CEM India - Abstract

Abstract Information :

Conventional air pollution monitoring at a single location involves measurements by a suite of sensors having different technologies from different manufacturers - integrated and housed in a rather bulky shipping container. The air sample to be monitored is generally sucked in from a particular location by means of a pump, & routed into different chambers through conduits connected to the suite of sensors. The air sub-samples are then analyzed by different sensors to give in-vitro estimates of the different parameters constituting the air quality. Therefore, monitoring of air pollution at a single location with the disparate sensors of varying sensitivities, accuracies and temporal responses, not only pose significant challenges in data acquisition and assimilation, but also, involves significantly high costs in order to arrive at digestible information for researchers, policy makers as well as the common public. Against this backdrop, we present an innovative photonic system capable of real time remote monitoring of various air parameters simultaneously, to arrive at the in-situ air quality at a particular location, or the spatial patterns (contours) of air quality covering a substantial Euclidian space - thus enabling eulerian as well as lagrangian measurements having multifarious advantages. The photonic system was extensively evaluated in the laboratory as well as in the field, and was found to be good; yielding air quality estimates at very high sampling frequencies with higher sensitivity and accuracy. The photonic system was designed and developed using COTS (commercially-off-the-shelf) technologies, thus making it significantly cheaper for wider deployment, in consonance with WHO’s roadmap. The uniqueness and novelty of the system lies in its ability to innovatively apply the concepts of laser back scattering, artificial intelligence and machine (deep) learning to identify, classify and quantify various air pollutants simultaneously.