Method of error detection during evaluation of sensor signals

• US 6,550,311 B2
• Filed: 05/02/2001
• Issued: 04/22/2003
• Est. Priority Date: 05/05/2000
• Status: Active Grant

• Current Assignee: Robert Bosch GmbH
• Original Assignee: Robert Bosch GmbH
• Inventors: Sloboda, Robert
• Primary Examiner(s): Larkin, Daniel S.
• Assistant Examiner(s): MILLER, ROSE MARY
• Application Number:
  US09/847,829
• Publication Number:
  US 20010042398A1
• Time in Patent Office: 720 Days
• Field of Search: 123/406.14, 123/406.16, 123/479, 731/172, 731/173, 731/181, 73/350-- 3
• US Class Current: 73/35.04
• CPC Class Codes: F02D 35/027 using knock sensors
• CPC Class Codes: F02D 41/222 relating to the failure of sensors or parameter detection devices
• CPC Class Codes: G01D 3/08 with provision for safeguarding the apparatus, e.g. against abnormal operation, against breakdown
• CPC Class Codes: G01L 23/221 for detecting or indicating knocks in internal combustion engines
• CPC Class Codes: G01L 23/225 circuit arrangements therefor
• CPC Class Codes: Y02T 10/40 Engine management systems

Abstract

A method for error detection during an evaluation of sensor signals, in particular during an evaluation of output signals of knock sensors in an internal combustion engine, with at least one sensor and an evaluating device connected therewith, the method includes the steps of forming at least one variable reference level from output signals of a sensor or sensors; comparing the variable reference level for error detection with upper and/or limiting values; making a conclusion about an error in the case of a predeterminable exceeding of the upper and/or exceeding of the lower limiting value by the actual reference level; and forming at least one limiting value in dependence on preceding reference values.

Claims

1. A method for error detection during an evaluation of output signals of knock sensor or sensors in an internal combustion engine, with at least one sensor and an evaluating device connected therewith, the method comprising the steps of forming at least one variable reference level from output signals of said sensor or sensors;

comparing the variable reference level for error detection with upper and lower limiting values;

making a conclusion about an error in the case of a predeterminable exceeding of the upper limiting value or falling below the lower limiting value by the actual reference level; and

forming at least one of the upper and lower limiting value in dependence on preceding reference values.

2. A method as defined in claim 1;

and further comprising using the factors depending on a rotary speed.

3. A method as defined in claim 1;

and further comprising forming the upper and lower limiting value as self-learning characteristic lines UDKSNO and UDKSNU depending on preceding standardized reference values.

4. A method as defined in claim 3;

and further comprising forming the both self-learning characteristic lines UDKSNO and UDKSNU rotary speed-dependent and by a slow filtering from standardized reference levels.

5. A method as defined in claim 3;

and further comprising forming self-learning characteristic lines UDKSNU and UDKSNQ via reference level, and both.

6. A method as defined in claim 5;

and further comprising forming the limiting values in accordance with the equation;

7. A method as defined in claim 1;

and further comprising providing an error detection first above an average rotary speed and/or an average value for the standardized reference level.

8. A method as defined in claim 1;

and further comprising a releasing an adaptation of the limiting value only when the learned value for the standardized reference level changes only significantly and/or is located above a minimal threshold.