SPIE Startup Challenge 2015 Founding Partner - JENOPTIK Get updates from SPIE Newsroom
  • Newsroom Home
  • Astronomy
  • Biomedical Optics & Medical Imaging
  • Defense & Security
  • Electronic Imaging & Signal Processing
  • Illumination & Displays
  • Lasers & Sources
  • Micro/Nano Lithography
  • Nanotechnology
  • Optical Design & Engineering
  • Optoelectronics & Communications
  • Remote Sensing
  • Sensing & Measurement
  • Solar & Alternative Energy
  • Sign up for Newsroom E-Alerts
  • Information for:

SPIE Photonics West 2017 | Register Today

SPIE Defense + Commercial Sensing 2017 | Call for Papers

Get Down (loaded) - SPIE Journals OPEN ACCESS


Print PageEmail PageView PDF

Sensing & Measurement

High-resolution distributed fiber-optic sensing for dynamic structural monitoring

Scanning laser interferometry with high-performance digital signal processors can measure distributed strain and characterize structural dynamics with millimeter resolution and at hundreds of Hertz.
14 June 2013, SPIE Newsroom. DOI: 10.1117/2.1201305.004826

First-generation fiber-optic sensing instruments had scan and data transmission times of several seconds.1–3 By eliminating data processing bottlenecks we have dramatically decreased the scan time and latency while maintaining high sensitivity and spatial resolution. Decreased scan time helps us accurately track fast-moving mechanical responses to an abrupt stimulus such as an impact. When the sensor output is used as feedback to control a structure (for example, when using fiber-optic shape sensor data to remotely guide a catheter to a desired location in the body), low latency allows the control system to respond more quickly and precisely to abrupt environmental stimuli or user inputs.

Taken in combination with optical fiber's light weight, small size, easy installation onto (or within) complex parts, and high-resolution continuous sensing capability, these higher acquisition rates open up a wide range of dynamic sensing opportunities. Potential applications include monitoring the mechanical response to changing loading forces as well as detecting and tracking structural defects in wind turbine blades, aircraft wings and fuselages, and composite pressure vessels. Luna also uses this technology to monitor strain in multi-core optical fibers in order to compute shape and track position in minimally invasive surgical devices.

The scanning-laser interferometric technique measures the sensor's backscattering amplitude and phase versus distance down the fiber with micron-level spatial resolution. The reflected pattern from Rayleigh backscattering or Fiber Bragg Gratings (FBGs) can be analyzed to determine local changes in optical phase induced by a strain or temperature change along the fiber. Rayleigh backscattering in optical fiber is caused by inhomogeneities along the fiber length and is intrinsic to every optical fiber. FBGs are made up of periodic index-of-refraction variations, intentionally written onto the core of the optical fiber.

We moved complex computations onto digital signal processors onboard the scanning laser interferometer, took advantage of high-speed PCI (peripheral component interconnect) Express bus transfer to a graphics processor unit, and used a high-performance laser built to our specifications. These improvements allowed us to demonstrate up to 500Hz acquisition rates with one-clock-cycle latency and spatial resolution down to 2mm in the experiment.

To demonstrate both high speed and high spatial resolution capabilities, we adhered sensing fiber using strain gage epoxy to two golf club shafts and recorded strain measurements during a swing. We used a Luna ODiSI B system to record the data from Rayleigh backscattering from Club A, routing the sensor fiber straight along the side and then back along the trailing edge of the club shaft. The system produced strain data with 5mm spatial resolution at 250Hz over a spatial range of 2m. A single scan of the strain profile along the side edge of the club at the end of a downswing is shown in Figure 1. Here the strain is proportional to the shaft curvature and can be used to calculate shaft shape. The largest strain response is near the grip because of the shaft taper and higher rigidity near the head. Strain vs. time for a point near the club grip is shown in Figure 2: strain evolution tells us how the shaft shape varies over the course of the swing. In this data series the ball was not struck.

