Although methane emissions from Alberta’s oil and gas sector have decreased in recent years, monitoring these emissions using Continuous Emission Monitoring Systems (CEMS) can be costly. Predictive Emissions Monitoring Systems (PEMS), powered by machine learning, offer an alternative to or can supplement CEMS. However, effective machine learning models for methane emissions prediction rely heavily on the amount of training data. To address this, we compare the prediction performance of different neural network models, including Long Short-Term Memory (LSTM), Stacked LSTM, Gated Recurrent Unit (GRU), and Bidirectional LSTM (BiLSTM), using varying time intervals for training of methane concentration data from Alberta airshed stations. The results showed that the GRU model performed better with shorter datasets, whereas the LSTM and Stacked LSTM models outperformed the GRU and BiLSTM models when trained with more historical data. However, the study found that more training data did not necessarily result in significantly better prediction models.
We experimentally study the effects of normal stress differences in the immiscible radial viscous fingering instability in a Hele–Shaw cell. Dilute low molecular weight poly(ethylene oxide) (PEO) solutions are used as the displaced fluid to focus on elastic effects without shear-thinning behavior. The molecular weight of PEO is varied to investigate the effects of normal stress differences. The experimental observations reveal that nonmonotonic and opposing effects are evident depending on the molecular weight of the PEO and the stage of the radial viscous fingering evolution. Decreases in the PEO molecular weight reduce the number of fingers and widen the finger width in the early stage. However, the increase in the PEO molecular weight promotes tip splitting and narrows finger width in the early stage but suppresses tip splitting in the intermediate stage. Weissenberg numbers are estimated at different stages of radial viscous fingering instabilities. Tip splitting occurs at the highest Weissenberg number covered in this study and suppression of tip splitting is observed at intermediate Weissenberg numbers. At low Weissenberg numbers, we observe an increased finger width and a reduced number of fingers.