Method and apparatus for engine analysis and remote engine analysis

• US 10,436,676 B2
• Filed: 08/09/2012
• Issued: 10/08/2019
• Est. Priority Date: 08/10/2011
• Status: Active Grant

• Current Assignee: Thompson Automotive Labs, LLC
• Original Assignee: Thompson Automotive Labs, LLC
• Inventors: Horlbeck, Eric G., Thompson, John, McDonald, Daniel, Chatham, Gregory Keith, Carr, Emery, Grubbs, William A.
• Primary Examiner(s): Mo, Xiao En
• Application Number:
  US14/235,575
• Publication Number:
  US 20140297157A1
• Time in Patent Office: 2,616 Days
• CPC Class Codes: F02D 2041/286 comprising means for signal processing
• CPC Class Codes: F02D 2041/288 for performing a transformation into the frequency domain, e.g. Fourier transformation
• CPC Class Codes: F02D 2200/1015 Engines misfires
• CPC Class Codes: F02D 41/1497 With detection of the mechanical response of the engine
• CPC Class Codes: G01L 1/10 by measuring variations of frequency of stressed vibrating elements, e.g. of stressed strings using resistance strain gauges G01L1/22
• CPC Class Codes: G01L 5/00 Apparatus for, or methods of, measuring force, work, mechanical power, or torque, specially adapted for specific purposes
• CPC Class Codes: G01M 15/11 by detecting misfire
• CPC Class Codes: G01M 15/12 by monitoring vibrations

Abstract

Methods of analyzing a cyclo-mechanical engine include detecting an engine signal associated with a plurality of cycles of the cyclo-mechanical engine, comparing a first sample of the engine signal with a second sample of the engine signal to determine a cycle length of the cyclo-mechanical engine, and analyzing the engine signal to detect a variation in the cycle length of the cyclo-mechanical engine over time based on the determined cycle length. Related systems and computer program products are also disclosed.

Claims

1. A method performed by at least one processor of a computer system of analyzing a cyclo-mechanical engine, comprising:

detecting, via a sensor, an engine signal associated with a plurality of cycles of the cyclo-mechanical engine;

determining a complexity of the engine signal;

selecting an operation based on the complexity of the engine signal;

performing the operation by comparing a first sample of the engine signal with a second sample of the engine signal to determine a cycle length of the cyclo-mechanical engine;

analyzing the engine signal;

detecting a variation in the cycle length of the cyclo-mechanical engine over time based on the cycle length based on the analyzing of the engine signal;

determining whether the variation in the cycle length is outside of a threshold; and

outputting a signal indicating an abnormal condition of the cyclo-mechanical engine responsive to the variation in the cycle length being outside of the threshold.

2. The method of claim 1, wherein the operation comprises a correlation operation and wherein comparing the first sample of the engine signal with the second sample of the engine signal comprises correlating the first sample with the second sample.

3. The method of claim 1, wherein determining the complexity of the engine signal comprises:

generating a frequency spectrum of the engine signal; and

analyzing the frequency spectrum of the engine signal to determine if the frequency spectrum of the engine signal has a single dominant peak or multiple dominant peaks; and

wherein selecting the operation based on the complexity of the engine signal comprises;

in response to determining that the frequency spectrum has multiple dominant peaks, selecting a first correlation operation; and

in response to determining that the frequency spectrum has a single dominant peak, selecting a second correlation operation that is different from the first correlation operation.

4. The method of claim 3, wherein the first correlation operation comprises maximum adjacent correlation.

5. The method of claim 4, wherein performing maximum adjacent correlation comprises:

determining a minimum expected cycle length and a maximum expected cycle length;

choosing a plurality of sections of the engine signal for analysis;

defining a first correlation window at each section of the engine signal, the first correlation window having a length equal to the minimum expected cycle length;

defining a second correlation window at each section of the engine signal, the second correlation window having a length equal to the first correlation window;

correlating a portion of each section of the engine signal covered by the first correlation window with a portion of each section of the engine signal covered by the second correlation window;

repeating steps of;

(a) increasing sizes of the first correlation window and the second correlation window in each section of the engine signal; and

b) correlating the portion of each section of the engine signal covered by the first correlation window with the portion of each section of the engine signal covered by the second correlation window;

until the sizes of the first correlation window and the second correlation window are equal to the maximum expected cycle length.

