Method and system for detecting the misfire of an internal combustion engine utilizing angular velocity fluctuations

• US 5,239,473 A
• Filed: 08/21/1991
• Issued: 08/24/1993
• Est. Priority Date: 04/20/1990
• Status: Expired due to Fees

• Current Assignee: University of Michigan
• Original Assignee: Board of Regents of the University of Michigan (University of Michigan)
• Inventors: Rizzoni, Giorgio, Ribbens, William B.
• Primary Examiner(s): Harvey, Jack B.
• Assistant Examiner(s): Pipala, E. J.
• Application Number:
  US07/741,338
• Time in Patent Office: 734 Days
• Field of Search: 364/431.01, 364/431.03, 364/431.07, 364/431.08, 364/508, 364/511, 364/550, 364/551.01, 364/484, 364/572, 364/574, 123/416-419, 123/428, 123/436, 123/480, 324/379, 324/391, 324/392, 324/77 R, 73/116, 73/117.3
• US Class Current: 701/111
• CPC Class Codes: F02B 2075/025 two
• CPC Class Codes: F02D 2041/288 for performing a transformation into the frequency domain, e.g. Fourier transformation
• CPC Class Codes: F02D 2200/1004 Estimation of the output torque
• CPC Class Codes: G01M 15/11 by detecting misfire

Abstract

A hybrid method and system are disclosed for the detection of internal combustion engine misfires. The method and system are capable of detecting misfires even at very low occurrence rates in real time with inexpensive analog and digital (i.e. hybrid) electronics suitable for use on-board a vehicle, thus satisfying current and proposed exhaust emission control regulations. The method and system exploit a measurement of engine crankshaft angular velocity in conjunction with hybrid electronic signal processing. Once the angular velocity signal is conditioned to minimize the effects of random error and external disturbances, four alternate computationally efficient methods may be used to extract information pertaining to individual cylinder torque productions. Two of the methods employ extremal samples of the estimated torque or velocity waveform to obtain a random torque nonuniformity index or metric. The other two methods utilize a transformation into the frequency domain after the input data is first sampled and windowed to substantially lower probability of error. All of the methods provide an M-dimensional torque non-uniformity vector of individual cylinder performance for each individual engine cycle. The non-uniformity vector represents a compact and efficient measurement to which statistical decision theory is applied.

Claims

1. A method for detecting the misfire of a reciprocating internal combustion engine having a plurality of reciprocating components, at least one cylinder and a crankshaft, the method comprising the steps of:

measuring the angular velocity of the crankshaft to obtain an electrical signal representation thereof, the electrical signal containing data;

sampling the data synchronously with crankshaft angular position at a sampling rate to obtain sampled data;

windowing the sampled data;

filtering the data to minimize the effects of random error and disturbances in the electrical signal;

calculating an M-dimensional torque nonuniformity vector derived from the windowed and filtered data, wherein the step of calculating includes the step of transforming the windowed and filtered data to a frequency domain equivalent; and

applying a statistical decision algorithm to the torque nonuniformity vector, which algorithm statistically distinguishes between a true misfire and a normal cyclic variability which characterizes the combustion process in each cylinder of the engine, to thereby obtain a misfire signal.

3. The method as claimed in claim 1 or claim 2 wherein the step of transforming includes the step of applying a discrete Fourier transform to the windowed and filtered data.

4. The method as claimed in claim 1 or claim 2 wherein the step of windowing includes the step of applying a Hanning window to the sampled data.

5. The method as claimed in claim 1 or claim 2 further comprising the step of removing the average angular velocity of the crankshaft from the measured angular velocity prior to the step of windowing.

6. The method as claimed in claim 1 or claim 2 further comprising the step of computing the torque due to combustion energy effectively exciting the crankshaft to thereby account for the torque attributable to the reciprocating forces.

11. The system as claimed in claim 1 or claim 2 or claim 7 or claim 8 wherein the statistical decision algorithm includes a statistical pattern recognition algorithm.

2. A method for detecting the misfire of a reciprocating internal combustion engine having a plurality of reciprocating components, at least one cylinder and a crankshaft, the method comprising the steps of:

measuring the angular velocity of the crankshaft to obtain an electrical signal representation thereof, the electrical signal containing data;

sampling the data synchronously with crankshaft angular position at a sampling rate to obtain sampled data;

windowing the sampled data;

filtering the data to minimize the effects of random error and disturbances in the electrical signal to obtain a filtered angular velocity signal;

calculating an M-dimensional torque nonuniformity vector derived from the windowed and filtered data, wherein the step of calculating includes the step of transforming the windowed and filtered data to be a frequency domain equivalent;

computing an index of torque nonuniformity based on the M-dimensional vector, the index of torque nonuniformity representing the variation in torque from cylinder to cylinder and from cycle to cycle; and

applying a statistical decision algorithm to one of the torque nonuniformity vector and the index of torque nonuniformity, which algorithm statistically distinguishes between a true misfire and normal cyclic variability which characterizes the combustion process in each cylinder of the engine, to thereby obtain a misfire signal.

7. A system for detecting the misfire of a reciprocating internal combustion engine having a plurality of reciprocating components, at least one cylinder and a crankshaft, the system comprising:

means for measuring the angular velocity of the crankshaft to obtain an electrical signal representation thereof, the electrical signal containing data;

means for sampling the data synchronously with the crankshaft angular position at a sampling rate to obtain sampled data;

means for windowing the sampled data;

means for filtering the data to minimize the effects of random error and disturbances in the electrical signal;

means for calculating an M-dimensional torque nonuniformity vector derived from the windowed and filtered data, wherein the means for calculating includes means for transforming the windowed and filtered data to a frequency domain equivalent; and

means for applying a statistical decision algorithm to the torque nonuniformity vector, which algorithm statistically distinguishes between a true misfire and normal cyclic variability which characterizes the combustion process in each cylinder of the engine, to thereby obtain a misfire signal.

9. The system of claim 7 or claim 8 further comprising means for removing the average angular velocity of the crankshaft from the measured angular velocity prior to the step of windowing.

10. The system as claimed in claim 7 or claim 8 further comprising computing means for computing the torque which is due to combustion energy effectively exciting the crankshaft to thereby account for the torque attributable to the reciprocating forces.

12. The system as claimed in claim 7 or claim 8 wherein the means for transforming includes means for applying a discrete Fourier transform.

13. The system as claimed in claim 7 or claim 8 wherein the means for windowing includes means for applying a Hanning window.

14. The system as claimed in claim 7 or claim 8 wherein the means for filtering includes an analog filter.

8. A system for detecting the misfire of a reciprocating internal combustion engine having a plurality of reciprocating components, at least one cylinder and a crankshaft, the system comprising:

means for measuring the angular velocity of the crankshaft to obtain an electrical signal representation thereof, the electrical signal containing data;

means for filtering the data to minimize the effects of random error and disturbances in the electrical signal;

means for sampling the data synchronously with crankshaft angular position at a sampling rate to obtain sampled data;

means for windowing the sampled data;

means for calculating an M-dimensional torque nonuniformity vector derived from the windowed and filtered data, wherein the means for calculating includes means for transforming the windowed and filtered data to a frequency domain equivalent;

means for computing an index of torque nonuniformity based on the M-dimensional vector, the index of torque nonuniformity representing the variation in torque from cylinder to cylinder and from cycle to cycle; and

means for applying a statistical decision algorithm to one of the torque nonuniformity vector rand the index of torque nonuniformity, which algorithm statistically distinguishes between a true misfire and normal cyclic variability which characterized the combustion process in each cylinder of the engine, to thereby obtain a misfire signal.