Multi sensor transducer and weight factor

• US 6,757,641 B1
• Filed: 06/28/2002
• Issued: 06/29/2004
• Est. Priority Date: 06/28/2002
• Status: Expired due to Fees

• Current Assignee: National Aeronautics and Space Administration US Government As Represented By The Administrator of The
• Original Assignee: United States Of America As Represented By The Administrator Of The National Aeronautics And Space Administration
• Inventors: Lane, John, Perotti, Jose M., Immer, Christopher D., Eckhoff, Anthony J.
• Primary Examiner(s): Barlow, John
• Assistant Examiner(s): Cherry, Stephen J.
• Application Number:
  US10/185,830
• Time in Patent Office: 732 Days
• Field of Search: 702/101, 702/150, 702/179, 702/190, 702/191, 702/194, 702/199, 700/45, 600/481, 602/77, 602/85, 340/573
• US Class Current: 702/194
• CPC Class Codes: G01D 3/08 with provision for safeguarding the apparatus, e.g. against abnormal operation, against breakdown

Abstract

A multi-sensor transducer and processing method allow insitu monitoring of the senor accuracy and transducer ‘health’. In one embodiment, the transducer has multiple sensors to provide corresponding output signals in response to a stimulus, such as pressure. A processor applies individual weight factors to reach of the output signals and provide a single transducer output that reduces the contribution from inaccurate sensors. The weight factors can be updated and stored. The processor can use the weight factors to provide a ‘health’ of the transducer based upon the number of accurate versus in-accurate sensors in the transducer.

Claims

1. A transducer comprising:

a plurality of sensors providing a plurality of corresponding output signals in response to a stimulus; and

a processor for calculating a health of the transducer by applying a plurality of weight factors to each of the plurality of corresponding output signals for calculating a number of inaccurate sensors of the plurality of sensors and for providing a single transducer output.

2. The transducer of claim 1 further comprising a multiplex circuit to selectively couple the plurality of corresponding output signals to the processor.

3. The transducer of claim 1 wherein the processor uses the weight factors of the plurality of sensors to identify a transducer failure.

4. A transducer comprising:

a plurality of sensors to provide a plurality of corresponding output signals in response to a stimulus;

a processor to apply a plurality of weight factors to each of the plurality of corresponding output signals and provide a single transducer output; and

said transducer having the weight factors for each of the plurality of sensors calculated by ${\left(W_i\right)}^{-1}=\frac{1}{2<br><br>N}<br><br>\underset{j=1}{\stackrel{N}{\sum <br><br>}}<br><br>\left(\frac{x_i}{x_j}+\frac{x_j}{x_i}\right),$

 

where W_i is the corresponding weight factor for sensor i providing output signal x_i.

5. The transducer of claim 4 wherein the single transducer output, Xoutput, is calculated by $X_{\mathrm{Output}}=\frac{\sum ^{}_{}}{}$

Ni=1

xi

Wi∑

Nj=1

Wj.

6. A transducer comprising:

a plurality of N sensors for providing a plurality of corresponding output signals in response to a stimulus;

a memory for storing a plurality of weight factors corresponding to the plurality of sensors;

a processor for calculating the health of the transducer by applying a plurality of weight factors to each of the plurality of corresponding output signals for calculating a number of inaccurate sensors of the plurality of sensors and providing a single transducer output; and

a multiplex circuit for selectively coupling the plurality of corresponding output signals to the processor.

7. The transducer of claim 6 wherein the plurality of sensors are pressure sensors.

8. A transducer comprising:

a plurality of N sensors to provide a plurality of corresponding output signals in response to a stimulus;

a memory to store a plurality of weight factors corresponding to the plurality of sensors;

a processor to apply a plurality of weight factors to each of the plurality of corresponding output signals and provide a singe transducer output;

a multiplex circuit to selectively couple the plurality of corresponding output signals to the processor; and

said transducer having the plurality of weight factors for each of the plurality of sensors is calculated by ${\left(W_i\right)}^{-1}=\frac{1}{2<br><br>N}<br><br>\underset{j=1}{\stackrel{N}{\sum <br><br>}}<br><br>\left(\frac{x_i}{x_j}+\frac{x_j}{x_i}\right),$

 

where W_i is the corresponding weight factor for sensor i providing output signal x_i, and the single transducer output, Xoutput, calculated by $X_{\mathrm{output}}=\frac{\underset{i=1}{\stackrel{N}{\sum <br><br>}}<br><br>x_i<br><br>W_i}{\underset{j=1}{\stackrel{N}{\sum <br><br>}}<br><br>W_j}.$

9. A method of providing an output from a multiple sensor transducer comprising:

sampling an output signal from each of the multiple sensors, wherein the multiple sensors are N sensors;

determining a weight factor for each sensor based upon the sampled output signals, wherein the weight factor represents an accuracy level of the corresponding sensor;

calculating a single output signal by applying the weight factors to the sampled output signals;

calculating a health of the transducer by calculating a number of inaccurate sensors of the multiple sensors.

10. The method of claim 9 further comprises:

storing the weight factors for subsequent use; and

updating the stored weight factors such that the transducer learns which sensors of the N sensors are reliable and which are unreliable via the weight factors.

11. A method of testing a multiple sensor transducer comprising:

sampling an output signal from each of the multiple sensors;

determining a weight factor for each sensor based upon the sampled output signals, wherein the weight factor represents an accuracy level of the corresponding sensor; and

calculating a health of the transducer by calculating a number of inaccurate sensors of the multiple sensors.

12. The method of claim 11 wherein the multiple sensors are N sensors, and the weight factors are calculated by the following algorithm, ${\left(W_i\right)}^{-1}=\frac{1}{2}$

N

∑

Nj=1

(xixj+xjxi),

where W_i is the corresponding weight factor for sensor i providing output signal x_i.

13. The method of claim 11 further comprises:

applying the weight factor to each of the sampled output signals; and

calculating a single transducer output using the weighted output signals.

14. The method of claim 13 wherein the weight factors are calculated by the following algorithm, ${\left(W_i\right)}^{-1}=\frac{1}{2}$

N

∑

Nj=1

(xixj+xjxi),

where W_i is the corresponding weight factor for sensor i providing output signal x_i.

15. The method of claim 14 wherein the single output signal, Xoutput, is calculated by the following algorithm, $X_{\mathrm{output}}=\frac{\sum ^{}_{}}{}$

Ni=1

xi

Wi∑

Nj=1

Wj.