6. The method of claim 5, further comprising:

identifying a correlation peak from results of the correlations.

7. The method of claim 1, further comprising:

sampling the engine signal at a first sampling rate, wherein the sampling of the engine signal at the first sampling rate comprises;

processing the engine signal at a second sampling rate that is lower than the first sampling rate to determine changes in engine speed over time;

analyzing a plurality of changes in engine speed to develop an estimate of engine speed variations over time; and

processing the engine signal at the first sampling rate to determine a variation in engine speed over time in response to the estimate of engine speed variations over time.

8. The method of claim 7, further comprising:

discarding a subset of the plurality of changes in engine speed, resulting in a discarded subset and a remaining subset, wherein each change in engine speed in the discarded subset occurs outside a predetermined range of engine speeds;

wherein analyzing changes in engine speed to develop the estimate of engine speed variations over time comprises analyzing the remaining subset.

9. The method of claim 1, wherein detecting the engine signal comprises detecting an acoustic signal associated with the cyclo-mechanical engine and converting the acoustic signal into an electrical signal.

10. The method of claim 9, wherein the acoustic signal comprises an exhaust signal.

11. The method of claim 1, wherein detecting the engine signal comprises at least one of detecting an electrical signal, an RF signal, a heat signal, an infrared signal, a vibration signal, an optical signal, an airflow signal, and/or a mass flow signal, associated with the cyclo-mechanical engine.

12. The method of claim 1, wherein detecting the engine signal comprises detecting an alternator output signal.

13. The method of claim 1, wherein the cyclo-mechanical engine comprises a four-stroke internal combustion engine or a two-stroke internal combustion engine, and wherein analyzing the engine signal comprises analyzing a cycle corresponding to a 720-degree rotation of a crankshaft in the cyclo-mechanical engine where the cyclo-mechanical engine comprises a four-stroke internal combustion engine or analyzing a cycle corresponding to a 360-degree rotation of a crankshaft in the cyclo-mechanical engine where the cyclo-mechanical engine comprises a two-stroke internal combustion engine.

14. The method of claim 1, wherein analyzing the engine signal to detect a variation in the cycle length of the cyclo-mechanical engine comprises generating a plurality of engine cycle length estimates, the method further comprising refining the engine cycle length estimates by aligning engine signal sections from adjacent durations.

15. The method of claim 14, wherein aligning engine signal sections from adjacent durations comprises:

choosing a starting point for a first cycle duration;

choosing a starting point for a second cycle duration adjacent the first cycle duration by starting with a duration length of the first cycle duration and shifting the second cycle duration by a number of samples that increases a correlation between the first duration cycle and the second cycle duration.

16. The method of claim 15, wherein determining lengths of third and subsequent cycle durations comprises shifting the first cycle duration and the second cycle duration by a number of samples that increases a correlation with any of a previous N number of cycle durations, where N>

1.

19. The method of claim 1, further comprising:

receiving a cylinder timing signal from the cyclo-mechanical engine; and

comparing the cylinder timing signal with the engine signal to identify which of a plurality of cylinders in the cyclo-mechanical engine is misfiring.

20. The method of claim 19, wherein the cylinder timing signal comprises a spark plug signal, a signal output by a cam sensor, or a signal output by a crank sensor in the cyclo-mechanical engine.

23. The method of claim 1, further comprising:

identifying misfire activity of the cyclo-mechanical engine from the variation in the cycle length of the cyclo-mechanical engine over time; and

identifying a strength and/or composition of a chemical cleaner that is recommended to remediate the misfire activity for the cyclo-mechanical engine being analyzed.

37. The method of claim 1, wherein determining a complexity of the engine signal comprises determining a number of cylinders of the cyclo-mechanical engine.

