Fibre Optic (FO) sensors are regarded as a novel and effective tool for thermal pipe leakage detection. However, in the early stage of leakage, the spatial range of the abnormal temperature is below FO’s resolution, which leads to unrealistic temperature measurements.
In Bingyu’s research, several mathematical methods, including; weighted average, linear equation system, genetic algorithm, are adopted to tackle the mapping relationship between the FO measurements and the real temperature. Among these methods, genetic algorithm is found to be the most effective one, though not perfect. This work will help the engineers to predict the realistic leakage situation from FO sensing data.
Bingyu Zhao is currently a PhD student under the supervision of Prof. Kenichi Soga in the Geotechnical and Environmental Group at the Department of Engineering, University of Cambridge. Bingyu is a member of Churchill College and her PhD studies are funded by the Cambridge International Student Scholarship (CISS).
Discover more videos from our latest academic seminar series: CHUtalks.