Adaptive learning control for engine intake air flow

• US 5,080,064 A
• Filed: 04/29/1991
• Issued: 01/14/1992
• Est. Priority Date: 04/29/1991
• Status: Expired due to Fees

• Current Assignee: General Motors Corporation (General Motors Company)
• Original Assignee: General Motors Corporation (General Motors Company)
• Inventors: Buslepp, Kenneth J., Kosowski, Mark G., Trombley, Douglas E.
• Primary Examiner(s): Nelli, Raymond A.
• Application Number:
  US07/693,105
• Time in Patent Office: 260 Days
• Field of Search: 123/399, 123/486, 123/440, 123/480, 123/489, 123/429, 60/274, 364/431.05
• US Class Current: 123/399
• CPC Class Codes: F02D 2011/102 at least one throttle being moved only by an electric actuator
• CPC Class Codes: F02D 41/0002 Controlling intake air
• CPC Class Codes: F02D 41/008 Controlling each cylinder individually
• CPC Class Codes: F02D 41/1406 with use of a optimisation method, e.g. iteration
• CPC Class Codes: F02D 41/2451 characterised by what is learned or calibrated
• CPC Class Codes: F02D 41/2464 Characteristics of actuators
• CPC Class Codes: F02D 43/00 Conjoint electrical control of two or more functions, e.g. ignition, fuel-air mixture, recirculation, supercharging or exhaust-gas treatment
• CPC Class Codes: Y02T 10/40 Engine management systems

Abstract

An adaptive learning control method is described for regulating the quantity of air inducted into a fuel injected internal combustion engine to achieve a desired air-fuel ratio, where the amount of fuel supplied to each engine cylinder is determined directly as a function of operator demand for engine output. The rotational speed of the engine is sensed, and indication are derived for the actual mass of air inducted into each engine cylinder and for a desired mass of air for each engine cylinder. For the region in which the engine is operating, open-loop and closed-loop values are determined for controlling engine air flow. A previously stored learning correction value corresponding to the engine operating region is obtained, and the quantity of air flowing to the engine is adjusted in accordance with the sum of the open-loop value, the closed-loop value, and the learning correction value. The learning correction value is periodically updated as a function of the difference between the indicated actual and desired mass of air for each engine cylinder, when the engine continuously operates in the region and a predetermined time has elapsed since the engine first began continuous operation in the region.

Claims

1. An adaptive learning control method for regulating the quantity of intake air flowing into a fuel injected internal combustion engine to achieve a scheduled air-fuel ratio, where the amount of fuel supplied to each engine cylinder is determined directly as a function of the operator demand for engine output, with the engine having a plurality of operating regions defined by its operating parameters, the steps of the method comprising:

deriving an indication of the actual mass of air inducted into each engine cylinder;

deriving an indication of a desired mass of air for each engine cylinder, which corresponds to the scheduled air-fuel ratio;

determining the region in which the engine is operating based upon the engine operating parameters;

obtaining a value for a previously stored learning correction that corresponds to the region in which the engine is operating;

adjusting the quantity of intake air flowing into the engine as a function of at least the learning correction value, to reduce the difference between the indications for the actual and desired mass of air for each engine cylinder; and

updating the value of the stored learning correction as a function of the difference between the indications for the actual and desired mass of air for each engine cylinder.

2. The method described in claim 1, wherein the stored learning coefficient is updated, when the engine continuously operates in the region and a predetermined time has elapsed since the engine first began continuous operation in the region.

3. The method described in claim 1, wherein the indication for the desired mass of air for each engine cylinder is derived as a function of at least the amount of fuel supplied to each cylinder.

4. The method described in claim 1, wherein the indication for the desired mass of air for each engine cylinder is derived as a function of at least the amount of fuel supplied to each each cylinder and the rotational speed of the engine.

5. The method described in claim 1, wherein the step of deriving the indication of the actual mass of air inducted into each engine cylinder includes sensing the mass air flow into the engine.

6. The method described in claim 1, wherein the step of deriving the indication of the actual mass of air inducted into each engine cylinder includes sensing the engine operating air-fuel ratio.

7. An adaptive learning control method for regulating the quantity of intake air flowing into a fuel injected internal combustion engine to achieve a scheduled air-fuel ratio, where the amount of fuel supplied to each engine cylinder is determined directly as a function of the operator demand for engine output, with the engine having a plurality of operating regions defined by its operating parameters, the steps of the method comprising:

sensing the rotational speed of the engine;

deriving an indication of the actual mass of air inducted into each engine cylinder;

deriving an indication of a desired mass of air for each engine cylinder, which corresponds to the scheduled air-fuel ratio;

determining the region in which the engine is operating based upon the engine operating parameters;

determining an open-loop value for controlling engine air flow based upon the rotational speed of the engine and the indication for the desired mass of air for each engine cylinder;

determining a closed-loop value for controlling engine air flow based upon the difference between the indications for the actual and desired mass of air for each engine cylinder;

obtaining a value for a previously stored learning correction that corresponds to the region in which the engine is operating;

adjusting the quantity of intake air flowing into the engine in accordance with the sum of the open-loop value, the closed-loop value, and learning correction value; and

updating the value of the stored learning correction periodically, as a function of the difference between the indications of actual and desired mass of air for each engine cylinder, when the engine continuously operates in the region and a predetermined time has elapsed since the engine first began continuous operation in the region.

8. The method described in claim 7, wherein the stored learning correction is periodically updated based upon the integration of the difference between the indications for the actual and desired mass of air for each engine cylinder.

9. The method described in claim 7, wherein the indication for the desired mass of air for each engine cylinder corresponding to the scheduled air-fuel ratio is derived at least as a function of the amount of fuel supplied to each cylinder.

10. The method described in claim 7, wherein the indication for the desired mass of air for each engine cylinder is derived as a function of at least the amount of fuel supplied to each cylinder and the rotational speed of the engine.

11. The method described in claim 7, wherein the step of deriving the indication of the actual mass of air inducted into each engine cylinder includes sensing the mass air flow into the engine.

12. The method described in claim 7, wherein the step of deriving the indication of the actual mass of air inducted into each engine cylinder includes sensing the engine operating air-fuel ratio.