Data-driven modeling to optimize the injection well placement for waterflooding in heterogeneous reservoirs applying artificial neural networks and reducing observation cost

Secondary recovery methods such as waterflooding are often applied to depleted reservoirs for enhancing oil and gas production. Given that a large number of discretized elements are required in the numerical simulations of heterogeneous reservoirs, it is not feasible to run multiple full-physics simulations. In this regard, we propose a data-driven modeling approach to efficiently predict the hydrocarbon production and greatly reduce the computational and observation cost in such problems. We predict the fluid productions as a function of heterogeneity and injection well placement by applying artificial neural network with small number of training dataset, which are obtained with full-physics simulation models. To improve the accuracy of predictions, we utilize well data at producer and injector to achieve economic and efficient prediction without requiring any geological information on reservoir. The suggested artificial neural network modeling approach only utilizing well data enables the efficient decision making with reduced computational and observation cost.

[1]  John E. McCray,et al.  A quantitative methodology to assess the risks to human health from CO2 leakage into groundwater , 2010 .

[2]  Vassilis Gaganis,et al.  Machine Learning Methods to Speed up Compositional Reservoir Simulation , 2012 .

[3]  J. Bruyelle,et al.  Uncertainty Assessment in Production Forecast with an Optimal Artificial Neural Network , 2017 .

[4]  J. Ramon,et al.  Machine learning techniques to examine large patient databases. , 2009, Best practice & research. Clinical anaesthesiology.

[5]  Trevor Hastie,et al.  An Introduction to Statistical Learning , 2013, Springer Texts in Statistics.

[6]  Larry W. Lake,et al.  Fast evaluation of well placements in heterogeneous reservoir models using machine learning , 2018 .

[7]  Pierre Baldi,et al.  Learning Activation Functions to Improve Deep Neural Networks , 2014, ICLR.

[8]  Bernard Widrow,et al.  A comparison of adaptive algorithms based on the methods of steepest descent and random search , 1976 .

[9]  R. W. Wedderburn.  Quasi-likelihood functions, generalized linear models, and the Gauss-Newton method , 1974 .

[10]  K. Pruess,et al.  TOUGH2 User's Guide Version 2 , 1999 .

[11]  Guigang Zhang,et al.  Deep Learning , 2016, Int. J. Semantic Comput..

[12]  Qiang Sun,et al.  Forecasting Well Performance in a Discontinuous Tight Oil Reservoir Using Artificial Neural Networks , 2013 .

[13]  T.,et al.  Training Feedforward Networks with the Marquardt Algorithm , 2004 .

[14]  Stefan Finsterle,et al.  Estimation of Landfill Gas Generation Rate and Gas Permeability Field of Refuse Using Inverse Modeling , 2011 .

[15]  Anil K. Jain,et al.  Artificial Neural Networks: A Tutorial , 1996, Computer.

[16]  George Cybenko,et al.  Approximation by superpositions of a sigmoidal function , 1989, Math. Control. Signals Syst..

[17]  J. J. Moré,et al.  Levenberg--Marquardt algorithm: implementation and theory , 1977 .

[18]  Heaton T. Jeff,et al.  Introduction to Neural Networks with Java , 2005 .

[19]  Geoffrey E. Hinton,et al.  Deep Learning , 2015, Nature.

[20]  Bill Fulkerson,et al.  Machine Learning, Neural and Statistical Classification , 1995 .

[21]  Bernhard Schölkopf,et al.  A tutorial on support vector regression , 2004, Stat. Comput..