Adaptive Water Sensor Signal Processing: Experimental Results and Implications for Online Contaminant Warning Systems

A real-time event adaptive detection, identification and warning (READiw) system has two principal functions. First, signal treatment using adaptive algorithms reduces background noise and enhances contaminant signals, leading to accurate detection of water quality changes of as low as 1%. Second, its forensic classification technique relates changes of water quality parameters to the reactivity of contaminants and hereby their chemical classes. To test these detection functionalities, contaminant transport experiments in a pilot-scale single pass pipe were conducted for 16 herbicides and pesticide, inorganic and biological contaminants. Sensor outputs (free and total chlorine, chloride, pH, DO, conductivity, ORP, and turbidity) were analyzed with the adaptive procedures. The results show unique changes of water quality parameters and the reactivity differences among the tested contaminants, based on which an effective READiw system can be configured.