Water quality change detection: multivariate algorithms

Katherine A. Klise; Sean A. McKenna

doi:10.1117/12.665019

9 May 2006 Water quality change detection: multivariate algorithms

Katherine A. Klise, Sean A. McKenna

Proceedings Volume 6203, Optics and Photonics in Global Homeland Security II; 62030J (2006) https://doi.org/10.1117/12.665019
Event: Defense and Security Symposium, 2006, Orlando (Kissimmee), Florida, United States

Abstract

In light of growing concern over the safety and security of our nation's drinking water, increased attention has been focused on advanced monitoring of water distribution systems. The key to these advanced monitoring systems lies in the combination of real time data and robust statistical analysis. Currently available data streams from sensors provide near real time information on water quality. Combining these data streams with change detection algorithms, this project aims to develop automated monitoring techniques that will classify real time data and denote anomalous water types. Here, water quality data in 1 hour increments over 3000 hours at 4 locations are used to test multivariate algorithms to detect anomalous water quality events. The algorithms use all available water quality sensors to measure deviation from expected water quality. Simulated anomalous water quality events are added to the measured data to test three approaches to measure this deviation. These approaches include multivariate distance measures to 1) the previous observation, 2) the closest observation in multivariate space, and 3) the closest cluster of previous water quality observations. Clusters are established using kmeans classification. Each approach uses a moving window of previous water quality measurements to classify the current measurement as normal or anomalous. Receiver Operating Characteristic (ROC) curves test the ability of each approach to discriminate between normal and anomalous water quality using a variety of thresholds and simulated anomalous events. These analyses result in a better understanding of the deviation from normal water quality that is necessary to sound an alarm.

Citation Download Citation

Katherine A. Klise and Sean A. McKenna "Water quality change detection: multivariate algorithms", Proc. SPIE 6203, Optics and Photonics in Global Homeland Security II, 62030J (9 May 2006); https://doi.org/10.1117/12.665019

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available