Method for cylinder-selective knock control in internal combustion engines

• US 5,386,367 A
• Filed: 05/28/1991
• Issued: 01/31/1995
• Est. Priority Date: 05/28/1990
• Status: Expired due to Term

• Current Assignee: Continental Automotive GmbH (Continental AG)
• Original Assignee: Siemens AG
• Inventors: Schindler, Peter, Ziegler, Herbert
• Primary Examiner(s): Teska, Kevin J.
• Assistant Examiner(s): Louis-Jacques, Jacques Harold
• Application Number:
  US07/706,248
• Time in Patent Office: 1,344 Days
• Field of Search: 364/431.01-431.12, 123/425, 123/435, 123/416, 123/414, 123/417, 123/415, 73/35
• US Class Current: 701/111
• CPC Class Codes: F02P 5/1522 with particular means concerning an individual cylinder
• CPC Class Codes: G01L 23/225 circuit arrangements therefor
• CPC Class Codes: Y02T 10/40 Engine management systems

Abstract

The method includes sampling a sensor signal of at least one knocking sensor after amplification and preparation at a predetermined frequency within a measurement window being predetermined for each firing cycle, to form sampled values. A knocking value is determined from a mean value of a predetermined number of greatest sampled values. A noise value is ascertained by forming a sliding mean value from a predetermined number of previous knocking values. An actual noise value is formed from the cylinder-selective noise values. A command noise value being dependent on at least one noise-determining engine parameter is read out of a performance graph. A difference between or a ratio of the actual noise value and the command noise value is formed. The comparison is processed in a controller to produce a correction factor. The correction factor is linked with a pilot value being read out of a pilot performance graph and being dependent on at least one noise-determining engine parameter, to produce a total amplification factor for the sensor signal.

Claims

1. A method for cylinder-selective knock control in internal combustion engines, which comprises:

sampling a sensor signal of at least one knocking sensor after amplification and preparation at a predetermined frequency within a measurement window being predetermined for each firing cycle, to form sampled values,

determining a knocking value from a mean value of a predetermined number of greatest sampled values,

ascertaining a noise value by forming a sliding mean value from a predetermined number of previous knocking values,

forming an actual noise value from the cylinder-selective noise values,

reading a command noise value being dependent on at least one noise-determining engine parameter out of a performance graph,

comparing the actual noise value and the command noise value.

processing the comparison in a controller to produce a correction factor, and

linking the correction factor with a pilot value being read out of a pilot performance graph and being dependent on at least one noise-determining engine parameter, to produce a total amplification factor for the sensor signal.

2. The method according to claim 1, which comprises performing the comparison step by forming a difference between the actual noise value and the command noise value.

3. The method according to claim 1, which comprises performing the comparison step by forming a ratio of the actual noise value and the command noise value.

4. The method according to claim 1, which comprises setting the actual noise value equal to the greatest cylinder-selective noise value.

5. The method according to claim 1, which comprises setting the actual noise value equal to the least cylinder-selective noise value.

6. The method according to claim 1, which comprises setting the actual noise value equal to the greatest sensor-selective noise value.

7. The method according to claim 1, which comprises setting the actual noise value equal to the least sensor-selective noise value.

8. The method according to claim 1, which comprises setting the actual noise value equal to a mean value of all of the cylinder-selective noise values.

9. The method according to claim 1, which comprises predetermining a control speed of the controller in the form of an integral controller with a control constant.

10. The method according to claim 1, which comprises setting the correction factor to a value "1" in certain operating states of the engine.

11. The method according to claim 1, which comprises limiting the correction factor to a range of values within predeterminable limit values.

12. The method according to claim 1, which comprises storing the limited correction factor in an adaptive performance graph.

13. The method according to claim 12, which comprises drawing the limited correction factor from the adaptive performance graph.

14. The method according to claim 1, which comprises forming the total amplification factor as a product of the pilot value and the correction factor.