameters can be used in a control loop to adjust the various parameters of the engine to obtain the desired luminosity characteristics and uniform combustion conditions from cycle to cycle in a given combustion chamber and in the combustion chambers of a multi chamber engine.

Description

"ENGINE CONDITION DETERMINING AND OPERATING METHOD"

BACKGROUND OF THE INVENTION

This invention relates to a method and apparatus for operating an engine, and more particularly to an improved method and apparatus for operating an engine in response to actual conditions sensed in the combustion chamber during each combustion cycle or on average, and for determining certain operating parameters and running conditions of the engine.

This invention also relates to a method and apparatus for determining certain operating and running parameters such as the time in the combustion cycle at which peak rate of heat release occurs in the combustion chamber and for determining N0_χ emissions based on the detected luminosity in the combustion chamber of an engine. This information can be then be used for operating and controlling the engine.

With modern technology and electronics, many of the components and running conditions of an internal combustion engine can be controlled more accurately than with previous mechanical systems. For example, the control of the air/fuel ratio, spark timing, fuel injection timing and pulse, and other adjustable factors of engine operation are greatly facilitated through the use of electronic components and electronic computers. However, in order to accurately sense the running of the engine and the various phenomena occurring within the combustion chamber, it is necessary to provide a sensor that is directly positioned within the combustion chamber or in proximity to it and which senses the actual combustion conditions or process in the engine. Most engine controls employ external devices such as oxygen sensors or knock sensors which actually sense only average conditions due to their inherent nature. It has been understood that knocking can be determined by an optical sensor that operates within the combustion chamber and which senses the luminosity of the gases in that chamber. A wide variety of knock detectors have been proposed that employ such sensors. However, the inventors have discovered that detected luminosity in the combustion chamber and, in particular, various parameters of the luminosity signal or curve generated from this detected luminosity can indicate a much wider range of engine operating and running conditions than previously realized. For example, these luminosity curve parameters can be used to predict the time at which the peak rate of heat release occurs in the combustion cycle and to predict N0_χ emissions from an internal combustion engine. These particular predictions were developed using a large bore, lean burn, highly boosted natural gas engine.

It is, therefore, an object of this invention to provide an improved apparatus and method for operating an engine wherein a luminosity detector and particular gain independent parameters of its luminosity signal or curve are used to determine certain combustion conditions occurring in the combustion chamber, including start and end of combustion, and to determine certain operating parameters and engine running conditions.

It is a further object of this invention to provide an improved system and method for operating an engine wherein the engine's adjustable parameters such as air/fuel ratio, spark timing, fuel injection, etc. can be varied in response to luminosity curve parameters and to actual sensed conditions in the combustion chamber so as to provide better running of the engine.

It is a still further object of this invention to provide engine control systems wherein the engine can be controlled in response to the luminosity signal. The control system should be predicated on certain measured parameters of the engine. For example, it is very desirable to be able to obtain and measure such engine parameters and running characteristics as peak cylinder pressure in relation to output shaft or crank angle, air/fuel ratio, indicated mean effective pressure (IMEP)

(which is in effect the same as measuring engine torque or power) , N0_χ emissions and the gas temperature at exhaust valve opening.

It is yet another object of this invention to provide an improved apparatus and method for operating an engine, wherein a luminosity detector and particular parameters of its luminosity signal or curve are used to determine when peak rate of heat release occurs in the combustion chamber and to determine N0_χ emissions from the engine. These predicted values can be determined during each cycle of operation, or an average value can be determined over a period of cycles.

It is a further object of this invention to provide an improved system and method for operating and controlling an engine, wherein the engine's adjustable parameters such as air/fuel ratio, spark timing, fuel injection, etc. , can be varied in response to the parameters of the luminosity signal or in response to certain predicted operating and running parameters of the engine such as the time in the combustion cycle at which peak heat release rate occurs, air/fuel ratio or N0_χ emissions so as to provide better running of the engine and/or reduce cycle to cycle variation.

A type of engine sensor has been proposed that senses the actual luminosity of the gases within the combustion chamber. A wide variety of patents illustrating and describing the use of such sensors have issued including the following:

4,358,952; 4,369,748; 4,377,086; 4,393,687;

4,409,815; 4,412,446; 4,413,509; 4,419,212; 4,422,321; 4,422,323; 4,425,788; 4,468,949;

4,444,043; 4,515,132.

For the most part, these patents disclose arrangements wherein the sensor is utilized to sense only total luminosity and to equate the luminosity signal to a knocking signal.

However, as previously noted, the. inventors have discovered that this luminosity signal or curve and, in particular, various gain independent parameters of the luminosity signal can also be employed to determine ^'particular phenomena occurring in the combustion chamber which then provides a basis for adjusting various engine parameters and running characteristics. These luminosity parameters can also provide an indication of various engine parameters and running characteristics. Thus, these gain indepenent luminosity parameters can be used to control the engine parameters to obtain better running and to obtain consistent running from cylinder to cylinder and cycle to cycle.

SUMMARY OF THE INVENTION

This invention is adapted to be embodied in a method for operating an internal combustion engine and an apparatus therefor that has a combustion chamber and means for forming a combustible air/fuel mixture within the combustion chamber.

In accordance with embodiments of the invention, the luminosity of the gases in the combustion chamber are- sensed or detected, a curve based on the detected luminosity is generated, the characteristic of at least one gain independent parameter of the luminosity curve is determined such as location of peak luminosity or its derivative, preferably in relation to output shaft or crank angle, and at least one parameter of the engine is adjusted to obtain a desired characteristic for that particular gain independent luminosity parameter at at least one point on the curve, preferably in relation to output shaft or crank angle. Rather than, or in addition to, adjusting at least one parameter of the engine, another feature of the invention involves measuring at least one engine parameter such as indicated mean effective pressure or air/fuel ratio based on the determined characteristic of at least one gain independent parameter.

Another feature of the invention is also adapted to be embodied in a method for operating an internal combustion engine and an apparatus therefor having a combustion chamber and means for causing combustion to occur in the combustion chamber. In accordance with this feature of the invention, the luminosity of the gases in the combustion chamber are detected during each combustion cycle and the engine is adjusted to minimize cyclic variations.

In accordance with another feature, the luminosity of the gases in the combustion chamber are detected, a curve is generated based on the detected luminosity, the characteristic of at least one gain independent parameter of the luminosity curve is determined, preferably in relation to output shaft or crank angle, and an engine parameter is measured and adjusted to attain a desired relationship between the characteristic of the particular gain independent parameter and output shaft angle.

Yet another feature of the invention is adapted to be embodied in a method and apparatus for operating a multi- combustion chamber internal combustion engine that includes means for effecting combustion in each of the combustion chambers. In accordance with this feature of the invention, the luminosity of the gases are detected in each of the combustion chambers and adjustment is made to the engine so as to minimize chamber to chamber variations for that particular combustion condition.

In accordance with a further feature of the invention, the luminosity of gases in the combustion chamber are sensed or detected, a curve is generated based on the detected luminosity, the location, preferably in relation to crank angle, of a gain independent parameter on the luminosity curve is determined, such as location of peak luminosity, and the time in the combustion cycle at which peak heat release rate occurs is predicted, preferably in relation to crank angle, based on the determined location of the gain independent parameter. The locations of peak luminosity and peak heat release rate can also be used to estimate air/fuel ratio and N0_χ emissions.

