Method for Improving Spark Ignited Internal Combustion Engine Starting and Idling Using Poor Driveability Fuels

1. Field of the Invention

The present invention relates generally to methods of starting and idling an internal combustion engine for an automotive vehicle. More particularly, the present invention relates to a method for starting and idling an internal combustion engine utilizing a dynamic fuel source.

2. Description of the Related Art

Conventional fuel delivery systems for internal combustion engines adjust fuel delivered by a feedback signal created by an exhaust gas oxygen (EGO) sensor to maintain desired stoichiometric combinations. During starting and cold idling, such feedback from the EGO sensor is not available. Therefore, only open loop calculations of fuel requirements are available. A problem with open loop calculation is that such an open loop calculation will not compensate or vary as a function of the fuel blend currently being consumed. This insensitivity to fuel blend varies the operation of the internal combustion engine.

An example of an open loop system is U.S. Pat. No. 5,229,946 which discloses a method for optimizing engine performance for internal combustion engines. This method accounts for different blends of fuel; namely, pure fuels and different blends of fuel and alcohol. This method utilizes specific engine parameters to determine what type of fuel is being combusted. This method utilizes a different engine map for each blend of fuel. This approach is not flexible in that it requires a specific blend of fuel before it can look up a value in a specific map. This method also relies on sensing the amount of fuel in a fuel tank to determine whether a sensing event should even occur.

The method disclosed in U.S. Pat. No. 5,229,946 which fails to immediately determine the composition of the fuel to better enable the internal combustion engine to operate during start-up and idling situations. In fact, this disclosed method does not identify the fuel composition until the fuel tank is refilled. Further, there is no provision to measure the performance of the internal combustion engine. The method merely estimates the performance based on the last identification of fuel composition.

Accordingly, a method for maintaining a rotational speed of a crankshaft of an internal combustion engine is disclosed. The internal combustion engine includes a plurality of cylinders, each having a spark plug. A predetermined amount of fuel is delivered to be combusted in each of the plurality of cylinders with each rotation of the crankshaft or camshaft. The method includes the step of starting the internal combustion engine. The method also includes the step of measuring the rotational speed of the crankshaft. The method further includes the step of defining an expected engine speed. The method also includes the step of calculating a speed error as the rotational speed of the crankshaft less the expected the engine speed. The method also includes the step of adjusting the predetermined amount of fuel delivered to be combusted in each of the plurality of cylinders to reduce the speed error.

One advantage associated with the present invention is the ability to operate an internal combustion engine smoothly during start-up and cold idling regardless of the fuel quality. Another advantage associated with the present invention is the ability to reduce the speed error as soon as it is determined that the rotational speed of the crankshaft is not at a value that it should be. Yet another advantage associated with the present invention is the correction of the speed error independently of any parameter of the engine condition other than the rotational speed of the crankshaft. Still another advantage associated with the present invention is the ability to reduce the speed error to zero in a manner which does not require additional hardware, thus reducing the cost.

The above advantages of the invention will be more clearly understood by reading an example of an embodiment in which the invention is used to advantage with reference to the attached drawings wherein:

FIG. 1 is a perspective view partially cut away of an internal combustion engine;

FIG. 2 is graphic representation of engine speed as a function of time;

FIG. 3 is a graphic representation of engine speed trajectories as a function of time;

FIG. 4 is a graphic representation of engine speed signature analysis as a function of time;

FIG. 5 is a fuzzy input matrix for fuel control magnitude;

FIG. 6 is a fuzzy input matrix for spark offset control; and

FIG. 7 is a flow chart of one embodiment of the method according to the present invention.

Referring to FIG. 1, an internal combustion engine is generally indicated at 11. Although the internal combustion engine 11 is depicted and discussed as being a part of a motor vehicle (not shown), it should be appreciated by those skilled in the art that the internal combustion engine 11 may be used in any environment requiring the power generated thereby. The internal combustion engine 11 receives air through an air inlet port 13. A fuel injector (not shown) injects fuel for a plurality of cylinders 17. A fuel air mixture is drawn into each cylinder 17 through a plurality of inlet valves 19. The valves, inlet 19 and outlet 21, are moved between an open position and a closed position during different portions of a four stroke cycle. The opening and closing thereof is timed by a camshaft 23 which is rotated through a timing mechanism. When the air/fuel mixture is ignited by a spark plug (not shown), one associated with each of the cylinders 17, a piston 27 within each of the cylinders 17 is forced to move downwardly. This downward action rotates a crankshaft 29 which, in turn, transfers the power generated by the combustion of the air/fuel mixture into a mechanical rotating force to be controlled and used.

Referring to FIG. 2, characteristics of an engine speed as a function of time is shown for a type of fuel which is typically referred to as "hesitation fuel" or "fringe fuel." Hesitation or fringe fuels are fuels that are defined by a high driveability index based on the distillation characteristics of the fuel or are of a low grade or quality. The internal combustion engines must be capable of operating while combusting these fringe fuels. A first line 10 represents the engine speed as a function of time wherein the engine maintains a speed greater than zero. This speed is, however, lower than desired which results from low power output and, in turn, exhibits objectionable vibrations, noise and longer warm up time periods. The second line 12 represents the engine speed of an internal combustion engine using a hypothetical fuel which is of such composition that the internal combustion engine may stall in a period of less than five seconds. It should be appreciated by those skilled in the art that this is an undesirable situation.

