Gated pipelined folding ADC based low power sensor for large-scale radiometric partial discharge monitoring

David W. Upton, Richard P. Haigh, Peter J. Mather, Pavlos I. Lazaridis, Keyur Mistry, Zaharias D. Zaharis, Christos Tachtatzis, Robert C. Atkinson

Research output: Contribution to journalArticle

3 Downloads (Pure)

Abstract

Partial discharge is a well-established metric for condition assessment of high-voltage plant equipment. Traditional techniques for partial discharge detection involve physical connection of sensors to the device under observation, limiting sensors to monitoring of individual apparatus, and therefore, limiting coverage. Wireless measurement provides an attractive low-cost alternative. The measurement of the radiometric signal propagated from a partial discharge source allows for multiple plant items to be observed by a single sensor, without any physical connection to the plant. Moreover, the implementation of a large-scale wireless sensor network for radiometric monitoring facilitates a simple approach to high voltage fault diagnostics. However, accurate measurement typically requires fast data conversion rates to ensure accurate measurement of faults. The use of high-speed conversion requires continuous high-power dissipation, degrading sensor efficiency and increasing cost and complexity. Thus, we propose a radiometric sensor which utilizes a gated, pipelined, sample-and-hold based folding analogue-todigital converter structure that only samples when a signal is received, reducing the power consumption and increasing the efficiency of the sensor. A proof of concept circuit has been developed using discrete components to evaluate the performance and power consumption of the system.
Original languageEnglish
Number of pages10
JournalIEEE Sensors Journal
Early online date23 Mar 2020
DOIs
Publication statusE-pub ahead of print - 23 Mar 2020

    Fingerprint

Keywords

  • analog-digital conversion
  • partial discharge measurement
  • radiometers
  • UHF measurements
  • wireless sensor network

Cite this