Full Description

Timely deployment of contaminant warning systemsrequires on-line sensors and advancement of dataanalysis and decision support systems to accuratelydetect water quality changes. Many utilities are beginningto install on-line water quality sensors or areexpanding their current on-line monitoring capabilities.As more and more data from these systems becomeavailable, various algorithms to extract useful informationfrom them are being developed.These new algorithms must be quantitatively evaluatedin order to understand their capabilities and limitationsas well as to determine whether they provideadded value. This study was prompted by the need toevaluate the performance of new algorithms for thedetection of changes in water quality. In addition tobeing assessed on how well they identified water qualitychanges, algorithms were also assessed on how fewfalse alarms they generated.The results of this study are likely to affectapproaches to distribution system security with respectto a possible accidental or deliberate contaminationevent. The steps necessary for evaluating detectiontools are detailed using both previously collected dataand simulated events. Includes 28 references, tables, figures.

Product Details

Edition: Vol. 100 - No. 1 Published: 01/01/2008 Number of Pages: 12File Size: 1 file , 410 KB