Method and apparatus for reconstructing in-cylinder pressure and correcting for signal decay

• US 8,396,649 B2
• Filed: 01/27/2012
• Issued: 03/12/2013
• Est. Priority Date: 07/31/2009
• Status: Active Grant

• Current Assignee: Westport Fuel Systems Canada Inc. (Westport Fuel Systems, Inc.)
• Original Assignee: Westport Power Inc. (Westport Fuel Systems, Inc.)
• Inventors: Huang, Jian
• Primary Examiner(s): Cronin, Stephen K
• Assistant Examiner(s): NAJMUDDIN, RAZA NEHAL
• Application Number:
  US13/360,519
• Publication Number:
  US 20120150414A1
• Time in Patent Office: 410 Days
• Field of Search: 123/435, 123/568.16, 123/406.17, 123/406.22, 123/406.41, 123/406.27, 702/104, 702/98, 702/90, 702/182, 702/183, 702/185, 701/101, 701/102, 701/103, 701/104, 701/105, 701/106, 701/108, 701/110, 701/111, 701/114, 701/115, 73/350.1, 73/350.3, 73/350.4, 73/350.5, 73/351.2, 731/141.6, 731/141.7, 731/141.8, 731/141.9
• US Class Current: 701/111
• CPC Class Codes: F02D 2041/1432 the system including a filter, e.g. a low pass or high pass filter
• CPC Class Codes: F02D 2041/281 between sensors and control unit
• 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 35/024 using an estimation
• CPC Class Codes: F02D 35/027 using knock sensors
• CPC Class Codes: F02D 41/28 Interface circuits
• CPC Class Codes: G01M 15/08 by monitoring pressure in cylinders
• CPC Class Codes: G01M 15/12 by monitoring vibrations

Abstract

A method comprises steps for reconstructing in-cylinder pressure data from a vibration signal collected from a vibration sensor mounted on an engine component where it can generate a signal with a high signal-to-noise ratio, and correcting the vibration signal for errors introduced by vibration signal charge decay and sensor sensitivity. The correction factors are determined as a function of estimated motoring pressure and the measured vibration signal itself with each of these being associated with the same engine cycle. Accordingly, the method corrects for charge decay and changes in sensor sensitivity responsive to different engine conditions to allow greater accuracy in the reconstructed in-cylinder pressure data. An apparatus is also disclosed for practicing the disclosed method, comprising a vibration sensor, a data acquisition unit for receiving the vibration signal, a computer processing unit for processing the acquired signal and a controller for controlling the engine operation based on the reconstructed in-cylinder pressure.

Claims

1. A method of processing a vibration signal collected from a vibration sensor mounted on an engine component outside of a combustion chamber to reconstruct in-cylinder pressure data for an engine cycle of an internal combustion engine, the method comprising the steps of:

(a) collecting said vibration signal from said vibration sensor during said engine cycle and transmitting said vibration signal to a data acquisition unit;

(b) processing said vibration signal in a computer processing unit by;

(i) filtering said vibration signal to produce filtered data with frequencies in a frequency range associated with combustion effects inside said combustion chamber;

(ii) processing said filtered data for said engine cycle to produce processed data that is corrected for errors introduced by charge decay and sensor sensitivity by applying at least one correction factor determined as a function of an estimated motoring pressure associated with said engine cycle, being the same engine cycle from which said vibration signal was collected; and

(c) in said computer processing unit reconstructing said in-cylinder pressure data from said processed data for at least a portion of said engine cycle.

2. The method of claim 1 wherein said vibration signal is collected in a voltage mode, said processing step further comprises applying a gain error correction factor to said filtered data based on a ratio between a value for motoring pressure determined from said estimated motoring pressure at a predetermined timing during said engine cycle before the timing for start of combustion;

and a value for said vibration signal also at said predetermined timing.

3. The method of claim 2 wherein said predetermined timing is between 35 and 20 crank angle degrees before top dead center.

4. The method of claim 1 wherein said vibration signal is collected in a voltage mode, said processing step further comprises:

(iii) differentiating said filtered data with respect to engine cycle timing measured in crank angle degrees;

(iv) differentiating said estimated motoring pressure with respect to engine cycle timing measured in crank angle degrees;

(v) correcting for charge decay by applying a charge decay error correction factor to said filtered data by determining a difference on a time scale between timing for a peak differentiated motoring pressure and timing for a peak differentiated vibration signal, and as a function of said difference, determining said charge decay error by calculating a characteristic time scale that defines charge decay rate.

5. The method of claim 4 further comprising correcting said filtered data by applying a gain error correction factor, which is calculated based on a ratio between said peak differentiated motoring pressure and said peak differentiated vibration signal.