38. The method of claim 1, wherein outputting the signal indicating the abnormal condition responsive to the variation the cycle length being outside the threshold comprises outputting the signal indicating the abnormal condition responsive to a variation in the cycle length of adjacent durations of the engine signal being outside the threshold.

17. A method of analyzing a cyclo-mechanical engine having a fundamental basis cycle length, the method comprising:

detecting an engine signal associated with a plurality of fundamental basis cycles of the cyclo-mechanical engine;

determining a complexity of the engine signal;

analyzing, based on the complexity, the engine signal;

detecting a variation in a cycle length of the fundamental basis cycle of the cyclo-mechanical engine;

determining whether the variation in the cycle length is outside of a threshold; and

outputting a signal indicating an abnormal condition of the cyclo-mechanical engine responsive to the variation in the cycle length being outside of a threshold.

18. The method of claim 17, wherein the cyclo-mechanical engine is an internal combustion engine having a 720 degree fundamental basis cycle.

21. A method of analyzing a cyclo-mechanical engine, comprising:

detecting, via a sensor, an engine signal associated with a plurality of cycles of the cyclo-mechanical engine;

determining a complexity of the engine signal;

analyzing the engine signal;

estimating cycle lengths of the cyclo-mechanical engine at a plurality of locations in the engine signal based on analyzing of the engine signal, wherein a first location and a second location of the plurality of locations are selected based on the complexity of the engine signal;

refining the cycle length estimates to generate refined cycle length estimates;

identifying cycle length variations in response to the refined cycle length estimates;

wherein refining the cycle length estimates comprises aligning engine signal sections from adjacent durations by choosing a starting point for a first duration, and choosing a starting point for a second duration adjacent the first duration by starting with a duration length of the first duration and shifting the second duration by a number of samples that increases a correlation between the first duration and the second duration;

determining whether a plurality of cycle length variations is outside of a threshold and

outputting a signal indicating an abnormal condition of the cyclo-mechanical engine responsive to the plurality of the cycle length variations being outside of the threshold.

22. The method of claim 21, wherein determining lengths of third and subsequent durations comprises shifting the first duration and the second duration by a number of samples that increases a correlation with any of a previous N number of durations, where N>

1.

24. A computer program product for analyzing a cyclo-mechanical engine comprising a tangible computer readable storage medium having computer readable program code embodied in the tangible computer readable storage medium, the computer readable program code comprising:

computer readable program code configured to detect, via a sensor, an engine signal associated with a plurality of cycles of the cyclo-mechanical engine;

computer readable program code configured to determine a complexity of the engine signal;

computer readable program code configured to select a correlation operation based on the complexity of the engine signal;

computer readable program code configured to compare a first sample of the engine signal with a second sample of the engine signal to determine a cycle length of the cyclo-mechanical engine, wherein the computer readable program code configured to compare a first sample of the engine signal with a second sample of the engine signal comprises computer readable program code configured to perform the correlation operation to correlate the first sample with the second sample;

computer readable program code configured to analyze the engine signal;

computer readable program code to detect a variation in the cycle length of the cyclo-mechanical engine over time based on the cycle length;

computer readable program code to determine whether the variation in the cycle length is outside of a threshold; and

computer readable program code configured to output a signal indicating an abnormal condition of the cyclo-mechanical engine responsive to the variation in the cycle length being outside of a threshold.

25. A system for analyzing a cyclo-mechanical engine, comprising:

a sensor configured to generate an engine signal associated with a plurality of cycles of the cyclo-mechanical engine; and

an engine speed analyzer configured to;

determine a complexity of the engine signal;

select a correlation operation based on the complexity of the engine signal;

compare, via performance of the correlation operation, a first sample of the engine signal with a second sample of the engine signal to determine a cycle length of the cyclo-mechanical engine;

analyze the engine signal;

detect a variation in the cycle length of the cyclo-mechanical engine over time based on the cycle length based on the analyzing of the engine signal;

determine whether the variation in the cycle length is outside of a threshold; and

output a signal indicating an abnormal condition of the cyclo-mechanical engine responsive to the variation in the cycle length being outside of a threshold.