A still further feature of the invention is also adapted to be embodied in a method for operating an internal combustion engine and an apparatus therefor having at least one combustion chamber and means for forming a combustible air/fuel mixture within the combustion chamber. In accordance with this feature of the invention, the luminosity of the gases- in the combustion chamber are detected, a curve is generated based on the detected luminosity, at least one parameter of the luminosity curve is determined, and N0_χ emissions are predicted based on the determination of at least one parameter.

BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1 illustrates a cross-sectional view taken through a single combustion chamber of a multi-cylinder internal combustion engine constructed and operated in accordance with embodiments of the invention.

Figure 2 shows comparison of luminosity, pressure and heat release rate as a function of crank angle.

Figure 3 shows the correlation between start of combustion (SOC) in crank angle degrees and the location at which the luminosity is 5% of the peak luminosity (ca 5% L/L_max) in crank angle degrees.

Figure 4 shows the correlation between start of combustion (SOC) in crank angle degrees and the location at which the luminosity is 5% of the total integrated luminosity (ca 5% iL) in crank angle degrees.

Figure 5 depicts the correlation between end of combustion (EOC) in crank angle degrees and the location of the minimum luminosity derivative (cadL^) in crank angle degrees.

Figure 6 illustrates the correlation between the location of peak rate of heat release (cadQ.^) in crank angle degrees and the location of the peak luminosity derivative cadL^) in crank angle degrees. Figure 7 shows the correlation between the location of peak pressure in the combustion chamber (caP^) in crank angle degrees and the location of the peak luminosity (caL^) in crank angle degrees.

Figure 8 shows a diagram of a spark timing control loop using the luminosity signal.

Figures 9 and 17 show the correlation between observed and predicted location of peak pressure in the combustion cha ber (caP^) in crank angle degrees using gain independent luminosity parameters and engine parameters.

Figures 10 and 18 show the correlation between observed and predicted indicated mean effective pressure (IMEP) in kPa using gain independent luminosity parameters and engine parameters.

Figures 11 and 19 illustrate the correlation between observed and predicted air/fuel ratio using gain independent luminosity parameters and engine parameters. Figures 12 and 20 depict the correlation between observed and predicted NO_χ exhaust emission using gain independent luminosity parameters and engine parameters.

Figures 13 and 21 show the correlation between observed and predicted gas temperature at exhaust valve opening using gain independent luminosity parameters and engine parameters.

Figures 14 and 22 illustrate the correlation between observed and predicted start of combustion in crank angle degrees using gain independent luminosity parameters and engine parameters.

Figures 15 and 23 depict the correlation between observed and predicted end of combustion in crank angle degrees using gain independent luminosity parameters and engine parameters. Figure 16 shows a diagram of an air/fuel ratio control loop using the luminosity signal and a zirconia sensor.

Figure 24 illustrates a cross sectional view taken through a single combustion chamber of a multi-cylinder internal combustion engine constructed and operated in accordance with embodiments of the invention. This figure also illustrates a schematic diagram of the processing of the luminosity signal in accordance with embodiments of the invention.

Figure 25 shows a 10mm threaded optic probe for use with embodiments of this invention.

Figure 26 shows a spark plug mounted probe for use with embodiments of this invention. Figure 27 is a circuit diagram of a transimpedance amplifier for use with embodiments of this invention.

Figure 28 shows a filtered mean radiation trace or luminosity curve measured in volts plotted against crank angle and its gradient or derivative measured in volts/degrees, also plotted against crank angle.

Figure 29 shows 100 cycle ensemble average luminosity (radiation) , pressure and heat release rate curves.

Figure 30 depicts the relationship between the location of minimum luminosity derivative (location of minimum radiation gradient) and location of peak heat release rate both measured in crank angle degrees for different engine speeds.

Figure 31 shows the relationship between location of peak luminosity or radiation and location of peak heat release rate both measured in crank angle degrees for different air/fuel ratios.

Figure 32 shows a comparison of peak luminosity (maximum detector output) and location of peak luminosity or maximum output in crank angle degrees versus brake specific N0_χ.

Figure 33 shows the location of peak heat release rate in crank angle degrees versus brake specific NO_χ.

Figure 34 shows the correlation between measured and predicted brake specific N0_χ using intake pressure, air/fuel ratio and location of minimum luminosity derivative to determine the predicted values.

Figure 35 shows the relationship between the location of peak luminosity and location of peak heat release rate both measured in crank angle degrees for different brake specific NO_χ.

Figure 36 shows the correlation between measured and predicted brake specific N0_χ using intake pressure, location of peak luminosity and location of minimum luminosity derivative to determine the predicted values.

Figure 37 shows the correlation between measured and predicted brake specific N0_χ using various luminosity parameters to determine the predicted values. Figure 38 shows individual cycle N0_χ estimates for 100 consecutive cycles.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

Referring to Figures 1 and 24 of the drawings, a multi-cylinder internal combustion engine is identified generally by the reference numeral 11. It is to be understood that, although the invention has particular utility in multi-cylinder engines, certain facets of the invention may find application in single cylinder engines as well. Also, although the invention is described in conjunction with a reciprocating type engine, the principles of the engine may be utilized with engines of the non-reciprocating type, such as rotary engines, and with engines operating on either two stroke or four stroke cycles.

Inasmuch as the invention is concerned primarily with the combustion chamber and the conditions therein, only a cross-sectional view taken through one of the combustion chambers is believed to be necessary to understand the invention and the environment in which it can be practiced. This cross-sectional view shows a cylinder block 12 having a cylinder bore 13 in which a piston 14 is supported for reciprocation. The piston 14 is connected by means of a connecting rod (not shown) to a crankshaft for providing output power from the engine 11.

A cylinder head 15 is affixed in a known manner to the cylinder block 12 and has a recess 16 which cooperates with the cylinder bore 13 and head of the piston 14 to provide a chamber of variable volume, sometimes referred to hereinafter as the combustion chamber.

An intake port 17 and an exhaust port (not shown) extend through the cylinder head 15 and have their communication with the combustion chamber 16 controlled by a poppet type intake valve, identified by the reference numeral 21 in Figure 24, and a poppet type exhaust valve (not shown) for admitting a charge to the combustion chamber 16 and for discharging the burnt charge from the combustion chamber 16. It is to be understood, of course, that the combustion chamber 16 may have a plurality of intake and exhaust valves, or may employ ports in lieu of valves and that the engine 11 may include a plurality of combustion chambers 16.

The charge admitted to the combustion chamber 16 may comprise pure air or an air/fuel mixture that is formed by a suitable charge former such as a port or throttle body type fuel injector, or carburetor. Alternatively, if pure air is delivered or injected, direct cylinder or manifold injection may be employed for delivering or injecting fuel into the combustion chamber 16 to form the air/fuel mixture. The air/fuel ratio may be controlled in a wide variety of known manners such as by means of throttle valves, fuel control valves, injector pulse width, injection duration, injection timing, injection pulse, etc. Although an important feature of the invention is the parameters under which the air/fuel ratio is controlled, the actual physical hardware for adjusting the air/fuel ratio is not part of the invention. However, in accordance with the invention, these engine parameters may be adjusted manually or automatically as part of an engine control loop so as to obtain the desired combustion characteristics or air/fuel ratio, which can be a function of engine speed and/or engine load.