6. The method of claim 1 wherein when said vibration signal is collected in a current mode, said processing step further comprises:

(iii) differentiating said estimated motoring pressure with respect to engine cycle timing measured in crank angle degrees;

(iv) correcting for charge decay by applying a charge decay error correction factor to said filtered data by determining a difference on a time scale between timing for a peak differentiated motoring pressure and timing for a peak vibration signal, and as a function of said difference, determining said charge decay error by calculating a characteristic time scale that defines charge decay rate.

7. The method of claim 6 further comprising the steps of:

(d) integrating said vibration signal with respect to engine cycle timing measured in crank angle degrees;

(e) correcting said filtered data by applying a gain error correction factor to said filtered data based on a ratio between a value for motoring pressure determined from said estimated motoring pressure at a predetermined timing during said engine cycle before the timing for start of combustion; and

a value for the integrated vibration signal also at said predetermined timing.

8. The method of claim 1 further comprising calculating said estimated motoring pressure for said engine cycle as a function of intake manifold pressure measurements.

9. The method of claim 1 further comprising determining said estimated motoring pressure by calculating a motoring pressure curve that is calculated as a polytropic curve that is a function of sampled engine intake manifold pressures.

10. The method of claim 1 further comprising retrieving said estimated motoring pressure from a look up table of predetermined motoring pressure curves accessed as a function of intake manifold pressure.

11. The method of claim 1 wherein said filtering step comprises applying a low-pass filter to said vibration signal.

12. The method of claim 1 wherein said processing step further comprises calculating heat release rate data from said filtered data after it has been corrected for errors introduced by charge decay and sensor sensitivity, wherein heat release data becomes part of said processed data.

13. The method of claim 12 further comprising the steps of:

(d) at a predetermined time after peak combustion pressure, correcting out-of-range values of said heat release rate data with predetermined values to produce corrected heat release rate values, and

(e) computing corrected processed data based on said corrected heat release rate values.

14. The method of claim 13 wherein said predetermined time after peak combustion pressure is between 30 and 40 degrees after top dead center.

15. The method of claim 1 further comprising reconstructing an in-cylinder pressure curve for a portion of said engine cycle between 60 crank angle degrees before top dead center and 60 crank angle degrees after top dead center from said in-cylinder pressure data.

16. The method of claim 1 further comprising reconstructing an in-cylinder pressure curve for a portion of said engine cycle between 30 crank angle degrees before top dead center and 60 crank angle degrees after top dead center from said in-cylinder pressure data.

17. The method of claim 1 wherein said processing step further comprises:

(iii) separating said filtered vibration signal data into a low frequency component and a medium frequency component;

(iv) correcting said low frequency component to produce a corrected low frequency component by;

(aa) applying a gain error correction factor to said low frequency component based on a ratio between said motoring pressure and said vibration signal;

(bb) estimating a charge decay error caused by charge decay in said vibration signal in said engine cycle by determining a difference on a time scale between a value of said motoring pressure and a corresponding value of said vibration signal; and

(cc) based on said charge decay error, calculating a characteristic time scale that defines charge decay rate and correcting for charge decay by applying to said low frequency component a charge decay error correction factor based on said characteristic time scale;

(v) correcting said medium frequency component to produce a corrected medium frequency component by applying said gain error correction factor to said medium frequency component; and

(vi) combining said corrected low frequency component with said corrected medium frequency component.

18. The method of claim 17 further comprising using a Fast Fourier Transform to separate said filtered data into said low frequency component and said medium frequency component.

19. The method of claim 1 wherein said processing step employs a transfer function to determine said at least one correction factor, wherein said transfer function is selected from a map of calibrated transfer functions based on engine intake manifold pressure and engine speed.

20. The method of claim 1 wherein said vibration sensor is mounted on a bearing cap associated with a crankshaft of said engine.

21. The method of claim 20 further comprising applying a predetermined correction factor to said vibration signal before said processing step, to compensate for a difference in magnitude between a force applied to a piston of an engine cylinder and a force applied to said bearing cap.

22. The method of claim 1 wherein said vibration signal has a signal-to-noise ratio that is at least 4:

1.

23. The method of claim 1 wherein said vibration signal has a signal-to-noise ratio that is at least 6:

1.

24. A method of controlling combustion of fuel within a combustion chamber of an internal combustion engine comprising the steps of:

(a) acquiring a vibration signal for an engine cycle from a vibration sensor mounted on an engine component outside of said combustion chamber wherein said vibration signal comprises a component that correlates to in-cylinder pressure, and transmitting said vibration signal to a data acquisition unit;

(b) collecting engine data for said engine cycle, being the same engine cycle from which said vibration signal is acquired, said engine data comprising the engine speed, timing for top dead center and intake manifold pressure, and transmitting said engine data to said data acquisition unit;

(c) processing said vibration signal in a computer processing unit to produce processed data that is corrected for an estimated error introduced by signal decay and sensor sensitivity in said vibration signal by applying at least one correction factor determined as a function of said engine data and said vibration signal, and reconstructing an in-cylinder pressure curve for at least a portion of said engine cycle from said processed data; and

(d) transmitting said reconstructed in-cylinder pressure data to a controller for controlling the combustion within said combustion chamber based on said reconstructed in-cylinder pressure data.