Referring to FIG. 3, an engine speed graph as a function of time is represented. A solid line 14 represents the engine speed of an internal combustion engine using a certification fuel, a fuel used as a standard which may be found in the marketplace having known properties. A dotted line 16 is the idle speed control set point. In one embodiment, the idle speed control set point 16 is substantially constant at approximately 1200 RPM. After the internal combustion engine passes a run-up point 18, the engine speed of the internal combustion engine rapidly approaches the idle speed control set point, as it is designed to do. A dashed line 20 having its own run-up point 22 represents the engine speed of an internal combustion engine using a fringe fuel. After the internal combustion engine reaches its run-up point 22 with the fringe fuel, the engine speed of the internal combustion engine rapidly approaches a 300 RPM level. This level is too low as it results in an insufficient and irregular level of power output.

Referring to FIG. 4, a fringe fuel detection is graphically represented. The low grade fuel is combusted to create an engine speed along a dashed line 24 with a run-up point 26. An expected speed value 28 is graphically represented. The expected speed 28 is defined as the minimum of either a run-up speed, graphically represented as a heavy dotted line 30, or the idle speed control set point 32. Because the run-up speed trajectory 30 is greater than the idle speed control set point 32, the expected speed 28 becomes the idle speed set point 32. The difference between the actual speed 24 and the expected speed 28 is calculated to be a speed error. More specifically, the speed error is the difference between the minimum desired speed and the actual speed.

Referring to FIG. 7, the method for maintaining the rotational speed of the crankshaft of the internal combustion engine is disclosed. The internal combustion engine is operated during an initiation period. The initiation period typically includes steps of starting and idling the internal combustion engine. It should be appreciated that other events may occur during the initiation period.

The method maintains the rotational speed by reducing the speed error in the initiation period of the internal combustion engine and is generally shown at 34. The initiation period for the internal combustion engine varies with the temperature thereof. For example, the initiation period for the internal combustion engine at seventy degrees Fahrenheit is approximately two minutes. This time is inversely related to the temperature. Therefore, as the temperature of the internal combustion engine decreases, the initiation period typically increases.

The method begins at 36. The coolant temperature of the internal combustion engine is sensed and normalized at 38. The time from when the internal combustion engine begins a first revolution of cranking is measured and normalized at 40. It should be appreciated by those skilled in the art that these parameters may be replaced or augmented with other engine parameters.

Once these two normalized values are calculated, they are used in a fuzzy logic matrix or look-up table to produce a calibrated minimum run-up speed as a function of time, at 42. A minimum idle speed value is calculated as the idle speed set point minus a calibrated dead band, at 44. The run-up and the minimum idle speed are compared at 46. If the run-up speed is less than the minimum idle speed, an expected engine speed is defined as the run-up speed at 48. If, however, the run-up speed is greater than or equal to the minimum idle speed, the expected engine speed becomes the minimum idle speed at 50. In other words, the expected engine speed of the method 34 becomes the minimum of the either the run-up or the minimum idle speed. The speed error is calculated at 52 as being the actual rotational speed of the crankshaft minus the expected engine speed, whether it be the minimum idle speed or the run-up speed. The expected engine speed is then normalized at 54. The speed error is normalized at 56.

A fuel scalar is calculated at 58 using the fuzzy input matrix shown in FIG. 5. The fuel scalar is used to adjust the predetermined amount of fuel which is to be combusted in each of the cylinders. In one embodiment, the fuel scalar is calculated by the normalized expected speed and normalized speed error. These two values are used in a fuzzy input matrix, one shown in FIG. 5, to determine what the fuel scalar at time k should be. As the fuel scalar decreases, the amount of fuel delivered to the internal combustion engine is adjusted or, increased. The previous frame time or the "old" value of the fuel scalar is preserved as the fuel scalar at time k-1. FIG. 5 shows a value of 1.0 that produces no change in the amount of fuel delivered to the internal combustion engine because the speed error has a value of zero.

The fuel scalar at time k is compared to the fuel scalar at time k-1 at 60. If the fuel scalar.sub.k is greater than or equal to the previous fuel scalar.sub.k-1 , the fuel scalar at time k is assigned a value corresponding to its value at time k-1 with a first order exponential decay approximated by a rolling average filter at 62. If not, the fuel scalar at time k is unfiltered. The difference in modulating the filtering is provided to insure fast fuel scalar changes in the presence of a speed error and slowly diminishing fuel scalar changes once the speed error is corrected. It has been determined that it is more desirable to modulate the fuel such that the fuel scalar rapidly correct a speed error but not to rapidly remove corrections when the speed error does not exist. Therefore, when the speed error is being corrected, i.e., being reduced to zero, the fuel scalar is modulated such that it gradually increases to 1.0 in this embodiment.

A spark offset is added to a firing timing of each of the spark plugs to aid in the reduction of the speed error. The spark offset is calculated as a function of the expected speed and the speed error via the look-up table at 64. The spark offset fuzzy input matrix is shown in FIG. 6. As may be seen from viewing FIG. 6, the offset, an addition to the firing time in which the spark is to occur, is zero when there is no speed error. More specifically, there is no need to offset the desired spark timing when the speed error is non-existent.

The spark offset at time k is compared with the previous spark offset at time k-1 at 66. If the spark offset at time k is less than the previous spark offset at time k-1, the spark offset at time k is assigned a value corresponding to its value at time k-1 with a first order exponential decay approximated by a rolling average filter similar to the fuel scalar filter, at 68. More specifically, the spark offset is modulated rapidly to correct for the speed error but slowly once the speed error is eliminated. If not, the spark offset at time k is not filtered and the method is ended at 70. As noted above, and similarly to the fuel scalar, the spark offset is modulated to adjust the spark offset depending on the direction it is going. Once the method 34 has been completed, the method returns the control of the combustion of the fuel to the fuel and spark managing system (not shown) until the method is again invoked during the next controller background or frame time interval.

This concludes a description of an example of operation which the invention claimed herein is used to advantage. Those skilled in the art will bring to mind many modifications and alterations to the example presented herein without departing from the spirit and scope of the invention. Accordingly, it is intended that the invention be limited only by the following claims.