The engine 11 is preferably of the spark ignited type. However, the types of controls exercised and the nature of luminosity detecting or sensing may vary with different engines. In an engine 11 of the spark ignited type, a spark plug identified by the reference numeral 22 in Figure 24 will be carried in the cylinder head 15 and have its gap exposed in the combustion chamber 16. The spark timing is controlled by a suitable mechanism which may be of any conventional type; however, the timing of the spark firing can be varied in accordance with parameters as hereinafter described.

As has been previously noted, the invention is capable of embodiment in any of a wide variety of conventional types of internal combustion engines and, for that reason, the details of the engine construction are not necessary to understand how the invention can be practiced by those skilled in the art. However, in accordance with the invention there is provided in the combustion chamber 16, a luminosity detector, indicated generally by the reference numeral 18. The luminosity detector 18 includes a fiber optic probe 19 or other type of optical access which extends through the cylinder head 15 and has its end terminating at or within the combustion chamber 16. The detector 18 and fiber optic probe 19 is preferably of the type described in the application entitled "Luminosity Detector", Serial Number 284,193, filed December 14, 1988 and in the continuation- in-part application of the same title, Serial Number 467,883, filed January 22, 1990, both in the names of Donald J. Remboski et al and assigned to the assignee of this application, the disclosures of which are incorporated herein by reference. Other suitable detectors and optical probes may also be used.

Referring now to Figure 1, the probe 19 can be formed from a relatively inexpensive material such as synthetic sapphire (A1₂0₃) or other materials having similar characteristics. A probe having a diameter of 0.06" has been found to be practical and makes it relatively easy to install in the cylinder head 15.

The fiber optic probe 19 is held in place by means of a compression fitting 23 and has its outer end disposed within a light sealed housing 24 in proximity to a silicon photo detector 25.

Various luminosity spectra may be detected by the probe 19 or merely a total luminosity signal may be read. It has been found that certain constituents of the glowing gases in the combustion chamber 16 glow at different spectral ranges and this may be utilized to sense the amount and condition of such components in the combustion chamber 16 during each cycle of operation. Depending on the particular gas or gases to be detected or sensed, it may be desirable to provide a monochromator or an optical filter in front of the silicon photo detector 25 so as to select the desired wavelength of light which is being measured. For this application, the probe 19 is employed to measure the overall radiant emission from products of combustion (primarily H₂0) in the near infrared region between 850nm and lOOOnm. For the correlations shown in Figures 3-7, 9-15 and 17-23 a wavelength band centered at 927.7nm +/-20nm was used. The near infrared region is monitored because it is not strongly influenced by radiant emission from the walls of the combustion chamber 16, nor is it sensitive to emission from the flame surface. In addition, this particular wavelength band coincides with the peak spectral response of the silicon photo detector 25.

The detector 25 is connected to a remotely positioned computer control unit by means of conductors 20. In addition to converting the voltage signal from the silicon photo detector 25 to an output indicative of luminosity, the remotely positioned control unit which may measure certain engine parameters as well, may also receive input signals from other sensors normally employed on the engine, for example, air/fuel ratio, intake manifold pressure and temperature, engine speed, and spark timing angle sensors. These types of sensors are normally employed with modern internal combustion engines and their signals can be processed in conjunction with the luminosity signal to provide certain measured characteristics of the engine operation. The remotely positioned control unit may be of any suitable type and is particularly adapted to transmit the signal from the detector 25 into an output indicative of luminosity within the combustion chamber 16. A typical luminosity signal or curve as a function of crank angle is shown in Figure 2. The measurement of the luminosity curve and the location of its various gain independent parameters in relation to crank angle assumes a reciprocating type engine. However, in non-reciprocating engines these measurements can be done in relation to output shaft angle. Accordingly, output shaft angle is used in the claims and is intended to refer to crank angle as well.

It has been discovered that a wide variety of combustion conditions and other engine operating and running characteristics can be determined by the luminosity detector 18 and the luminosity curve generated as a result of the luminosity detected in the combustion chamber 16 employed to adjust the parameters and running conditions of the engine to obtain optimum performance. In addition, this luminosity curve has various parameters which are gain independent, and characteristics of these parameters have been found to provide very good indications of certain combustion conditions and engine operating and running conditions.

Gain independent luminosity parameters, as the name implies, are not affected by variations in the gain of the system, as are gain dependent parameters. An example of a parameter which is gain dependent is peak luminosity (L^) . The advantage associated with using gain independent parameters to determine various combustion and engine conditions is that the luminosity measurements are not affected when the gain of the system decreases over time due to probe deposits, etc. The inventors have discovered that the use of gain independent luminosity parameters to correlate various combustion characteristics and engine conditions avoids the problem of a diminishing luminosity signal over time, since gain independent correlations are not affected by variations in the gain of the system. An example of a parameter that is gain independent is the location of peak luminosity (caL_maχ) . As long as the luminosity signal maintains a sufficient signal to noise ratio, the location of peak luminosity in crank angle degrees will remain the same regardless of gain changes. In other words, the inventors are using the shape of the luminosity curve rather than the amplitude to correlate their data. Location of peak luminosity can be used as a combustion phasing signal to control spark timing for each cylinder on a cycle-to-cycle basis under a wide variety of engine operating and running conditions. This luminosity parameter of a generated luminosity curve can be used to continually adjust spark timing as part of an engine control loop as shown in Figure 8 in order to obtain the proper phasing of the luminosity curve profile. This can be accomplished by comparing the determined caL^ from the detected luminosity with a desired caL^ which can be a function of engine speed. The spark timing is then adjusted to obtain the desired caL^. Knock can also be detected from the luminosity signal and is detected by a high frequency signature on the luminosity signal at detonation. This characteristic of the luminosity signal can be employed in the control loop to retard spark timing until knock is eliminated. In this case, caL.^ is used to adjust spark timing under no knock conditions.

Data for the gain independent correlations illustrated in Figures 3-7, 9-15 and 71-23 was obtained using a single cylinder of a 2.2 liter, 4 cylinder port fuel injected, spark ignited, automotive engine of the side flow type having a compression ratio of 8.9, a bore of 87.7mm, a stroke of 92.0mm and two valves per cylinder. The data was also obtained under various engine speeds (i.e., 750, 1500, and 2400 rpm) , intake manifold pressures (i.e., 70kPa, 85kPa, wide open throttle), air/fuel ratios (i.e., 13.0, 14.6, 16.5, 18.0), percent EGR (i.e., 0, 5, 10, 15) and spark timing (i.e., minimum advance for best torque [MBT] - 10°, MBT, MBT +10°).

The inventors have found that the characteristics of certain gain independent parameters of the luminosity curve can be used to give a good indication of start and end of combustion and combustion duration. As shown in Figure 3, start of combustion (SOC) can be correlated to the gain independent luminosity parameter (ca 5% L/L-^..) which is the location in crank angle degrees at which the luminosity is 5% of the peak luminosity. Figure 4 shows that start of combustion can also be correlated to (ca 5%iL) which is defined as the location in crank angle degrees at which the luminosity is 5% of the total integrated luminosity. 5% of peak and total integrated luminosity is exemplary only and other percentages early in the luminosity curve can also be used for example, 1-10%.