25. The method of claim 24 further comprising estimating motoring pressure data for said engine cycle from said engine data and determining a gain error correction factor and a charge decay correction factor for correcting said vibration signal in said processing step as a function of said motoring pressure data.

26. The method of claim 25 wherein said vibration signal is collected in a voltage mode and said method further comprises differentiating said motoring pressure and differentiating said vibration signal, each with respect to engine cycle timing measured in crank angle degrees, and said gain error correction factor is calculated as a ratio between a value for motoring pressure determined from said estimated motoring pressure at a predetermined timing during said engine cycle before the timing for start of combustion;

and a value for said vibration signal also at said predetermined timing.

27. The method of claim 25 wherein said vibration signal is collected in a voltage mode and said method further comprises differentiating said motoring pressure and differentiating said vibration signal, each with respect to engine cycle timing measured in crank angle degrees, and said charge decay correction factor is determined by calculating the difference between a peak differentiated motoring pressure and a peak differentiated vibration signal, and determining a characteristic time scale defining charge decay rate, and determining said charge decay correction factor as a function of said characteristic time scale.

28. The method of claim 25 wherein said motoring pressure is calculated based on engine intake manifold pressure sampling assuming a polytropic curve.

29. The method of claim 25 wherein said at least one correction factor is determined by selecting a transfer function from a map of calibrated transfer functions that correct for gain and charge decay errors based on engine intake manifold pressure, engine speed and said vibration signal.

30. The method of claim 25 wherein said vibration signal is collected in a current mode and said method further comprises integrating said vibration signal with respect to the crank angle and determining a gain error correction factor based on a ratio between a value for said motoring pressure determined from said estimated motoring pressure data at a predetermined timing during said engine cycle before the timing for start of combustion;

and a value for said integrated vibration signal also at said predetermined timing, and applying said gain error correction factor to said vibration signal when processing said vibration signal in said computer processing unit.

31. The method of claim 25 wherein said vibration signal is collected in a current mode and said method further comprises differentiating said motoring pressure data with respect to engine timing measured in crank angle degrees and said charge decay correction factor is determined from a calculated charge decay error which is determined from a characteristic time scale that defines charge decay rate that is determined as a function of a time difference on a time scale between timing for a peak differentiated motoring pressure and timing for a peak vibration signal in the same engine cycle.

32. The method of claim 24 wherein said vibration signal has a signal-to-noise ratio that is at least 4:

1.

33. The method of claim 24 wherein said vibration signal has a signal-to-noise ratio that is at least 6:

1.

34. A control system for an internal combustion engine comprising:

(a) a vibration sensor mounted on an engine component outside of a combustion chamber where said vibration sensor is operative to generate a vibration signal when combustion occurs within said combustion chamber during an engine cycle,

(b) a data acquisition unit that is operative to receive said vibration signal transmitted from said vibration sensor and to correlate it with engine data associated with said engine cycle, being the same engine cycle associated with said vibration signal, said engine data comprising engine intake manifold pressure and parameters characteristic of engine operation comprising at least one of engine speed and timing for top dead center;

(c) a computer processing unit which is operative to receive and process data from said data acquisition unit to estimate an error introduced by vibration signal decay, and apply a correction factor determined as a function of said engine data and said vibration signal to said vibration signal to generate a corrected vibration signal based on said error, which is determined for said engine cycle, and thereby reconstruct in-cylinder pressure data for at least a portion of said engine cycle from said corrected vibration signal; and

,

(d) a controller that is operative to control engine operation based on said reconstructed in-cylinder pressure data and said parameters characteristic of engine operation.

35. The control system of claim 34 further comprising a resistor electrically connected in parallel to said vibration sensor.

36. The control system of claim 34 wherein said error has a component that is estimated by comparing a modeled motoring pressure curve based on engine intake manifold pressure to said vibration signal.

37. The control system of claim 36 wherein said modeled motoring pressure curve is an estimated polytropic curve based on engine intake manifold pressure sampling.

38. The control system of claim 36 wherein said error has a component that is estimated using a transfer function selected from a map of calibrated transfer functions based on engine intake manifold pressure and engine speed.

39. The control system of claim 34 wherein said engine component to which said vibration sensor is mounted is in a direct force transfer path from said combustion chamber.

40. The control system of claim 34 wherein said engine component to which said vibration sensor is mounted is a bearing cap for an engine crankshaft.