Transcript – Modeling, IoT and Machine Learning for Smart Greenhouse
Howdy, I’m Gabriela Ramos and I’m Tatiana Baig and we were graduate students at the Biological and Agricultural Engineering Department representing on Iot and machine learning combined with modeling for a smart greenhouse, the objective is to develop a smart greenhouse that uses automated remote monitoring sensors to maintain optimal environmental conditions for plant growth by introducing continuous, viable bio aerosol collection systems and air flow modeling to prevent the spread of antibiotic resistant pathogens from contaminated soil and water.
In this study, an actual greenhouse in a smart greenhouse in a wind tunnel setting have been tested for bacteria aerosols at different environmental conditions. For the methods, there are seven air sampling locations inside the actual greenhouse and leave samples were collected in random locations. Bacteria, aerosols and leave samples were quantitate by plaiting and analyzed the Kearby Bauer test using eight paper disks impregnated with commonly used antibiotics to assess the antibiotic resistance shown in table one. The smart greenhouse in the wind tunnel set up had a nebulizer free Kaleigh aerosol generation, a blower to create the average wind speed of six meters per second. In Texas, a bio aerosol collector and plants exposed to the wind. Each test had a five-minute period for aerosol collection and plant exposure. With the nebulizer operating for the entire 15-minute experiment, leaves were collected from the top and bottom part of each plant in the bacterial counts on the plant. Leaves were compared to the bio aerosol collections and the actual greenhouse.
Over 50 percent of the aerosol samples exhibited resistance to antibiotics compared to the twenty five percent resistance and the bacteria from the leaf surfaces. The dominant bacteria was bacillus serious airflow model trace the best Lissa’s contamination to the air conditioner condensate from where it becomes aerosolized and spreads throughout the greenhouse and the smart greenhouse consistently.
For each part, the bottom leaves and the front had higher bacteria counts, indicating that the turbulent air flow may play a role in impaction. Current work focuses on particle tracking velocimetry testing to determine the effect of different environmental conditions on the size, distribution and pathway of aerosolized bacteria and complete in remote sensor system for the smart greenhouse. Thank you for supporting our T3 project.