In spark ignited, direct injected engines strong charge stratification may lead to diffusion burning and soot production. In these cases, the in-cylinder luminosity measurement is a good indicator of the presence of soot in a diffusion burn and the start of the diffusion burning. Thus, the luminosity signal may also be used to control diffusion burning in spark ignited direct fuel injection engines. By determining the degree of which diffusion burning exists, various engine parameters may be adjusted to affect a change in the burning process. For example, injection pulse width, injection timing, injection pressure, rate of fuel injection and spark timing may all be modified based on the relative degree of diffusion burning that is to be avoided, or the level of pre-mixed burning desired.

For a "normal burn" that occurs near top dead center, the inventors have also determined that the end of combustion (EOC) can be correlated to (cadL_mn) which is defined as the location of the minimum luminosity derivative in crank angle degrees. This correlation is shown in Figure 5. That is, the end of combustion can be defined by determining the location of the minimum rate of change of luminosity in crank angle degrees. A late in the cycle can indicate incomplete burns, high CO emission and/or high hydrocarbon emission. Various engine parameters such as spark timing can be adjusted so that the cadl^^ occurs at the desired crank angle location so as to eliminate or minimize these undesirable effects or emissions.

Once the start and end of combustion are determined the duration of combustion in crank angle degrees can also be determined by subtracting the start of combustion measurement from the end of combustion measurement.

These gain independent luminosity parameters may then be employed to adjust or control engine parameters such as spark timing so as to control when combustion begins to obtain optimum performance characteristics. These adjustments can be made manually or automatically as part of an engine control loop to obtain better running of the engine. The control circuitry and mechanism for adjusting spark timing or other adjustable parameters of the engine in response to the luminosity parameters are believed to be within the scope of those skilled in the art once they understand that these gain independent luminosity parameters are indicative of the start and end of combustion, and also that these parameters can be used so as to control the start of combustion. Some of these luminosity parameters may also be utilized to determine the amount of fuel which has been or should be injected or delivered into the cylinder from a carburetor or port type injector so as to obtain the desired combustion duration or burn time.

The inventors have also determined that for a "normal burn" that occurs near top dead center, the location in crank angle degrees of the peak rate of heat release

(cadQ.^..) can be determined by monitoring the location of the peak luminosity derivative in crank angle degrees (cadL-^) as shown in Figure 6. The inventors have found that this relationship has good correlation under normal burn conditions and can be used to control burn phasing of the engine. Control of burn phasing can be accomplished by adjusting spark timing so that cadL^ occurs at the desired crank angle location.

The inventors have further determined that the location of peak luminosity in crank angle degrees (caL_max) can be related to the location of peak pressure in the combustion chamber in crank angle degrees (caP_m).) . This correlation is shown in Figure 7. As with the location of the peak heat release rate, location of peak pressure is a phasing parameter that can be used for spark timing control. Thus, in the 'normal burn" situation the spark timing can be adjusted based on the caL_maχ parameter so that parameter and the location of peak pressure occur at the desired crank angles. Various empirical correlations have also been developed by the inventors which relate the characteristics of various parameters of the luminosity curve to such engine operating parameters and running conditions as location of peak pressure (caP^) , indicated mean effective pressure (IMEP) , air/fuel ratio, NO_χ emissions and burned gas temperature at exhaust valve opening (T_evo) . Empirical correlations have been developed between luminosity parameters and start and end of combustion as well. These empirical correlations were established using a curve fitting routine which performs a multiple regression analysis using a quadratic response surface model for the function: Y = A + B^ + B₂X₂ + ... + B_nX_n + C. ₂ (X X₂) + C.₃ (X.X₃) + ... + D^ + D₂X₂ ² + ... + D_nX_n ². The routine calculates regression coefficients for linear, cross product and squared terms. The correlations involve a combination of gain independent luminosity parameters plus selected engine parameters such as engine speed in revolutions per minute (speed) , spark timing in crank angle degrees (spk) , intake manifold pressure in kPa (P_int) intake manifold temperature in degrees Celsius (T_int) , coolant temperature in degrees Celsius (TH₂0) . X₁ through X represent the various luminosity and engine parameters. The various A, B, C and D terms are curve fitting parameters generated in response to the luminosity and engine data set to generate a curve fit. Each data point in the following correlations shown in Figures 9-15 and 17- 23 represents a 100 cycle ensemble average. All correlations depicted in these figures and used to predict the various engine operating parameters and running conditions take the form of the above quadratic model. The inventors then tested their predicted values for the various engine operating parameters and running conditions by comparing those values with observed values. These comparisons using a full set of linear, cross product and squared terms in the quadratic model are shown in Figures 9-15. Comparisons using a reduced set of terms are shown in Figures 17-23. The inventors have found that the luminosity curve can be used to give an indication of the location of peak pressure (caP^) in crank angle degrees. The predicted value of caP_maχ has been found to be a function of speed,

^Pint ' ^{Ca] CadL}max ' V (^CadLmin ^{" S k}) ^{whe e CaL}max ^{is the} location in crank angle degrees of peak luminosity, cadL^ is the location in crank angle degrees of peak luminosity derivative, cadL_m]-_n - spk is the phase difference between the location of minimum luminosity derivative and spark timing in crank angle degrees. The relationship obtained from the data between the observed location of peak pressure determined by a pressure transducer and the predicted value based on these luminosity and engine parameters using the correlation routine are shown in Figure 9. The correlation coefficient between the predicted and observed location of peak pressure is 0.99 with a standard deviation of 0.16 degrees. Indicated Mean Effective Pressure (IMEP) can also be predicted using engine and luminosity parameters. The predicted IMEP has been found to be a function of speed,

^Pint ' ^{C L}max ' ^CadLmax ' ^CadLmin ' C cadL^ -S_Pk) Where CadL_mι-_n is the location of minimum luminosity derivative in crank angle degrees. The relationship between the observed IMEP determined by a pressure transducer and the predicted IMEP based on these luminosity and engine measurements and determinations is shown in Figure 10. The correlation between the predicted and observed IMEP is 0.99 with a standard deviation of 7.5 kPa.

This IMEP measurement based on the luminosity parameters can be used in managing engine torque as well as in cylinder balancing and traction control schemes. For example, the luminosity signal could be used to manage engine torque output during shifting of an automatic transmission. The luminosity parameters can also be used to give a good indication of the air/fuel ratio. The predicted air/fuel ratio is a function of speed, X , Y_cpN, P_int

^CaLmax ' ^CadLmax ' ^CadI ( cadL_mir. - Spk) . X_cp is the X- coordinate of the centroid of the luminosity signal in crank angle degrees and Y N is the normalized y-coordinate of the centroid of the luminosity signal against ^. The correlation between predicted air/fuel ratio using the regression analysis for these parameters and observed air/fuel ratio determined by exhaust emission measurement is 0.99 with a standard deviation of 0.16 as shown in Figure 11. The standard deviation of the correlation improves to 0.13 when EGR variations are omitted from the correlation. This luminosity measurement may also be able to provide an estimate of the air/fuel ratio of each cylin r of an engine on a cycle to cycle basis. This air/fuel ratio estimate could then be used in an engine control loop, as shown in Figure 16, to provide tighter control through transient and cold starts. The luminosity measurements could be incorporated in a control scheme along with a zirconia or oxygen sensor. The purpose of such a luminosity control loop is to provide an instantaneous response to a fuel injector control device when changes in the air/fuel ratio are observed. Longer term (ten cycles or more) control could be accomplished by the zirconia sensor which could be used to correct the luminosity loop. Logic could also be incorporated so that the primary signal for control of the air/fuel ratio during cold starts is from the luminosity measurement, thus yielding better control during this period than is now possible with the zirconia sensor inoperative.

