Treating Uncertainties in Reservoir Performance Prediction with Neural Networks

2005 ◽  
Author(s):  
Johann Peter Lechner ◽  
Georg Zangl
Keyword(s):  
Neural Networks ◽  
Performance Prediction ◽  
Reservoir Performance

scholarly journals Reservoir Performance Prediction using Integrated Production Modelling (MBAL Software)

2019 ◽  
Vol 8 (4) ◽  
pp. 1484-1489
Keyword(s):  
Performance Prediction ◽  
History Matching ◽  
Material Balance ◽  
Development Planning ◽  
Reservoir Rock ◽  
Rock Properties ◽  
Production Operation ◽  
Gas Field ◽  
Reservoir Performance ◽  
Fluid Properties

Reservoir performance prediction is important aspect of the oil & gas field development planning and reserves estimation which depicts the behavior of the reservoir in the future. Reservoir production success is dependent on precise illustration of reservoir rock properties, reservoir fluid properties, rock-fluid properties and reservoir flow performance. Petroleum engineers must have sound knowledge of the reservoir attributes, production operation optimization and more significant, to develop an analytical model that will adequately describe the physical processes which take place in the reservoir. Reservoir performance prediction based on material balance equation which is described by Several Authors such as Muskat, Craft and Hawkins, Tarner’s, Havlena & odeh, Tracy’s and Schilthuis. This paper compares estimation of reserve using dynamic simulation in MBAL software and predictive material balance method after history matching of both of this model. Results from this paper shows functionality of MBAL in terms of history matching and performance prediction. This paper objective is to set up the basic reservoir model, various models and algorithms for each technique are presented and validated with the case studies. Field data collected related to PVT analysis, Production and well data for quality check based on determining inconsistencies between data and physical reality with the help of correlations. Further this paper shows history matching to match original oil in place and aquifer size. In the end conclusion obtained from different plots between various parameters reflect the result in history match data, simulation result and Future performance of the reservoir system and observation of these results represent similar simulation and future prediction plots result.

Learning performance prediction via convolutional GRU and explainable neural networks in e-learning environments

Computing ◽  
2019 ◽  
Vol 101 (6) ◽  
pp. 587-604 ◽  
Author(s):  
Xizhe Wang ◽  
Pengze Wu ◽  
Guang Liu ◽  
Qionghao Huang ◽  
Xiaoling Hu ◽  
...  
Keyword(s):  
Neural Networks ◽  
Learning Environments ◽  
Performance Prediction ◽  
Learning Performance ◽  
E Learning
Sign in / Sign up

Export Citation Format

Share Document