Methods and systems to derive health of mating cylinder using knock sensors

Patent metadata

Field	Value
Publication number	US-2016258378-A1
Application number	US-201514639736-A
Country	US
Kind code	A1
Filing date	Mar 5, 2015
Priority date	Mar 5, 2015
Publication date	Sep 8, 2016
Grant date	—

Abstract

A method of deriving the health of a first cylinder in a reciprocating device includes receiving a first signal from a first knock sensor in proximity to the first cylinder, receiving a second signal from a second knock sensor in proximity to a second cylinder, processing the first signal and the second signal, and deriving the health of the first cylinder by determining whether the first signal is coherent with the second signal.

First claim

1 . A method of deriving the health of a first cylinder in a reciprocating device, comprising:
receiving a first signal from a first knock sensor in proximity to the first cylinder; receiving a second signal from a second knock sensor in proximity to a second cylinder; processing the first signal and the second signal; and deriving the health of the first cylinder by determining whether the first signal is coherent with the second signal.

2 . The method of claim 1 , wherein the first cylinder is 360 crank angle degrees out of phase with the second cylinder.

3 . The method of claim 1 , wherein processing the first signal and the second signal comprises:
deriving first and second combustion signatures from the first and second signals by applying a band pass or low pass filter; deriving first and second valve signatures from the first and second signals by applying a band pass filter; deriving one or more events from the first and second combustion signatures and first and second valve signatures; and shifting the first combustion signature and the first valve signature or the second combustion signature and the second valve signature by 360 crank angle degrees.

4 . The method of claim 3 , wherein the one or more events comprise peak firing pressure, intake valve closure, exhaust valve closure, or a combination thereof.

5 . The method of claim 3 , wherein processing the first signal and the second signal comprises smoothing the first and second combustion signatures and the first and second valve signatures;

6 . The method of claim 1 , wherein processing the first signal and the second signal comprises:
scaling the first and second signals, wherein each data point included in the first and second signals is multiplied by a multiplier such that each of the first and second signals has a maximum amplitude of 1; measuring a first period of time between a start of the scaled signal and a time at which the scaled signal reaches a maximum amplitude; measuring a second period of time between the time at which the scaled signal reaches the maximum amplitude and a second time at which the scaled signal runs down to a sustain level; measuring a third period of time during which the scaled signal sustains; and measuring a fourth period of time during which the scaled signal runs down from the sustain level to zero.

7 . The method of claim 1 , wherein processing the first signal and the second signal comprises applying machine learning techniques to predict an occurrence of one or more events in the first and second signals.

8 . The method of claim 7 , wherein applying machine learning techniques comprises:
applying a two-state model using feature vectors; and using a Gaussian mixture model to predict an occurrence of one or more events in the first or second cylinder.

9 . The method of claim 7 , wherein applying machine learning techniques comprises:
applying predictive frequency bands to the first and second signal; and applying short time Fourier transforms to the first and second signal to predict an occurrence of one or more events in the first or second cylinder.

10 . The method of claim 1 , further comprising:
determining that the first knock sensor has malfunctioned; and wherein deriving the health of the first cylinder comprises:
deriving a combustion signature from the second signal by applying a band pass or low pass filter;
deriving a valve signature from the second signal by applying a band pass filter;
deriving one or more events from the combustion signature and the valve signature; and
using a lookup table to determine which of the one or more events should be occurring during operation of the reciprocating device at a known time or at a known crankshaft angle position.

11 . A system, comprising:
a controller configured to control a reciprocating engine, wherein the controller comprises a processor configured to: receive a first signal from a first knock sensor in proximity to a first cylinder; receive a second signal from a second knock sensor in proximity to a second cylinder; process the first signal and the second signal; and derive the health of the first cylinder by determining whether the first signal is coherent with the second signal.

12 . The system of claim 11 , wherein the first cylinder is 360 crank angle degrees out of phase with the second cylinder.

13 . The system of claim 11 , wherein processing the first signal and the second signal comprises:
deriving first and second combustion signatures from the first and second signals by applying a band pass or low pass filter; deriving first and second valve signatures from the first and second signals by applying a band pass filter; smoothing the first and second combustion signatures and the first and second valve signatures; deriving one or more events from the first and second combustion signatures and first and second valve signatures; and shifting the first combustion signature and the first valve signature or the second combustion signature and the second valve signature by 360 crank angle degrees.

14 . The system of claim 13 , wherein the one or more events comprise peak firing pressure, intake valve closure, exhaust valve closure, or a combination thereof.

15 . The system of claim 11 , wherein processing the first signal and the second signal comprises:
scaling the first and second signals, wherein each data point included in the first and second signals is multiplied by a multiplier such that each of the first and second signals has a maximum amplitude of 1; measuring a first period of time between a start of the scaled signal and a time at which the scaled signal reaches a maximum amplitude; measuring a second period of time between the time at which the scaled signal reaches the maximum amplitude and a second time at which the scaled signal runs down to a sustain level; measuring a third period of time during which the scaled signal sustains; and measuring a fourth period of time during which the preconditioned signal runs down from the sustain level to zero.

16 . The system of claim 11 , wherein processing the first signal and the second signal comprises applying feature vectors or predictive frequency bands to predict an occurrence of one or more events in the first and second signals.

17 . A non-transitory computer readable medium comprising executable instructions that when executed cause a processor to:
receive a first signal from a first knock sensor in proximity to a first cylinder; receive a second signal from a second knock sensor in proximity to a second cylinder, wherein the first cylinder is 360 crank angle degrees out of phase with the second cylinder; process the first signal and the second signal; and derive the health of the first cylinder comprising determining whether the first signal is coherent with the second signal.

18 . The non-transitory computer readable medium of claim 17 , wherein processing the first signal and the second signal comprises:
deriving first and second combustion signatures from the first and second signals by applying a band pass or low pass filter; deriving first and second valve signatures from the first and second signals by applying a band pass filter; deriving one or more events from the first and second combustion signatures and first and second valve signatures; and shifting the first combustion signature and the first valve signature or the second combustion signature and the second valve signature by 360 crank angle degrees.

Assignees

• Gen Electric

Inventors

• Bizub Jeffrey Jacob
• Matthews Brett Alexander
• Batal Iyad

Classifications

• F02D41/2438

  Active learning methods · CPC title

• F02D41/2474 Primary

  Characteristics of sensors · CPC title

• F02D41/26

  using computer, e.g. microprocessor · CPC title

• F02D41/2441

  characterised by the learning conditions · CPC title

• F02D41/2422

  Selective use of one or more tables · CPC title