Fiber specklegram sensors (FSSs) traditionally use statistical methods to analyze specklegrams obtained from fibers for sensing purposes, but can suffer from limitations such as vulnerability to noise and lack of dynamic range. In this paper we demonstrate that deep learning improves the analysis of specklegrams for sensing, which we show here for both air temperature and water immersion length measurements. Two deep neural networks (DNNs); a convolutional neural network and a multi-layer perceptron network, are used and compared to a traditional correlation technique on data obtained from a multimode fiber exposed-core fiber. The ability for the DNNs to be trained against a random noise source such as specklegram translations is also demonstrated.

An optical fiber specklegram chemical sensor based on a concatenated multimode-no-core-multimode structure is reported. The 30 mm no-core section is immersed in the analyzed liquid so the number of excited modes is modulated by the surrounding refractive index value, affecting the interference of the several propagating modes and consequently the output specklegram, whose changes are quantified by the zero mean normalized cross-correlation coefficient. The sensor was evaluated on the measurement of ethanol-water solutions, resulting in 18.7 RIU-1 sensitivity (5×10-4 resolution) for the 1.33 to 1.36 refractive index interval, besides 0.07°C-1 temperature sensitivity within the 26 to 32°C range. Moreover, the system was also tested on liquid level measurements, yielding 0.05 mm-1 sensitivity and 0.2 mm resolution. The results are comparable to previously reported multimodal interferometers based on more complicated interrogation schemes, indicating that the specklegram sensor is a feasible alternative for the sensitive and straightforward measurement of liquid samples.

The inception of photonic crystal fibers (PCFs) allowed for unprecedented tailoring of waveguide properties for specialty sensing probes. Exposed core microstructured fibers (ECFs) represent a natural evolution of the PCF design for practical liquid and gas sensing. Until now, to the best of our knowledge, only single-mode or few-modes ECFs have been explored. In this Letter, we demonstrate a highly multimode ECF with a lateral access that extends throughout the whole length of the fiber. The ECF is operated as a fiber specklegram sensor for assessing properties of fluids and interrogated using a simple and low-cost setup. The probe exhibits a refractive index resolution and sensitivity of at least 4.6 × 10 − 4 refractive index units (RIUs) and − 10.97 R I U − 1 , respectively. A maximum temperature resolution up to 0.017°C with a − 0.20 ∘ C − 1 temperature sensitivity over the 23°C–28°C range and a liquid level sensing resolution up to 0.12 mm with − 0.015 m m − 1 sensitivity over the 0.0–50.0 mm bathed the length range in water.

We present a multi-point curvature sensor based on optical fiber specklegram measurements. Apart from the current approaches, the proposed system uses an ordinary multimode fiber excited with visible light as a reflection-type probe. Besides, this method discretizes the waveguide into segments connected by joints and assumes sequential bend events, simplifying the specklegram referencing for correlation analyses and avoiding laborious deep learning processing. Sensor characterization yielded a linear response with ∼ 1.3 ∘ resolution for single curvatures, whereas shape prediction experiments in the plane resulted in maximum errors of ∼ 3.5 ∘ and ∼ 5.4 m m for angular and linear positioning, respectively. Furthermore, exploratory tests indicated errors < 2.3 ∘ regarding probe curvatures in the space. This research introduces a feasible, straightforward alternative to the available shape sensors, enabling applications in medical probes and soft robotics.

An optical fiber anemometer based on the Pitot-static tube is reported. The device was fabricated in a 3D printer and the pressure difference was assessed by means of a latex rubber diaphragm with a mounted fiber Bragg grating probe, so the membrane deflection is obtained in terms of the fiber curvature. The sensor characteristics were firstly obtained by simulation in order to determine the correlation between the Bragg wavelength shifts and the airflow speed. Next, the device was experimentally evaluated in a wind tunnel, allowing for the assessment of a 20 to 36 m/s dynamic range, with 24.9 pm/(m/s) maximum sensitivity and 0.04 m/s resolution. Finally, the system was tested on the measurement of cyclic airflow speeds for predicting the applied speed according to the calibration curve, yielding reproducible results with an average error of 0.98 m/s, which is comparable to other reported sensors based on more complex designs and interrogation setups.

Offshore oil and gas platforms present a harsh environment for their installed infrastructure, with pipelines that are subjected to both a corrosive atmosphere and transport of aggressive chemicals being the most critical. These conditions have prompted the industry to substitute metallic pipelines for composite counterparts, often made from fiber-reinforced plastics assembled with bonded joints. Various technologies have emerged in recent years to assess the health of these composite pipelines. In particular, robust speckle metrology techniques such as shearography, although not capable of long-term monitoring, have produced very satisfactory results. However, these inspection techniques require specialized equipment and trained personnel to be flown to offshore platforms, which can incur in non-trivial inspection costs. In this paper, we propose and demonstrate a robust and cost-effective approach to monitor pipeline bonded joints during assembly and operation using fiber Bragg grating (FBG) sensors embedded into the joints’ adhesive layer. This approach allows for informed decisions on when to perform targeted in-depth inspections (e.g., with shearography) based on both real-time and long-term feedback of the FBG sensors data, resulting in lower monitoring costs, a severe increase in monitoring uptime (up to full uptime), and increased operational security.