Innovation in Automation: Anomaly detection in in-situ water quality sensor data

We are currently going through a step-change in the way that measure water quality in streams and rivers. Inexpensive in-situ sensors are increasingly being used, which dramatically increases the spatial and temporal density of streams data. However, only a subset of water quality variables can be measured with inexpensive sensors at this stage and these sensors are relatively inaccurate and prone to sensor drift due to biofouling or battery failure. This goal of this project is to develop automated algorithms for detecting anomalies in multivariate water-quality data collected using in-situ sensors in near-real time.

 

 

 

Funding Body: Queensland Department of Environment and Science

 

Collaborators:

Omar Alsibai, QUT, ACEMS, IFE
Dr Catherine Leigh, QUT, ACEMS, IFE

DProf Kerrie MengersenQUT & ACEMS
Assoc Prof James McGree, QUT
Prof Kate Smith-Miles, University of Melbourne
Prof Rob HyndmanMonash University & ACEMS
Dr Sevvandi Kandanaarachchi, Monash University & ACEMS
Dilini Thalagala, Monash University
Prof Bronwyn Harch, QUT & IFE

Top
%d bloggers like this: