10th International Aerosol Conference September 2 - September 7, 2018 America's Center Convention Complex St. Louis, Missouri, USA
Abstract View
Intelligent PM2.5 Sensor Network Experience in Taiwan’s Campus and Industrial Park
HSUNLING BAI, Chungsying Lu, Shie-Yuan Wang, Wen-Chih Peng, Chun-Chia Hsu, Sihyu Liou, Chienchiao Hung, Yen-Chi Huang, Peiyu Lu, Wei So Sun, National Chiao Tung University
Abstract Number: 616 Working Group: Low-Cost and Portable Sensors
Abstract This study presents the collaborate work of a cross-domain research through environmental, cloud computing and big data analysis to create a new generation of air quality monitoring systems using low cost sensors. The intelligent sensing experience of ambient PM2.5 monitoring in a university campus and an industrial park was reported. The Plantower PMS7003 PM2.5 sensor (brief as "G7 sensor" hereafter) was selected in the intelligent sensor network because it outperformed other tested sensors in terms of precision and accuracy. After lab calibration, the G7 sensors were installed in the NCTU campus to construct a small PM2.5 sensing network via the Internet of Thing (IoT) technology. A small cloud system that can store and provide PM2.5 sensed data has also been built. The data loss rates during transmission were evaluated under different circumstances. And several data mining methodologies have been approached to simulate and predict the sensed data for the purposes of identifying outliers and pollution sources as well. Further field experience was then gained by building another intelligent sensing network in an industrial park. Field calibration of the PM2.5 low cost sensors with the FEM instrumentation data obtained from nearby air quality monitoring stations was established. And the field calibration results were compared with the original lab calibration results.