Misfire detection in an internal combustion engine using exhaust pressure

• US 5,193,513 A
• Filed: 06/03/1992
• Issued: 03/16/1993
• Est. Priority Date: 06/03/1992
• Status: Expired due to Term

• Current Assignee: Ford Global Technologies, LLC (Ford Motor Company)
• Original Assignee: Ford Motor Company
• Inventors: Marko, Kenneth A., Bryant, Bruce D.
• Primary Examiner(s): WOLFE JR, WILLIS RAY
• Application Number:
  US07/892,883
• Time in Patent Office: 286 Days
• Field of Search: 123/479, 123/481, 123/419, 123/425, 123/436, 123/571, 123/616, 73/116, 73/117.3, 364/424.03, 364/431.08
• US Class Current: 123/676
• CPC Class Codes: F02D 2200/1015 Engines misfires
• CPC Class Codes: F02D 41/1405 Neural network control
• CPC Class Codes: F02D 41/1448 the characteristics being an exhaust gas pressure
• CPC Class Codes: F02M 26/57 using electronic means, e.g. electromagnetic valves
• CPC Class Codes: F02M 26/61 in response to exhaust pressure
• CPC Class Codes: G01M 15/106 using pressure sensors

Abstract

Misfiring of individual cylinders in an internal combustion engine is detected based on sensing exhaust pressure during each operating cycle of the engine. A pressure signal is sampled and digitized during each cycle to form an input vector for a pattern classifier. Predetermined internal coefficients of the trained classifier enable the classifier to discriminate between firing and misfiring cylinders in real time with a high degree of accuracy. Overall system cost is reduced by sharing a pressure transducer with an exhaust gas recirculation system on the internal combustion engine.

Claims

1. Apparatus for detecting individual misfires in an internal combustion engine having an exhaust manifold, comprising:

pressure transducer means in communication with said exhaust manifold for generating a pressure signal responsive to exhaust pressure within said exhaust manifold;

position means for generating a plurality of position signals indicative of predetermined rotational positions within a cycle of said internal combustion engine;

an analog-to-digital converter coupled to said pressure transducer means and said position means for generating digitized pressure data corresponding to said predetermined rotational positions; and

a trained classifier having a data input coupled to said analog-to-digital converter and having a set of predetermined internal coefficients for processing said digitized pressure data to generate an output signal indicative of the detection of individual misfires.

2. The apparatus of claim 1 further comprising an exhaust gas recirculation system coupled to said pressure transducer means.

3. The apparatus of claim 1 further comprising register means coupled between said analog-to-digital converter and said trained classifier for formatting said digitized pressure data into respective input vectors for said trained classifier.

4. The apparatus of claim 3 wherein said register means is coupled to said position means, wherein said position means provides a reference signal for determining a full engine cycle, and wherein said input vectors each correspond to a respective engine cycle.

5. The apparatus of claim 3 further comprising translating means coupled to said trained classifier for translating said digitized pressure data to a more compact representation using a predetermined transformation function.

6. The apparatus of claim 5 wherein said transformation function is predetermined using principal component analysis.

7. The apparatus of claim 1 wherein said set of predetermined internal coefficients are determined in advance by training a trainable classifier on a substantially identical internal combustion engine using digitized pressure data containing known misfires.

8. A method for detecting individual misfires in an internal combustion engine having an exhaust manifold, comprising the steps of:

generating a pressure signal responsive to exhaust pressure within said exhaust manifold;

generating a plurality of position signals indicative of predetermined rotational positions within a cycle of said internal combustion engine;

generating digitized pressure data; and

processing said digitized pressure data using a set of predetermined internal coefficients to generate an output signal indicative of the detection of individual misfires.

9. The method of claim 8 further comprising the step of formatting said digitized pressure data into respective input vectors.

10. The method of claim 8 wherein said input vectors each correspond to a respective engine cycle.

11. The method of claim 8 further comprising the step of translating said digitized pressure data to a more compact representation using a predetermined transformation function.

12. The method of claim 11 wherein said transformation function is predetermined using principal component analysis.

13. The method of claim 8 wherein said set of predetermined internal coefficients are determined in advance by training a trainable classifier on a substantially identical internal combustion engine using digitized pressure data containing known misfires.