Use this URL to cite or link to this record in EThOS:
Title: Applying distributed embedded electronic instrumentation to the measurement and analysis of multiple acoustic and ultrasonic signals
Author: Hopkins, Mark B.
ISNI:       0000 0004 6497 5325
Awarding Body: University of Kent
Current Institution: University of Kent
Date of Award: 2018
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Restricted access.
Access from Institution:
In many industrial applications there is a requirement to monitor acoustic and ultrasonic signals at selected locations across large structures - and leak detection in power station boilers would be a good example of such an application. Conventional topologies dictate that the analogue output signals from the acoustic and ultrasonic sensors are routed individually back to a central computer in the control room, where they are digitised and then analysed for the purposes demanded. Such topologies require long (and expensive) individual runs of specialist cables (sometimes lengths in the order of 500 meters), over which the sensor signals suffer significant degradation and exposure to interference. The central computer also requires high levels of processing bandwidth, if the sensors signals are to be processed concurrently. This thesis describes an alternative and novel approach, with the design and development of distributed electronic instrumentation (containing embedded microcontrollers), utilised to locally measure and analyse acoustic and ultrasonic signals. These instruments are mounted on the structure adjacent to each sensor location (thereby maintaining very short transducer cable lengths). They also incorporate a digital interface to the central computer via an Ethernet IP network, or a 'multidrop' Profibus system, greatly simplifying the cabling requirements. The local signal processing largely obviates the need for real time high speed communications over this interface. A dedicated serial interface is also provided to stream live audio data back to the control room, where it may be monitored by the plant operators. This analysis of the sensor signals locally greatly improves digitised signal fidelity, markedly reduces noise, and provides a powerful dedicated processing resource for every sensor. The local digital signal analysis includes various techniques working in both the time and frequency domains - such as amplitude trends, bandpass filtering, autocorrelations and Fast Fourier Transforms. High speed synchronised collection of sensor data may implemented, for subsequent 'off line' processing (for applications such as acoustic signal source location, and Acoustic Emission analysis). This work also includes the development of a very low noise charge preamplifier for piezo electric transducers, much improving the sensitivity of measurements with these sensors.
Supervisor: Lee, Peter ; Waller, Winston Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral