City Research Online

Abstract

In nature, sensing critical flow events or flow signatures is often achieved through a coupled interaction between a fluid and arrays of slender flexible beams, such as whiskers and wind-hairs. Behavioural research on live sea lions has shown that they are able to discern the direction of oncoming vortices, even when they impact the animal’s whiskers contralaterally. These experiments show that the whiskers coupled with the animal’s neural processing can detect the direction of arrival of a flow event, just using information from the whiskers on one side of the head. Therefore, it is hypothesized that in highly noisy environments, important information is gained from the time differences between whisker stimulation. Herein, numerous iterations of bio-mimetic sensors, array designs, and the decoding of the relationship between the sensors are investigated.

A model sea lion head with slender plastic optical fibre whiskers was subjected to a mean flow with overlaid turbulent structures generated in the wake of a cylinder. A motion tracking camera observed the array of fibres to track the bending deformations of the fibres. The characteristic signature of the passage of a vortex core, a jerk-like motion of the whisker, is apparent in the leading sensor, and in the response of subsequent sensors, even in the wake of the first whisker. Cross-correlation of the time domain of the bending signal from pairs of whiskers proves that the detection of vortices and their passage along the animal’s head is possible even in noisy environments.

Following these observations, a theoretical model for tracking the path of an unknown flow disturbance is devised, adapted from multilateration principles used for electromagnetic and acoustic source localisation. The model focuses on tracking a signal of unknown size, speed, and direction as it passes the array using the time signatures and known distances between triplets of disturbed whiskers. The model is tested against experimental studies on the model seal head, and a simple 2D array with regular spacing. The results validate the principle behind the model, and highlight the sensitivity of the model to the array layout and spacing. An optimised array was produced based on these findings, with tuned array spacing, sensor length, and a new layout minimising co-linearity between the sensors. Further tests validated the efficacy of the design with a mean output angle RMSE of 1.9◦.

Finally, a new whisker sensor design is presented, using the enhanced sensitivity and performance of fibre Bragg Grating (FBG) engraved optical stress sensors. Its performance is evaluated against the original sensors, with the added benefit of removing the tracking camera from the process. The performance and improved sensitivity of these FBG-based fibre-optic whiskers is shown to produce reliable
direction of arrival and velocity estimations, at a typical SNR of around 2dB. The FBG sensors correct for disruptive variations in temperature using a second grating at the whisker tip, positioned at the point with the least bending stress, allowing for isolated bending stress measurements and identification of the hydrodynamic disturbances of interest.

The study herein further develops our understanding of flow sensing systems and how they can overcome high background noise levels with the application of the correlation principle. Further, the study explores array design principles and sensitivities, covering several methods to improve the signal output that have been trialled and shown to improve the performance of the array, with further recommendations made for various applications. Lastly a new sensor design is presented, with the objective of broadening deployment options while maximising sensitivity, expanding the range of suitable applications for whisker sensors.

Publication Type:	Thesis (Doctoral)
Subjects:	Q Science > QC Physics T Technology > T Technology (General) T Technology > TA Engineering (General). Civil engineering (General)
Departments:	School of Science & Technology > Engineering School of Science & Technology > School of Science & Technology Doctoral Theses Doctoral Theses

Export

Downloads