Figure 1. Strain spatial distribution on club A with the sensing fiber mounted in a straight line along the side of the golf club shaft. Strain is read out from Rayleigh scattering at 250Hz; trace is taken at the bottom of the front swing.

Figure 2. Strain time distribution for a spot near the grip of Golf Club A where stain due to curvature is highest. Local curvature changes dramatically throughout the player's swing.

To Club B we attached fiber containing FBGs with a peak reflectivity of −40dB, 10mm individual grating length, and a 0.1mm gap between gratings. The fiber was spiraled around the shaft at 40 turns/m, and thus was sensitive to shaft bending in any direction. Torque applied to the shaft also gave a clear strain response that indicated the torque direction and magnitude. We interrogated the FBGs on club B with a Luna HSDSS 8600, operating at 500Hz acquisition rate with 2mm spatial resolution and 1.6m spatial range. The strain distribution down the shaft near the bottom of a swing is shown in Figure 3. Here the shaft bending produces a sinusoidal signal with a period equal to the spiral period. The linear offset of the sinusoid is the response to the torque applied to the shaft from the club head as it accelerates towards the ball. Thus a single fiber can provide information on the distributed curvature magnitude and orientation as well as the distributed twist on the shaft due to torque.

Figure 3. Strain spatial distribution on club B with the sensing fiber mounted in a spiral around the golf club shaft. Strain is read out from Fiber Bragg Gratings (FBGs) at 500Hz; trace is taken during the front swing.

Strain vs. time for a point near the head of club B is shown in Figure 4. This part of the shaft exhibits little bending so most of the strain is due to torque. Positive strain on the downswing indicates that the head is being accelerated toward the ball. The positive impulse when the ball is struck indicates that momentum is being transferred from the club head to the ball. Ringing in the strain immediately after impact indicates that energy is imparted into the torsional resonance of the shaft. The swing paired with the strain profile can be seen online (see video4).

Figure 4. Strain time distribution for a spot near the head of golf club B where stain due to torque is highest. Local torques change dramatically throughout the player's swing, especially when the ball is hit.

Both Rayleigh scattering and grating based sensors perform well in dynamic environments with new instruments designed to eliminate data-processing bottlenecks. A key advantage of Rayleigh scattering is that the sensing element is ‘free’: every fiber exhibits it, so the sensor is inexpensive and it can be used for sensing in existing installations. Semi-continuous FBG sensors are more expensive but can have a data rate/resolution/noise performance benefit. These high-speed distributed strain measurements are valuable for dynamic structural monitoring and testing in civil, automotive, and aerospace industries, and are useful as feedback for automation and control loops in various manufacturing processes and medical applications. Our future work will concentrate on reducing the instrument's weight and power requirements to support in-flight and remote-site testing, as well as increasing its measurement range and speed.

A detailed version of this article was presented at the 2013 Defense, Security and Sensing Conferences in Baltimore, MD.

Stephen Kreger, Alex Sang, Naman Garg, Julia Michel
Luna Innovations
Blacksburg, VA

1. B. Soller, D. Gifford, M. Wolfe, M. Froggatt, High resolution optical frequency domain reflectometry for characterization of components and assemblies, Opt. Express 13, p. 666-674, 2005. doi:10.1364/OPEX.13.000666
2. D. Gifford, B. Soller, M. Wolfe, M. Froggatt, Distributed fiber-optic temperature sensing using Rayleigh backscatter, ECOC Conf. Proc. 3, p. 511-512, 2005. doi:10.1049/cp:20050584
3. D. Gifford, S. Kreger, A. Sang, M. Froggatt, R. Duncan, M. Wolfe, B. Soller, Swept-wavelength interferometric interrogation of fiber Rayleigh scatter for distributed sensing applications, Proc. SPIE 6770 , p. 67700F, 2007. doi:10.1117/12.734931