Asset Details
Do you wish to reserve the book?
Exploring the predictive performance of deep learning for fracturing fluid flowback and shale gas production
Hey, we have placed the reservation for you!
By the way, why not check out events that you can attend while you pick your title.
Oops! Something went wrong.
Looks like we were not able to place the reservation. Kindly try again later.
Are you sure you want to remove the book from the shelf?
Exploring the predictive performance of deep learning for fracturing fluid flowback and shale gas production
Do you wish to request the book?
Exploring the predictive performance of deep learning for fracturing fluid flowback and shale gas production
Please be aware that the book you have requested cannot be checked out. If you would like to checkout this book, you can reserve another copy
We have requested the book for you!
Your request is successful and it will be processed during the Library working hours. Please check the status of your request in My Requests.
Oops! Something went wrong.
Looks like we were not able to place your request. Kindly try again later.
Journal Article
Exploring the predictive performance of deep learning for fracturing fluid flowback and shale gas production
2025
Overview
The big data system has been developed to optimally combine numerical-model predictions with actual measurements from the gas shale play to create the best estimates of current shale gas conditions and their uncertainties, improving our ability to forecast and understand the shale gas production variations. However, considering the hydraulic fracturing fluid flowback in predicting shale gas production introduces new challenges. For example, complexities in the flow of gas-water two-phase fluid defy traditional numerical simulation, while flowback are complicated by strong environmental disruptions, and shale gas production encounters substantial noise interference. Here, we developed CNN-Transformer, a production and fluid flowback predicted system using deep learning, by integrating a convolutional neural network (CNN) and a Transformer network. CNN’s receptive field focus irregular observational data on the local region correlation of the sequence, while Transformer network extracts the state information of historical production and flowback volumes from previous time frames, and predicts future time states, aiming to capture temporal patterns. We contrast the performance of CNN-Transformer, CNN-LSTM, CNN-GRU-AM models in predicting shale gas production, flowback volume and the relationship between the two. We show that the CNN-Transformer outperforms the other models, with R²=0.644 and RMSE = 0.1424, compared to CNN-GRU-AM (R²=0.6068, RMSE = 0.1513) and CNN-LSTM (R²=0.5727, RMSE = 0.1618) in predicting flowback volume. We conclude that CNN-Transformer (R
2
= 0.72, RMSE = 0.3824) markedly reduces analysis error of the shale gas production, outperforming both CNN-LSTM (R
2
= 0.4911, RMSE = 0.44912) and CNN-GRU-AM (R
2
= 0.5705, RMSE = 0.4249). We deem our results to lay a foundation for further debates on striking a balance between fracturing fluid flowback volume and prediction of shale gas production.
