... absolute percentage error (SMAPE) are used to quantify the prediction accuracy for both ROP and downhole shock. The performance of the transformer-based model is compared with other commonly used deep learning architectures for time series forecasting, such as the deep learning autoregressive recurrent...

...URTeC: 4042557 Physics Informed Deep Learning Models for Improving Shale and Tight Forecast Scalability and Reliability Kainan Wang1, Lichi Deng1, Yuzhe Cai1, Guido Di Federico2, Keith Ramsaran1, MunHong Hui1, Hussein Alboudwarej1, Christian Hager1, Yuguang Chen1, Xian-Huan Wen1, 1. Chevron U.S.A...

...URTeC: 4040968 A Deep Learning Workflow for Integrated Geological, Petrophysical, and Geomechanical Interpretation Vanessa Simoes*1, Atul Katole1, Bhuvaneswari Sankaranarayanan1, Tao Zhao1, Aria Abubakar1, 1. SLB. Copyright 2024, Unconventional Resources Technology Conference (URTeC) DOI 10.15530...

... forecasting scenario geologist deep learning modeling & simulation energy economics time sery transformer application transforming early-stage oil clastic rock rock type natural language generative ai history gas production forecasting houston production forecasting machine learning...

... the written consent of URTeC is prohibited. Abstract This study aims to leverage deep learning (DL) to forecast oil production dynamics, focusing on the transition from conventional to unconventional reservoirs in the Shengli Oilfield, China. Due to the discrepancy in geological characteristics...

.... neural network deep learning geology geologist complex reservoir machine learning wine-rack geometry unconventional drilling unit dsus artificial intelligence dataset activation unconventional resource technology conference geometry urtec convolutional neural network forecasting houston...

... machine learning geology production forecasting urtec workflow large language model clastic rock rock type artificial intelligence geological subdiscipline operator information efficient field development decision driven evaluation deep learning asset and portfolio management natural...

... deep learning geologist complex reservoir reservoir surveillance neural network production monitoring machine learning production control dataset depletion function prediction forecast workflow geology artificial intelligence depletion urtec sensitivity accuracy validation...

... and, consequently, production behavior (Waite, 2021). geologist structural geology sedimentary rock clastic rock unconventional resource economics reservoir simulation deep learning fluid dynamics complex reservoir unconventional play mudrock production forecasting flow in porous media neural...

... of these models through new workflows to accommodate the distinct characteristics of unconventional reservoirs. complex reservoir geology natural language forecasting deep learning machine learning geologist artificial intelligence unconventional reservoir urtec scenario accuracy transformer...

... eagle ford geologist deep learning artificial intelligence sedimentary rock machine learning complex reservoir ultimate recovery information geology reservoir surveillance nrmse production forecasting eur lstm unconventional reservoir eur prediction clastic rock rock type...

... and pattern recognition. structural geology drillstem/well testing reservoir simulation deep learning pvt measurement drillstem testing neural network geologist risk management data quality reservoir geomechanics equation of state asset and portfolio management artificial intelligence...

... fluid dynamics geologist neural network clastic rock shale gas complex reservoir artificial intelligence deep learning mape rock type dataset modeling & simulation machine learning realization production forecasting accuracy prediction permeability sedimentary rock geology...

... stimulation efficiency. Introduction Fracturing and stimulation have multiple associated challenges similar to any technical domain. This paper applies high-fidelity measurements and deep-learning techniques to address some of the key challenges faced by operators today in the fracturing domain. Fiber optic...

...URTeC: 4046687 A Deep Learning Approach to Predicting Remaining Useful Life for Downhole Drilling Sensors Using Synthetic Data Generation Aamir Bader Shah*1, Yu Wen1, Jiefu Chen1, Xuqing Wu2, Xin Fu1, 1. Electrical and Computer Engineering, University of Houston, USA, 2. Department of Information...

... recovery. geology sagd enhanced recovery geological subdiscipline optimization workflow integrating geomechanics steam-assisted gravity drainage modeling & simulation geomechanics recovery geologist algorithm optimization problem artificial intelligence convergence deep learning...

...) condition is presented. A nested optimization based on the PI-performance model is provided. In the-inner shell, the optimum fracture parameters considering time dependency are achieved by searching the maximum PI based on unified fracture design (UFD). Coupling with deep learning and UFD practices...

... the nonlinearity as well as the impact of manual operations. Production history was obtained from 3 wells of the Eagle Ford shale (Frio, La Salle, and Zavala County) where a single well data was split into a 75:25 ratio for training and testing. Deep learning (DL) algorithms, long short-term memory (LSTM) and Bi...

... with small label values (in the lower left corner of the plot) that actually represent high liquid loading risk. The top performers of the two deep learning architectures, feed-forward NNs and RNNs, were able to correct the major deficiencies of the RF and GB models in liquid loading prediction. Compared...

... in the national greenhouse-gas inventory (Zavala-Araiza et al., 2015). Both anthropogenic and natural emissions of CH 4 are likely to increase if no mitigation plans are deployed. upstream oil & gas emission neural network united states government air emission deep learning ch 4 methane...