The luminosity loop could be used to attenuate transient errors and thus reduce exhaust emissions. Individual cylinder resolution could also be used to provide a more uniformed control of air/fuel ratio to reduce overall emissions. A closed loop fuel control system during cold start and warm-up may also act to reduce emissions during this period.

Various gain independent luminosity parameters in connection with engine parameters can also be used to predict N0_χ exhaust emission. Predicted N0_χ exhaust emission has been shown to be a function of speed, P_int, ^caLmax - ^SP ' l/ (-c dI^_fn - S_Pk) , X_cp, Y_cpN , I./L^,

M_j/I^_gj., where caL^ - spk is the phase difference between location of peak luminosity and spark timing in crank angle degrees, X is the x-coordinate of the centroid of the luminosity signal in crank angle degrees, Y N is the normalized Y-coordinate of the centroid of the luminosity signal against l>_max, ./l^^ is the normalized first moment of the luminosity signal against L^, ^/L.^ is the normalized second moment about X of the luminosity signal against L^ and M₃/L_maχ is the normalized third moment about X of the luminosity signal against L^.

The correlation between the observed N0_χ emission measured from the exhaust and the predicted N0_χ emission based on these luminosity and engine parameters is shown in Figure 12. The correlation coefficient is 0.98 with a standard deviation of 154ppm. The correlation coefficient is 73ppm when only lean air/fuel ratios are considered.

Another engine condition which can be predicted based on luminosit ^'and engine parameters is gas temperature at exhaust valve opening. This predicted temperature is a function of the same engine and luminosity parameters as N0_χ emission. The correlation coefficient between observed temperature using a heat release model in connection with a pressure transducer and the predicted temperature is 0.99, as shown in Figure 13, with a standard deviation of 13°K.

High exhaust gas temperature is generally associated with wide open throttle or particularly high duty cycle operation. This luminosity based measurement of gas temperature at exhaust valve opening can be incorporated into a control loop to adjust the air/fuel ratio and/or throttle position to prevent premature failure of exhaust valves, head gasket and turbocharger and to reduce engine wear and aging.

Empirical models for predicting start and end of combustion based on luminosity and engine parameters have also been developed. Start of combustion has been found t V..oW buec a a. f_.uunnctι.i__uonιι o«->f__. s__,p_^■e*_e,_d>_*,, PJ-

cadLm_i-n', (»caLmax - ca5%L/L_maχ), l/(cadL_min - spk) , where caL^ - ca 5% L/I^_χ is the phase difference between location of peak luminosity and the location at which the luminosity is 5% of the peak luminosity in crank angle degrees. 5% of peak luminosity is exemplary only and other percentages early in the luminosity curve can also be used, for example 1% -10%. The correlation between observed start of combustion determined by a heat release model using a pressure transducer and predicted start of combustion based on the luminosity and engine parameters is shown in Figure 14. The correlation coefficient is 0.99 with a standard deviation of 0.32 degrees.

End of combustion can be predicted and is a function of speed, P_1nt, cal^, cad^, cadL_mι-_n, l/(cadl^_in - spk). The relationship between the end of combustion predicted from this correlation and the observed end of combustion determined by a heat release model using a pressure transducer is shown in Figure 15. The correlation coefficient is 0.99 with a standard deviation of 0.65 degrees.

With respect to Figures 9 through 15, it should be noted that correlations can be obtained although not quite as good by eliminating certain parameters or terms from the equations which will effect the standard deviations. Correlations between observed and predicted values using a reduced set of terms in the quadratic model for location of peak pressure, IMEP, air/fuel ratio, N0_χ emission, gas temperature at exhaust valve opening and start and end of combustion are shown in Figures 17 - -23. Certain parameters and terms may also be eliminated depending on the type of engine used. On the other hand, the luminosity parameter l/(cadl^_in - spk) has been found to be very important in determining the foregoing engine parameters and running conditions.

Referring again to Figure 24 as well as to Figure 25, another luminosity detecting arrangement generally similar to that shown in Figure 1 is depicted. In Figures 24 and 25, the illustrated probe 19 is comprised of a mounting housing 26 that has a threaded end 27 that is received within a threaded opening formed at the base of a counterbore that extends through the cylinder head 15 to the combustion chamber 16. The housing 26 may be formed from any suitable material, such as those materials normally used for the body of a spark plug.

An optic element 31 is affixed within the housing 26 in a suitable manner and has a portion that extends beyond the tip of the threaded end 27 so as to protrude into the combustion chamber 16 as shown in Figure 24.

Light signals may be transferred from the optic element 31 through a fiber optic 33 that extends from the housing 26 into a housing 34 to a silicon photo detector 25 positioned within a threaded portion 35 of the probe 19. The fiber optic 33 is held in place by a set screw 32. Other arrangements may also be used where the silicon photo detector 25 is remotely positioned from the engine 11, in which case a fiber optic bundle can be used to conduct the light from the probe 19 to the remotely positioned detector 25.

The optic element 31 can be formed from a relatively inexpensive material such as synthetic sapphire (A1₂0₃) or other materials having similar characteristics. An optic element 31 having a diameter of 3mm has been found to be practical and makes it relatively easy to install in the cylinder head 15.

In addition, a spark plug mounted probe 41, as shown in Figure 26, may also be used instead of the detector 18, probe 19 and spark plug 22 shown in Figures 24 and 25. With this arrangement, light is collected by a sapphire window 45 and transmitted via a fiber optic cable 42 to the silicon photo detector 25. The spark plug mounted probe 41 further includes a base 43 adapted for threaded engagement with the cylinder head 15 of the engine 11. A gasket is provided for securely fitting the spark plug mounted probe 41 into the cylinder head 15. This base 43 has a hole drilled therein for housing the sapphire window 45 and a portion of the fiber optic cable 42. A clamping nut 46 is used to secure a spark plug ceramic and electrode 47 within the base 43. A cable strain relief 48 is provided to help prevent the fiber optic cable 42 from being disengaged from the spark plug mounted probe 41. A sub miniature applications (S.M.A.) type fiber optic connector 49 is used to connect the fiber optic cable 42 to the photo detector 25. In engines having a pre-chamber, this spark plug mounted probe 41 can be used in the pre-chamber and a separately installed luminosity detector 18 and probe 19 as previously described with reference to Figures 24 and 25 can be used in the main, chamber.

The silicon photo detector 25 in Figure 24 has a peak spectral response at approximately 900nm and is reversed biased to six volts to minimize noise and allow a temperature measurement of the detector 25. The silicon photo detector 25 converts the light received from the probe 19 or spark plug mounted probe 41 into a photo detector current. Various luminosity spectra may be detected by the probe 19 and/or 41 and photo detector 25 or merely a total luminosity signal may be read. It has been found that certain constituents of the glowing gases in the combustion chamber 16 glow at different spectral ranges and this may be utilized to sense the amount and condition of such components in the combustion chamber 16 during each cycle of operation.

