Efficient and data-driven prediction of water breakthrough in subsurface systems using deep long short-term memory machi
- PDF / 2,076,835 Bytes
- 13 Pages / 595.276 x 790.866 pts Page_size
- 62 Downloads / 230 Views
ORIGINAL PAPER
Efficient and data-driven prediction of water breakthrough in subsurface systems using deep long short-term memory machine learning Tao Bai 1 & Pejman Tahmasebi 1 Received: 1 February 2020 / Accepted: 14 September 2020 # Springer Nature Switzerland AG 2020
Abstract Water coning is one of the common issues in subsurface systems in which water flows into the production well through perforated zones. This phenomenon can cause severe problems in wellbore and surface facilities. Thus, accurate prediction of water breakthrough can help to adapt to the production mode and avoid such issues. Conducting flow simulations, as a conventional approach, can be very time demanding if one deals with large subsurface systems. Furthermore, several types of data are often collected during the life of a subsurface system each of which can help to predict the breakthrough and water coning. As such, it is very important to produce similar results using the time sequence data gathered from various geo-sensing tools. In this paper, a deep long short-term memory (LSTM) model is developed to predict the water cut and water breakthrough time for multiple production wells in a water flooding case. The dataset is generated by the Egg model with a multi-input-multioutput system. We found that the proposed model can capture the general trend of variation for the water cut time sequence data for a complex subsurface system. To evaluate the performance of our data-driven method, the results are compared with vanilla recurrent neural network (RNN), deep gated recurrent unit (GRU), and artificial neural network (ANN). The conducted comparison indicates that the proposed deep LSTM model outperforms the other three approaches when the results are compared with the numerical data. Keywords Machine learning . Watercut . Fluid flow . Porous media . Big data
1 Introduction Generally, water-drive reservoirs are bounded and supported by aquifers. In the production process, pressure around the wellbore is decreased with the increase of production rate, which results in the migration of oil-water interface toward the perforation interval and the interface changes from its original state into a cone shape, known as water coning [1]. This phenomenon mainly depends on the pressure gradients near the production well and the balance between viscous and gravitational forces [2]. Initially, the gravitational force is dominant, which prevents water to invade the oil-saturated region. Once the well starts to produce, the viscous force can push oil into the well by a pressure drawdown. The oil-water * Pejman Tahmasebi [email protected] 1
Department of Petroleum Engineering, University of Wyoming, Laramie, WY 82071, USA
interface maintains at a certain elevation where viscous force is balanced with gravity. Increasing the production rate can break the balance and the cone will be dragged up until invading the wellbore. After water breakthrough, continuous increment of water to oil production ratio results in the reduction of ultimate recovery rate. F
Data Loading...
