Self-learning fiber processing system and method
Assignee3SAE Technologies, Inc.
InventorsRobert G. Wiley, John Lower, Brett Clark
Life of this patent
Estimating the term…
Abstract
Provided is a system for and a method of processing an optical fiber, such as tapering an optical fiber. The method includes receiving fiber parameters defining characteristics of an optical fiber, modeling an idealized fiber based on the fiber parameters to establish modeled data, and establishing processing parameters. A processing operation is performed on the optical fiber according to the processing parameters to produce a resultant fiber. Aspects of the resultant fiber are measured to establish measured data. The measured data and the modeled data are normalized to a common axis and a difference between the two is determined. The processing parameters are adjusted based on the differences.
Claims20
A self-learning fiber tapering system, comprising:
- at least one processor, a computer memory, at least one heat source, at least one fiber holder, and one or more sensors;
- and
- fiber tapering program code executable by the at least one processor to:
- process received fiber parameters to define taper characteristics in the form of physical dimensions of a desired resultant fiber taper of an optical fiber;
- model and store a shape of the desired resultant fiber taper based on the received fiber parameters to establish modeled data;
- establish processing parameters from the received fiber parameters useful to control the at least one fiber holder and/or the at least one heat source in a tapering operation of the optical fiber to produce a resultant fiber taper;
- use the one or more sensors to measure physical dimensions of the resultant fiber taper to establish measured data, the measured data including fiber diameter over a tapered length of the optical fiber;
- determine differences in physical dimensions between the measured data and the modeled data;
- and
- adjust one or more of the processing parameters to form adjusted processing parameters useful by the processor to reduce the differences in physical dimension between the modeled data and measured data of a subsequent resultant fiber taper.
A self-learning optical fiber tapering apparatus comprising:
- at least one processor, a computer memory, at least one heat source, at least one fiber holder, and one or more sensors;
- and
- fiber tapering program code executable by the at least one processor to execute a fiber tapering method comprising:
- receiving tapering parameters defining characteristics of an optical fiber taper;
- computer-modeling an idealized fiber taper based on the fiber parameters to establish modeled data;
- from the tapering parameters, establishing fiber processing parameters for controlling aspects of the fiber tapering machine;
- performing a tapering operation on the optical fiber according to the processing parameters to produce a first resultant fiber taper;
- measuring aspects of the first resultant fiber taper to establish measured data, including collecting fiber diameter over length data in real time as the first resultant fiber taper is created using one or more sensors;
- determining differences between the first resultant fiber taper and the idealized fiber taper, including:
- from the modeled data, determining an idealized fiber taper diameter per unit time;
- from the measured data, determining a resultant fiber taper diameter per unit time;
- determining a difference between the idealized taper diameter and the resultant taper diameter per unit time;
- and
- determining at least one taper ratio based on the difference between the resultant taper diameter and the idealized taper diameter at one or more times;
- adjusting one or more of the processing parameters according to the at least one taper ratio;
- and
- performing a next tapering operation using the adjusted processing parameters to form a second resultant fiber taper of the optical fiber or a resultant fiber taper on a different optical fiber.
Dependencies are inferred from the claim text.