26. The system of claim 25, wherein the engine speed analyzer is configured to output engine speed variation data relating to engine speed variation of the cyclo-mechanical engine, the system further comprising a processor coupled to the engine speed analyzer and configured to determine a recommended engine service based on the engine speed variation data.

27. The system of claim 25, wherein the system further comprises a processor configured to identify a strength and/or composition of a chemical cleaner that is recommended to remediate misfire activity in the cyclo-mechanical engine based on an output of the engine speed analyzer.

28. A system for remotely analyzing a cyclo-mechanical engine, comprising:

a communication module coupled to a communication interface and configured to receive raw engine signal data associated with a plurality of cycles of a remotely located cyclo-mechanical engine over a communication interface; and

an engine speed analyzer configured to receive the raw engine signal data, to select a correlation operation based on a complexity of the raw engine signal data determined based on the raw engine signal data, and to compare, by performing the correlation operation, a first sample of the raw engine signal data with a second sample of the raw engine signal data to determine a cycle length of the cyclo-mechanical engine, to analyze the raw engine signal data, to detect a variation in the cycle length of the cyclo-mechanical engine over time based on the cycle length based on the analyzing of the raw engine signal data, to determine whether the variation in the cycle length is outside of a threshold, and to output a signal indicating an abnormal condition of the cyclo-mechanical engine responsive to the variation in the cycle length being outside of a threshold.

29. A method of analyzing a cyclo-mechanical engine, comprising:

detecting, via a sensor, an airflow signal associated with operation of the cyclo-mechanical engine;

determining a complexity of the airflow signal;

selecting a correlation operation based on the complexity of the airflow signal;

correlating a first sample of the airflow signal with a second sample of the airflow signal by performing the correlation operation to determine a cycle length of the cyclo-mechanical engine;

analyzing the airflow signal;

detecting a variation in the cycle length of the cyclo-mechanical engine over time based on the cycle length based on the analyzing of the engine signal;

determining whether the variation in the cycle length is outside of a threshold; and

outputting a signal indicating an abnormal condition of the cyclo-mechanical engine responsive to the variation in the cycle length being outside of a threshold.

30. The method of claim 29, wherein the airflow signal comprises an exhaust signal.

31. The method of claim 29, wherein detecting the airflow signal comprises detecting the airflow signal using an airflow sensor in an air intake or exhaust.

32. The method of claim 29, wherein the airflow signal comprises an acoustic signal, and wherein detecting the airflow signal comprises detecting the airflow signal using a microphone.

33. The method of claim 29, further comprising:

receiving a cylinder timing signal from the cyclo-mechanical engine; and

comparing the cylinder timing signal with the airflow signal to identify which of a plurality of cylinders in the cyclo-mechanical engine is misfiring.

34. A method of analyzing a cyclo-mechanical engine, comprising:

detecting an engine signal associated with operation of the cyclo-mechanical engine;

comparing, based on a determined complexity of the engine signal, a first sample of the engine signal with a second sample of the engine signal to determine a cycle length of the cyclo-mechanical engine;

analyzing the engine signal

detecting a variation in a shape of the engine signal based on the analyzing of the engine signal to identify an anomaly in a characteristic pattern of the engine signal that is indicative of incomplete combustion in a cylinder in the cyclo-mechanical engine;

determining whether the anomaly in the characteristic pattern is outside of a threshold and

outputting a signal indicating an abnormal condition of the cyclo-mechanical engine responsive to the anomaly being outside of a threshold.

35. The method of claim 34, wherein identifying an anomaly in a characteristic pattern of the engine signal that is indicative of incomplete combustion in a cylinder in the cyclo-mechanical engine comprises analyzing the engine signal to detect a variation in a period of at least a portion of the engine signal.

36. The method of claim 34, wherein identifying an anomaly in a characteristic pattern of the engine signal that is indicative of incomplete combustion in a cylinder in the cyclo-mechanical engine comprises analyzing the engine signal to determine a probability of a dirty or clogged valve or clogged fuel injector in the cyclo-mechanical engine.