Session: Government Agency Student Posters

Paper Number: 173037

Wireless Real-Time Microplastic Monitoring System 

Microplastics are a relatively new contaminant in the environment, and they have been shown to cause significant impacts, such as ingestion by marine organisms, disruption of food chains, and potential accumulation in human tissues.  There is a need to gather data on their concentrations in different areas, to help determine areas of high concentrations and how the environment is impacted there compared to areas of low concentrations.  To help gather the needed data to determine this impact, monitoring systems are necessary.  In this project, a real-time monitoring system is proposed and developed using a Raspberry Pi with various sensors.  The Raspberry Pi acts as an edge device and has a Sense Hat and a SDS011 sensor connected, to collect information on temperature, humidity, air pressure, PM 2.5, and PM 10 data.  It sends this information, as well as the GPS location and time the data was collected, to Google Cloud, which is then collected by a main node for processing and displaying on a map.  This map receives real-time data updates from the different locations that the edge devices are located, displaying the locations and the data collected through points, as well as displaying the particulate matter concentrations through a heatmap.  This creates an interconnected system of data collection, allowing data to be collected in many locations where WiFi is available.  If WiFi was not available, a Zigbee device could be used instead, where one Raspberry Pi uses the Zigbee device to send the data it collects over to another Raspberry Pi that has access to the internet for uploading to Google Cloud, allowing data to be collected in more of a variety of areas.  Furthermore, this system uses connections and correlations between PM 2.5 and PM 10 data and microplastics for monitoring the microplastic concentrations, which provides a solution that avoids microplastic sensors that are neither commercially nor academically available.  In addition, the following work in this project involves integrating artificial intelligence on the edge devices themselves, where Edge AI will be used to train a model and accurately predicate the microplastics concentrations based on all collected temperature, humidity, air pressure, PM 2.5, and PM 10 data to improve the performance of the system.  This model, along with the microplastic monitoring devices, can be used to determine areas of higher microplastic concentrations, which can aid the overall goal of viewing the impact that microplastics have on the environment and factors that may cause this increase.

Presenting Author: Madeline Potter University of South Alabama

Presenting Author Biography: Madeline Potter is a senior at the University of South Alabama, majoring in computer engineering. She has participated in many research projects in the field of Edge AI, and has plans to continue this research into a graduate degree. Outside of research, she is active in many extracurricular activities on campus, including being an Engineering Ambassador for her university and president of the Theta Lambda IEEE-HKN chapter.

Authors:

Madeline Potter University of South Alabama
Jinhui Wang University of South Alabama
Shenghua Wu University of South Alabama