Restoration of Missing Pressures in a Gas Well Using Recurrent Neural Networks with Long Short-Term Memory Cells

Abstract

This study proposes a data-driven method based on recurrent neural networks (RNNs) with long short-term memory (LSTM) cells for restoring missing pressure data from a gas production well. Pressure data recorded by gauges installed at the bottom hole and wellhead of a production well often contain abnormal or missing values as a result of gauge malfunctions, noise, outliers, and operational instability. RNNs employing LSTM cells to prevent long-term memory loss have been widely used to predict time series data. In this study, an RNN with the LSTM method was used to restore abnormal or missing wellhead and bottom-hole pressures in three intervals within a production sequence of more than eight years in duration. The pressure restoration was performed using various input features for RNNs with LSTM models based on the characteristics of the available data. It was carried out through three sequential processes and the results were acceptable with a mean absolute percentage error no more than 5.18%. The reliability of the proposed method was verified through a comparison with the results of a physical model.

Robust hybrid machine learning algorithms for gas flow rates prediction through wellhead chokes in gas condensate fields

“…Determination of multiphase flow rate is crucial in planning and adopting correct measures and reforms in production policies commensurate with the reservoir's performance during operation [15]. The back pressure applying by wellhead chokes has several advantages, such as stabilizing the multiphase flow rate [16], preventing further pressure drop at the bottom hole section and condensate drop out, avoiding to create the skin factor due to pressure drop, and preventing water coning in gas condensate reservoirs [17,18]. Numerous experimental and theoretical relationships have been introduced to estimate the multiphase flow rate through wellhead chokes.…”

Analysis of Uncertainty Trend for Estimated Ultimate Recovery Prediction of Shale Gas with Various Production Periods based on Machine Learning

“…머신러닝은 물리적 모델이 아닌 복잡 한 데이터를 학습하여 해당 시스템의 패턴을 파악하는 알 고리즘이다. 오일 및 가스개발에서는 저류층 특성변수 예 측 (Iturrar and Parra, 2014;Zerrouki et al, 2014), 암상분 류 (Silversides et al, 2015), 물리검층자료 복원 (Alizadeh et al, 2012), 저류층 히스토리 매칭 (Ahn et al, 2018;Kim et al, 2020), 생산압력 복원 (Ki et al, 2020) Latitude (°) (Bowker, 2007). 입력자료 특성 분석 다음은 RF 모델의 입력자료 구성을 위하여 Table 1의 입 력자료와 EUR과의 상관관계를 분석하였다(Fig.…”

Reservoir Production Prediction Model Based on a Stacked LSTM Network and Transfer Learning

“… 10 proposed a method using the LSTM network to establish a prediction model to predict the distribution of reservoir water saturation and oil production. Ki et al 11 proposed a data-driven method based on the LSTM network to recover the lost pressure data in gas wells. Li et al 12 used a bidirectional gated recurrent unit (Bi-GRU) and sparrow search algorithm (SSA) to improve the prediction accuracy of oil production.…”