This Campus Notice is more than 3 years old. Links and contact information may have changed.

MSc-SDE Defense: Hassan Afzaal

Posting Date(s)
Date

We are pleased to announce the upcoming Master of Science in Sustainable Design Engineering (MSc-SDE) Thesis Defense of Hassan Afzaal. 

Title: Applications of Artificial Intelligence and Deep Learning for Sustainable Water Management in Prince Edward Island

Date:  Wednesday, May 27, 1:00 to 4:00 PM


Abstract: The potato industry significantly promotes the economy of Prince Edward Island (PEI).  It contributes 10.8% to the GDP of the province with more than one billion direct and indirect economic benefits, including engaging 12% of the total work force on the island. The majority of potato production in PEI is rain-fed. Potato tubers are highly sensitive to drought conditions, which drastically reduce tuber yield if supplemental irrigation (SI) is not scheduled. This study considers rainfall patterns and evaluates the benefits of irrigation scheduling, deep learning modeling, and sustainable irrigation water management. Reference evapotranspiration was estimated with recurrent neural networks, namely long short term memory (LSTM) and Bidirectional LSTM. Four representative meteorological sites (North Cape, Summerside, Harrington and Saint Peters) were selected across PEI. Results suggested that July, August and September are relatively drier months of the year in PEI and SI may be required to meet crop water requirements. Pressurized irrigation systems including sprinkler, fertigation and drip irrigation were installed at a small-scale site to evaluate the impact of scheduled SI in offsetting the deficit in irrigation, to mimic drought conditions, as compared to conventional practice of rain-fed conditions; i.e., no irrigation practice (control). Significant potato yield differences were observed between control and irrigation methods used in this study. Furthermore, to assess the irrigation impacts on groundwater levels, three deep learning models, namely multilayer perceptron, LSTM and one dimensional convolutional neural networks were tested for two watersheds (Baltic River and Long Creek) in PEI. The results suggested that pumping impacts the groundwater levels in summer seasons. Sustainable options of SI are necessary without affecting the groundwater levels in PEI. 

The presentation and examination will be conducted via a web conference. Those who wish to "attend" should RSVP by Tuesday, May 26, with the Chair of the examination committee, Dr. Amy Hsiao, at ahsiao@upei.ca for specific details on calling-in.