Photoelastic neural torque sensor
First Claim
1. A method of quantizing flexure in a material, the method comprising the steps of:
- providing a photoelastic material;
flexing the photoelastic material into a first flexed orientation;
passing a first light signal through said photoelastic material disposed in said first flexed orientation, the first light signal exiting the photoelastic material at a plurality of points over the surface of the photoelastic material as a first modified light signal; and
,in a neural network based on said first modified light signal, generating a first output signal representative of a first degree of flexure in said photoelastic material disposed in said first orientation.
3 Assignments
0 Petitions
Accused Products
Abstract
An opto-mechanical torque sensing device suitable for use with rotary machinery integrates a photoelastic polymer detector, a light source, a photoelastic image sensor and an artificial intelligence neural network and algorithm. The photoelastic polymer is formed into a hollow cylinder and bonded to metal collars located at each end of the cylinder. The collars serve to readily place the cylinder detector over a machine shaft and affix it to the shaft using a keyway, setscrew, or spring pin. Alternatively, split collars and a split sleeve can also be used to clamp the detector to the shaft. In the presence of polarized light, the photoelastic polymer detector generates an optical fringe pattern that varies as a function of torque applied to the shaft or other machine part on which the detector is mounted. The artificial intelligence neural network learns the mapping relationship between the observed optical fringe pattern and the applied torque using a training procedure. Once trained, the neural network generates a signal representative of torque in the shaft based on observed fringe pattern.
57 Citations
24 Claims
-
1. A method of quantizing flexure in a material, the method comprising the steps of:
-
providing a photoelastic material; flexing the photoelastic material into a first flexed orientation; passing a first light signal through said photoelastic material disposed in said first flexed orientation, the first light signal exiting the photoelastic material at a plurality of points over the surface of the photoelastic material as a first modified light signal; and
,in a neural network based on said first modified light signal, generating a first output signal representative of a first degree of flexure in said photoelastic material disposed in said first orientation. - View Dependent Claims (2)
-
-
3. A method of quantizing flexure in a material comprising the steps of:
-
providing an elongate portion of photoelastic material; fixing a first end of said elongate portion of photoelastic material at a first region of a torsionally compliant shaft; fixing a second end of said elongate portion of photoelastic material at a second region of said torsionally compliant shaft; flexing the photoelastic material into a first flexed orientation by moving said first and second ends of the elongate portion of photoelastic material relative to one another by torsionally deforming said shaft between said first and second regions; passing a first light signal through said photoelastic material disposed in said first flexed orientation, the first light signal exiting the photoelastic material as a first modified light signal; and
,in a neural network based on said first modified light signal, generating a first output signal representative of a first degree of flexure in said photoelastic material disposed in said first orientation. - View Dependent Claims (4, 5, 6, 7)
-
-
8. A method of quantizing flexure in a material comprising the steps of:
-
providing a photoelastic material; flexing the photoelastic material into a first flexed orientation; passing a first light signal through said photoelastic material disposed in said first flexed orientation, the first light signal exiting the photoelastic material as a first modified light signal; while passing said first light signal through said photoelastic material, optically phase shifting said first light signal in said photoelastic material as said first modified light signal; detecting said first modified light signal in a photoelectric array; generating a set of output signals at said photoelectric array based on said first modified light signal; and
,in a neural network based on said set of output signals, generating a torque output signal in said neural network. - View Dependent Claims (9, 10, 11, 12)
-
-
13. A torque sensor for use in conjunction with a drive system, the torque sensor comprising:
-
a light source for generating a first light signal; a photoelastic member on at least a one of a drive member of the drive system and a driven member of the drive system, the photoelastic member being adapted to receive said first light signal and generate a phase shifted light signal based on a magnitude of flexure of the photoelastic member, the phase shifted light signal being a two dimensional phase shift function of position on the surface of the photoelastic material; a detector for receiving the phase shifted second light signal as said two dimensional phase shift function of position on the surface of the photoelastic material and generating an electric signal based on the phase shifted second light signal; and
,a processing apparatus adapted to receive said electric signal and generate a torque output signal representative of said magnitude of flexure in the photoelastic member. - View Dependent Claims (14, 15, 16)
-
-
17. A torque sensor apparatus comprising:
-
a light source for generating a first light signal; a first member including a photoelastic body portion, the first member having a first end and a second end, the photoelastic body portion being adapted to receive a first light signal and generate over the surface of the photoelastic body portion a phase shifted light signal based on a relative orientation of said first end to said second end; a set of sensors adapted to receive said phase shifted light signal and generate a set of output signals based on the phase shifted light signal; and
,a neural network adapted to receive said set of output signals and generate a torque output signal representative of said relative orientation of said first end of the first member to said second end of the first member. - View Dependent Claims (18)
-
-
19. A torque sensor apparatus comprising:
-
a light source for generating a first light signal; a photoelastic member having a first portion adapted to engage a first operatively associated external drive member, having a second portion adapted to engage a second operatively associated external drive member, and having a body portion connecting said first portion to said second portion, the body portion being adapted to receive said first light signal and generate a phase shifted light signal based on a relative orientation of said first portion to said second portion; a set of sensors adapted to receive said phase shifted light signal and generate a set of output signals based on the phase shifted light signal; and
,a neural network adapted to receive said set of output signals and generate a torque output signal representative of said relative orientation of said first portion of the photoelastic member to said second portion of the photoelastic member.
-
-
20. A method of determining torque between a drive member and a driven member comprising the steps of:
-
providing a first torque sensor element formed of a photoelastic material; fixing a first portion of said first torque sensor element to the drive member; fixing a second portion of said first torque sensor element to the driven member; flexing the photoelastic material into a first flexed orientation by moving said first portion and said second portion of the first torque sensor element relative to one another based on a movement of said drive member relative to said driven member; passing a first light signal through said photoelastic material disposed in said first flexed orientation, the first light signal exiting the photoelastic material at a plurality of points on the surface of the photoelastic material as a first modified light signal; and
,in a neural network based on said first modified light signal, generating a first torque output signal representative of a first degree of torque in said photoelastic material disposed in said first orientation. - View Dependent Claims (21, 22, 23, 24)
-
Specification