Depending on the particular gas or gases to be detected or sensed, it may be desirable to provide a monochromator or an optical filter in front of or on the silicon photo detector 25 so as to select a desired wavelength of light to be measured. For relationships in Figures 28-38, the probe 19 is employed to measure the overall radiant emission from products of combustion in the near infrared region between 850nm and lOOOnm. Thus, for the results generated in Figures 28-38, combustion radiation was filtered by an 850nm long pass filter. This filter in combination with the silicon photo detector's long wavelength cut-off of approximately lOOOnm gives the system a band pass from 850nm to lOOOnm. The near infrared region is monitored because it is not strongly influenced by radiant emission from the walls of the combustion chamber 16, nor is it sensitive to the emission from the flame surface. In addition, this particular wavelength band coincides with the peak spectral response of the silicon photo detector 25.

Referring again to Figure 24, the silicon photo detector 25 outputs its signal to a transimpedence amplifier 36 via a cable 37 which should be adequately supported to minimize engine vibration coupled to the cable 37. This amplifier 36 is used to amplify the current from the photo detector 25 and to convert that current to a voltage signal. The circuit diagram of an amplifier 36 for use with the arrangement of Figure 24 is shown in Figure 27. A low pass non recursive filter 38 can be used to further reduce ignition and vibrational noise if necessary. The filter 38 has a passband ripple of less than 0.1 dB, a stopband attenuation of greater than 40 dB and a cutoff at 5% of the sampling frequency. In such an arrangement the output of the transimpedence amplifier 36 can be fed to the low pass filter 38 via a cable or other suitable conducting means for further filtering. The low pass filter 38 then outputs its signal via conductors 39 to a computer control unit 40. In addition to converting the voltage signal from the transimpedence amplifier 36 and low pass filter 38 to an output indicative of luminosity, the remotely positioned control unit 40, which may measure certain engine parameters as well, may also receive input signals from other sensors normally employed on the engine, for example, air/fuel ratio, intake manifold pressure and temperature, engine speed and spark timing angle sensors. These types of sensors are normally employed with modern internal combustion engines and their signals can be processed in conjunction with the luminosity signal to provide certain measured characteristics of the engine operation. The remotely positioned control unit 40 may be of any suitable type and is particularly adapted to transmit the signal from the transimpedence amplifier 36 and low pass filter 38 into an output indicative of luminosity within the combustion chamber 16. A typical filtered mean radiation trace or luminosity curve measured in volts, and its gradient or derivative measured in volts/degree, both plotted against crank angle, are shown in Figure 28^". In non-reciprocating engines these measurements can be done in relation to output shaft angle. These measurements can also be done in relation to any timing indicator which gives an indication of time in the combustion cycle.

It has been determined that the time in the combustion cycle at which peak heat release rate occurs, and N0_χ emissions can be predicted using the luminosity detector 18 and/or the spark plug mounted probe 41 and the luminosity curve generated as a result of the detected luminosity in the combustion chamber 16 employed to adjust various parameters of the engine to obtain optimum performance.

Some of the luminosity parameters used, such as location of peak luminosity, are gain independent which means that they are not affected by variations in the gain of the signal which may decrease over time due to deposits forming on the probe 19 or sapphire window 45. As previously noted, gain independent luminosity parameters are typically measured in terms of time in the combustion cycle, which can be expressed as a location in crank angle degrees, instead of amplitude of the signal. Thus, as long as the luminosity signal maintains a sufficient signal to noise ratio, the location in crank angle degrees of a particular gain independent luminosity parameter, will remain the same regardless of gain changes in the signal. Data for the relationships and correlations set forth in Figures 28-38 were obtained using a single cylinder of a 69 liter, carbureted, spark ignited engine, natural gas engine having a compression ratio of 11.0, a bore of 170 mm, a crank radius of 95 mm, a stroke of 190 mm, and four valves per cylinder. The data was also obtained under various engine speeds (1200, 1350, and 1500 rp ) , lean air/fuel ratios and spark timing (21, 24, 27 degrees before top dead center) . The load measured in percent rated torque was varied between 50, 75 and 100. The engine was also run under different water jacket temperatures of 190°F, 210°F and 230°F. Spark plug gaps 0.015", 0.020" and 0.025" were used.

Figure 29 shows 100 cycle ensemble average luminosity (radiation) , pressure and heat release rate curves for a typical run condition. A normalized value is used for comparison. The pressure and heat release rate curves were generated using a water cooled pressure transducer. Heat release data was obtained from a single zone thermodynamic model. The relationship between pressure and luminosity is similar to that observed in a stoichiometric passenger car engine in that the peak pressure leads peak luminosity by a few crank angle degrees. As the figure indicates, peak heat release rate also leads peak luminosity. This is also the case in auto engines; however, in the auto engine peak heat release rate leads peak luminosity by a greater margin, reflecting the slower burn needed to minimize N0_χ production.

Figures 30 and 31 show two gain independent luminosity parameters which are effective in predicting the location in crank angle degrees of peak heat release rate (CADQCHMAX) . Figure 30 shows that the location in crank angle degrees of minimum radiation gradient or minimum luminosity derivative (CADLMIN) can be used to predict CADQCHMAX within +/- one (1) crank angle degree. As Figure 30 illustrates, the trend between these two parameters is linear with a linear correlation of 0.99. The data points are grouped according to engine speeds of 1200, 1350 and 1500 rpm.

Figure 31 shows the relationship between location in crank angle degrees of peak radiation or luminosity (CALMAX) and CADQCHMAX. CALMAX can be used to predict CADQCHMAX within +/- two (2) degrees. The trend is generally linear with a linear correlation of 0.96. The data is grouped by air/fuel ratio (A/F) , and as can be seen, this trend improves if air/fuel ratio is taken into account. The measurements are in crank angle degrees in the preferred embodiment, although as previously noted, these measurements can be based on other indicators which provide an indication of time in the combustion cycle.

Figure 32 shows the relationship of maximum detector output or peak luminosity and location in crank angle degrees of maximum detector output or peak luminosity vs. brake specific N0_X measured in g/hp-hr with the engine running at 1500 rpm and 100% load. NO_χ is reduced at constant torque by increasing air/fuel ratio and increasing intake pressure, that is, running leaner at higher pressure. Thus, for the data shown, the air/fuel ratio is lowest and the intake pressure is lowest at the highest NO_χ condition. It has been shown that peak luminosity increases with both decreasing air/fuel ratio and increasing intake pressure. Thus, the drop in peak luminosity at lower levels of N0_χ indicates that the change in air/fuel ratio had a stronger effect than the increase in intake pressure. The location of peak luminosity retards with decreasing NO_χ, as shown in Figure 32, as does the location of peak rate of heat release, as shown in Figure 33. Figure 33 shows the relationship between location of peak heat release rate in crank angle degrees and brake specific NO_χ.

Empirical correlations have also been developed by the inventors which use luminosity parameters in combination with various engine operating parameters such as engine speed in rpm, spark timing in crank angle degrees, intake manifold pressure in kPa, intake air temperature in degrees Celsius, coolant temperature in degrees Celsius, and air/fuel ratio to predict N0_χ emissions. These empirical correlations were established using a curve fitting routine which performs a multiple regression analysis using the quadratic response surface model as previously set forth. Each data point in the correlations of Figures 34-38 represents a 100 cycle ensemble average. The predicted values for NO_χ emissions determined by the quadratic model were compared with measured values determined by an emissions bench.

