Mechanical health monitor apparatus and method of operation therefor

• US 9,173,574 B2
• Filed: 04/28/2011
• Issued: 11/03/2015
• Est. Priority Date: 04/22/2009
• Status: Active Grant

• Current Assignee: Vital Metrix Inc.
• Original Assignee: Streamline Automation LLC
• Inventors: Teixeira, Rodrigo E., Reich, Alton
• Primary Examiner(s): Lau, Tung S
• Application Number:
  US13/096,845
• Publication Number:
  US 20120022805A1
• Time in Patent Office: 1,650 Days
• Field of Search: 702/50
• US Class Current: 1/1
• CPC Class Codes: A61B 5/0205 Simultaneously evaluating both cardiovascular conditions and different types of body conditions, e.g. heart and respiratory condition
• CPC Class Codes: A61B 5/14551 for measuring blood gases
• CPC Class Codes: A61B 5/318 Heart-related electrical modalities, e.g. electrocardiography [ECG]
• CPC Class Codes: A61B 5/7203 for noise prevention, reduction or removal
• CPC Class Codes: A61B 5/7214 using signal cancellation, e.g. based on input of two identical physiological sensors spaced apart, or based on two signals derived from the same sensor, for different optical wavelengths
• CPC Class Codes: A61B 5/7278 Artificial waveform generation or derivation, e.g. synthesising signals from measured signals A61B5/327 takes precedence

Abstract

A probabilistic data signal processor used to determine health of a system is described. Initial probability distribution functions are input to a dynamic state-space model, which iteratively operates on probability distribution functions, such as state and model probability distribution functions, to generate a prior probability distribution function, which is input to a probabilistic updater. The probabilistic updater integrates sensor data with the prior to generate a posterior probability distribution function passed to a probabilistic sampler, which estimates one or more parameters using the posterior, which is output or re-sampled and used as an input to the dynamic state-space model in the iterative algorithm. In various embodiments, the probabilistic data signal processor is used to filter output from any mechanical device using appropriate physical models, which optionally include chemical, electrical, optical, mechanical, or fluid based models. Examples to valve bearings and pipe systems are provided.

Claims

1. An apparatus for estimation of state of mechanical health of a mechanical element, comprising:

a probabilistic data signal processor embedded in a computer, said probabilistic data signal processor comprising;

a dynamic state-space model comprising;

a process model; and

an observation model;

a probabilistic updater configured to generate a posterior probability distribution function using both;

(1) a prior probability distribution function output from said dynamic state-space model and (2) readings from a load sensor, wherein the mechanical element comprises a valve bearing of a valve, said valve comprising a pipe fitting configured to regulate flow of a substance through a pipe, said load sensor configured to provide time dependent measurements related to load on the valve bearing of the valve; and

a probabilistic sample module configured to operate on said posterior probability distribution function to generate an output in the form of at least one probability of state of health of said valve.

2. The apparatus of claim 1, further comprising:

an iterative control module configured to;

control said load sensor; and

control physical rotational movement of said valve bearing using said posterior probability distribution function in an iterative control loop, said time dependent readings of said load sensor correlated with said movement of said valve bearing.

3. The apparatus of claim 1, wherein said time dependent readings comprise data related to load on said valve bearing, wherein output correlated with the posterior probability distribution function relates to an estimated remaining life of said valve.

4. The apparatus of claim 1, wherein the time dependent readings from said load sensor comprise deterministic readings, said probabilistic data signal processor configured to convert said deterministic readings into said output, said output comprising a probability distribution function representative of state of health of said valve.

5. The apparatus of claim 1, wherein said process model of said dynamic state-space model comprises a valve dynamics model, said valve dynamics model configured to operate on at least pressure readings from said load sensor.

6. The apparatus of claim 5, wherein said observation model of said dynamic state-space model comprises a load sensor noise model configured to use mathematical representations accounting for observation noise produced by said load sensor.

7. The apparatus of claim 5, further comprising:

an irregular sampling model configured to provide non-uniform sampling information to said valve dynamics model.

8. The apparatus of claim 1, wherein said process model of said dynamic state-space model comprises a pipe failure dynamic state-space model configured to operate on at least one of:

fluid velocity data;

stress data from a stress gauge; and

strain data from a strain gauge.

9. The apparatus of claim 8, wherein said output comprises a probability function estimating likelihood of pipe failure.

10. A method for estimating state of mechanical health of a mechanical element, comprising the steps of:

collecting with a sensor time dependent readings related to a mechanical state of the mechanical element, wherein the mechanical element comprises a valve bearing of a valve, said valve comprising a pipe fitting configured to regulate flow of a substance through a pipe, said sensor configured to measure load on said valve bearing;

calculating a prior probability distribution function with a probabilistic data signal processor, said probabilistic data signal processor comprising;

a dynamic state-space model comprising;

a process model; and

an observation model using the readings from said sensor,

determining a posterior probability distribution function through combination of;

(1) said prior probability distribution function output from said dynamic state-space model and (2) the readings from said sensor; and

using said posterior probability distribution function, generating an output comprising an estimate of state of health of the mechanical element.

11. The method of claim 10, wherein said estimate comprises a state of health probability function of the valve bearing.

12. The method of claim 10, wherein said sensor yields a first state parameter reading, wherein said output comprises an output to a second state parameter, wherein said first state parameter is different from said second state parameter.

13. The method of claim 12, wherein said first state parameter comprises a pressure reading, wherein said second state parameter comprises a measure related to a coefficient of friction of rotation of said valve bearing.

14. The method of claim 10, further comprising the step of:

operating on said posterior distribution function with a probabilistic sampler to generate an updated parameter, said updated parameter iteratively passed to said dynamic state-space model.

15. The method of claim 14, further comprising the step of:

resampling output of said probabilistic sampler to form a new probability distribution function.

16. The method of claim 15, said step of resampling using a sigma point Kalman filter.