A comparison between predicted and measured brake specific N0_χ (BSNO_χ) values is shown in Figure 34. The predicted values were calculated using intake pressure, air/fuel ratio and location in crank angle degrees of minimum luminosity derivative (minimum radiation gradient) as variables in the quadratic model. The correlation has an R-squared value of 0.93 and a standard deviation of residuals (S.D.R.) of 1.28 g/hp-hr.

Figure 35 shows the same data as Figure 8 only sorted by brake specific N0_χ. As can be seen, lines of constant N0_χ level result which are similar to the lines of constant air/fuel ratio shown in Figure 31. Figure 35 also shows that for a given location of peak heat release rate (CADQCHMAX) , a higher level of N0_χ is indicated by a later location of peak luminosity (CALMAX) . Thus, the phase shift between location of peak luminosity and location of peak rate of heat release can be used to give an estimate of N0_χ emissions and air/fuel ratio.

Using the relationship between location of peak rate of heat release and location of peak luminosity shown in Figure 31, the inventors have determined that the air/fuel ratio variable in the quadratic model used in Figure 34 can be replaced with location of peak luminosity. Figure 36 shows the correlation between predicted brake specific N0_χ emissions, using intake pressure, location of peak luminosity and location of minimum luminosity derivative as variables in the aforementioned quadratic model, and measured brake specific N0_χ. As Figure 36 shows, good correlation between measured and predicted values was obtained. The R-squared value for this correlation is 0.87 and the standard deviation of residuals is 1.35g/hp-hr. The measured values were determined using an emissions bench.

The correlation shown in Figure 36 can be further improved by including the following variables in the quadratic model: engine speed, intake air temperature, coolant temperature, intake pressure, location of peak luminosity in crank angle degrees, the phase difference between location of peak luminosity and spark timing in crank angle degrees, the x-coordinate of the centroid of the luminosity signal in crank angle degrees (X_cp) , the normalized y-coordinate of the centroid of the luminosity signal against peak luminosity, the normalized second moment about X of the luminosity signal against peak luminosity and the normalized third moment about X of the luminosity signal against peak luminosity. The correlation between predicted brake specific N0_X emissions using these variables in the quadratic model and measured brake specific NO_χ determined by an emissions bench is shown in Figure 37. This comparison was generated using a reduced set of linear, cross product and squared terms in the quadratic model. The R-squared value for this correlation is 0.99 and the S.D.R. is 0.30g/hp-hr.

Figure 38 illustrates the cyclic variability problem associated with N0_χ emissions. Individual cycle NO predicted values are shown for 100 consecutive cycles. As can be seen, a small number of cycles produce most of the N0_χ. It is therefore desirable that the NO_χ emissions be predicted on a cycle to cycle basis so as to reduce this variability. The predictions developed by the inventors can be made on a cycle to cycle basis.

It should be readily apparent that the use of luminosity curve and in particular various gain independent parameters of that curve are extremely effective in measuring numerous conditions occurring within the combustion chamber and variations from chamber to chamber and cycle to cycle and in determining various operating parameters and engine running conditions. It is to be understood that the relationships between the luminosity and engine parameters, and engine operating and running conditions such as air/fuel ratio are exemplary only.

Those skilled in the art and armed with this knowledge should be able to provide the various engine controls such as the timing of the spark ignition, fuel delivery including timing and duration of fuel injection, or changing of air/fuel ratios through premixing devices such as carburetors or port injectors so as to obtain optimum performance in response to the measured combustion conditions mentioned and to minimize cycle to cycle and cylinder to cylinder variations. For example, by measuring the location of peak cylinder pressure, it is possible to phase the burning by utilizing feedback control of the spark timing to improve fuel consumption and better achieve emission control. In the same manner, the proper phasing of the burn rate using feedback control of spark timing can improve fuel consumption and provide better emission control. The control of air/fuel ratio by the feedback control of fuel flow can improve exhaust emissions and fuel consumption as well. The luminosity signal may also be employed to adjust spark timing or air/fuel ratio to compensate for engine aging and manufacturing differences.

The foregoing description is that of preferred embodiments of the invention and various changes and modifications may be made without departing from the spirit and scope of the invention as defined by the appended claims.

Claims

1. A method for operating an internal combustion engine having at least one combustion chamber, means for forming a combustible air/fuel mixture within the combustion chamber, means for detecting the luminosity within the combustion chamber during at least a portion of the combustion cycle, generating a curve based on the detected luminosity, determining the characteristic of at least one gain independent parameter of the luminosity curve so as to determine a particular combustion condition within the combustion chamber, adjusting at least one parameter of the engine to obtain a desired characteristic for said gain independent luminosity parameter at at least one point on the curve. 2. A method for operating an internal combustion engine as recited in Claim 1, wherein the characteristic of at least one gain independent parameter of the luminosity curve is determined in relation to output shaft angle and at least one parameter of the engine is adjusted to obtain a desired relationship between the characteristic of at least one gain independent luminosity parameter and output shaft angle.

3. A method for operating an internal combustion engine as recited in Claim 2, wherein the location at which the luminosity is a particular percentage of the peak luminosity is determined so as to determine location of start of combustion.

4. A method for operating an internal combustion engine as recited in Claim 2, wherein the location at which the luminosity is a particular percentage of the total integrated luminosity is determined so as to determine location of start of combustion.

5. A method for operating an internal combustion engine as recited in Claim 2, wherein the location of the minimum luminosity derivative is determined so as to determine location of end of combustion. ^'6. A method for operating an internal combustion engine as recited in Claim 2, wherein the location of peak luminosity derivative is determined so as to determine location of peak rate of heat release in the combustion chamber.

7. A method for operating an internal combustion engine as recited in Claim 2, wherein the location of peak luminosity is determined so as to determine location of peak pressure in the combustion chamber.

8. A method for operating an internal combustion engine as recited in Claim 3, wherein spark timing is adjusted to obtain a desired relationship between the location at which luminosity is a particular percentage of the peak luminosity and output shaft angle.

9. A method for operating an internal combustion engine as recited in Claim 4, wherein spark timing is adjusted to obtain a desired relationship between the location at which luminosity is a particular percentage of the total integrated luminosity and output shaft angle.

10. A method for operating an internal combustion engine as recited in Claim 6, wherein spark timing is adjusted to obtain a desired relationship between the location of the peak luminosity derivative and output shaft angle.

11. A method for operating an internal combustion engine as recited in Claim 7, wherein spark timing is adjusted to obtain a desired relationship between the location of peak luminosity and output shaft angle. 12. A method for operating an internal combustion engine as recited in Claim 2, wherein air/fuel ratio is adjusted to obtain a desired relationship between the characteristic of said gain independent luminosity parameter and output shaft angle. 13. A method for operating an internal combustion engine as recited in Claim 2, wherein fuel delivery is adjusted to obtain a desired relationship between the characteristic of said gain independent luminosity parameter and output shaft angle. 14. A method for operating an internal combustion engine having at least one combustion chamber, means for causing combustion to occur in the combustion chamber, means for detecting the luminosity within the combustion chamber during each cycle of operation, generating a curve based on the detected luminosity, determining the characteristic of at least one gain independent parameter of the luminosity curve so as to determine a particular combustion condition for each cycle within the combustion chamber, adjusting at least one parameter of the engine to obtain uniformity in each cycle of operation for that particular combustion condition.

15. A method for operating an internal combustion engine having multiple combustion chambers, means for effecting combustion in each of the combustion chambers, means for detecting luminosity in each of the combustion chambers, generating a curve based on the detected luminosity, determining the characteristic of at least one gain independent parameter of the luminosity curve so as to determine a particular combustion condition within each of the combustion chambers, adjusting at least one parameter of the engine to minimize chamber to chamber variation for that particular combustion condition.

16. A method for operating an internal combustion engine having at least one combustion chamber, means for forming a combustible air/fuel mixture within the combustion chamber, means for detecting the luminosity within the combustion chamber, generating a curve based on the detected luminosity, determining the characteristic of at least one gain independent parameter of the luminosity curve, adjusting spark timing to obtain a desired characteristic for said gain independent luminosity parameter.

17. A method for operating an internal combustion engine as recited in Claim 16, wherein location of peak luminosity is determined and spark timing is adjusted to obtain a desired location of peak luminosity.

18. A method for operating an internal combustion engine as recited in Claim 17, wherein spark timing is adjusted such that engine -efficiency is maximized without experiencing engine knock.

19. A method for operating an internal combustion engine having at least one combustion chamber, means for forming a combustible air/fuel mixture within the combustion chamber, means for detecting the luminosity within the combustion chamber during at least a portion of the combustion cycle, adjusting at least one parameter of the engine to obtain the desired luminosity characteristics so as to control the start of diffusion burning.

20. A method for operating an internal combustion engine having at least one combustion chamber, means for forming a combustible air/fuel ratio within the combustion chamber, means for detecting the luminosity within the combustion chamber during each cycle of operation of the engine from prior to initiation of combustion until after completion of combustion, generating a curve based on the detected luminosity, determining a characteristic of at least one gain independent parameter of the luminosity curve in relation to output shaft angle and measuring at least one parameter of the engine based on the determined characteristic of at least one gain independent parameter.

21. A method for operating an internal combustion engine as recited in Claim 20, wherein the location of peak pressure is measured.

22. A method for operating an internal combustion engine as recited in Claim 20, wherein indicated mean effective pressure is measured. 23. A method for operating an internal combustion engine as recited in Claim 20, wherein air/fuel ratio is measured.

24. A method for operating an internal combustion engine as recited in Claim 20, wherein N0_χ emission is measured.

25. A method for operating an internal combustion engine as recited in Claim 20, wherein gas temperature at exhaust valve opening is measured. 26. A method for operating an internal combustion engine as recited in Claim 20, wherein start of combustion is measured.

27. A method for operating an internal combustion engine as recited in Claim 20, wherein end of combustion is measured.

28. A method for operating an internal combustion engine as recited in Claim 21, wherein location of peak luminosity, location of peak luminosity derivative, and location of minimum luminosity derivative are determined.

29. A method for operating an internal combustion engine as recited in Claim 22, wherein location of peak luminosity, location of peak luminosity derivative and location of minimum luminosity derivative are determined. 30. A method for operating an internal combustion engine as recited in Claim 23, wherein location of peak luminosity, location of peak luminosity derivative, location of minimum luminosity derivative, the X- coordinate of the centroid of the luminosity signal and the normalized Y-coordinate of the centroid of the luminosity signal against peak luminosity are determined.

31. A method for operating an internal combustion engine as recited in Claim 24, wherein location of peak luminosity, location of minimum luminosity derivative, the X-coordinate of the centroid of the luminosity signal, the normalized Y-coordinate of the centroid of the luminosity signal against peak luminosity, the normalized first moment of the luminosity signal against peak luminosity, the normalized second moment about the X-coordinate of the centroid of the luminosity signal against peak luminosity and the normalized third moment about the X-coordinate of the centroid of the luminosity signal against peak luminosity are determined.

32. A method for operating an internal combustion engine as recited in Claim 25, wherein location of peak luminosity, location of minimum luminosity derivative, the X-coordinate of the centroid of the luminosity signal, the normalized Y-coordinate of the centroid of the luminosity signal against peak luminosity, the normalized first moment of the luminosity signal against peak luminosity, the normalized second moment about the X-coordinate of the centroid of the luminosity signal against peak luminosity and the normalized third moment about the X-coordinate of the centroid of the luminosity signal against peak luminosity are determined.

33. A method for operating an internal combustion engine as recited in Claim 26, wherein location of peak luminosity, location of minimum luminosity derivative and the location at which the luminosity is a particular percentage of the peak luminosity are determined.

34. A method for operating an internal combustion engine as recited in Claim 27, wherein location of peak luminosity, location of peak luminosity derivative and location of minimum luminosity derivative are determined.

35. A method for operating an internal combustion engine as recited in Claim 23, further including calculating the air/fuel ratio based on signals from a zirconia sensor.

36. A method for operating an internal combustion engine as recited in Claim 20, further including adjusting at least one parameter of the engine to attain a desired relationship between the determined characteristic of at least one gain independent parameter and output shaft angle.

37. A method for operating an internal combustion engine having at least one combustion chamber, means for forming a combustible air/fuel mixture within the combustion chamber, means for detecting the luminosity within the combustion chamber, generating a curve based on the detected luminosity, adjusting at least one parameter of the engine to attain the desired characteristic of the luminosity curve to compensate for engine aging. 38. A method for operating an internal combustion engine as recited in Claim 37, wherein the spark timing is adjusted.

39. A method for operating an internal combustion engine as recited in Claim 37, wherein air/fuel ratio is adjusted.

40. A method for operating an internal combustion engine having at least one combustion chamber, means for forming a combustible air/fuel mixture within the combustion chamber, means for detecting the luminosity within the combustion chamber, generating a curve based on the detected luminosity, determining the location of a gain independent parameter on the luminosity curve and predicting the time in the combustion cycle at which peak heat release rate occurs based on the determined location of the gain independent parameter.

41. A method for operating an internal combustion engine as recited in Claim 40, wherein the predicted time in the combustion cycle at which peak heat release rate occurs is determined in relation to crank angle.

42. A method for operating an internal combustion engine as recited in Claim 41, wherein the location of minimum luminosity derivative is determined. 43. A method for operating an internal combustion engine as recited in Claim 41, wherein location of peak luminosity is determined.

44. A method for operating an internal combustion engine as recited in Claim 43, further comprising estimating air/fuel ratio based on the relationship between location of peak luminosity and location of peak heat release rate.

45. A method for operating an internal combustion engine as recited in Claim 43, further comprising estimating N0_χ emissions based on the relationship between location of peak luminosity and location of peak heat release rate.

46. A method for operating an internal combustion engine having at least one combustion chamber, means for forming a combustible air/fuel mixture within the combustion chamber, means for detecting the luminosity within the combustion chamber, generating a curve based on the detected luminosity, determining at least one parameter of the luminosity curve and predicting N0_χ emissions based on the determination of at least one parameter.

47. A method for operating an internal combustion engine as recited in Claim 46, wherein at least one parameter of the luminosity curve is determined in relation to crank angle.

48. A method for operating an internal combustion engine as recited in Claim 47, wherein the location of peak luminosity is determined. 49. A method for operating an internal combustion engine as recited in Claim 47, wherein the location of minimum luminosity derivative is determined.

50. A method for operating an internal combustion engine as recited in Claim 47, wherein the location of peak luminosity and the location of minimum luminosity derivative are determined.