Deep-Learning-Based Production Decline Curve Analysis in the Gas Reservoir through Sequence Learning Models

Shaohua Gu,Jiabao Wang,Yuetian Liu,Mingjin Yang,Bin Tu,Liang Xue

doi:10.32604/cmes.2022.019435

Abstract

Production performance prediction of tight gas reservoirs is crucial to the estimation of ultimate recovery, which has an important impact on gas field development planning and economic evaluation. Owing to the model’s simplicity, the decline curve analysis method has been widely used to predict production performance. The advancement of deep-learning methods provides an intelligent way of analyzing production performance in tight gas reservoirs. In this paper, a sequence learning method to improve the accuracy and efficiency of tight gas production forecasting is proposed. The sequence learning methods used in production performance analysis herein include the recurrent neural network (RNN), long short-term memory (LSTM) neural network, and gated recurrent unit (GRU) neural network, and their performance in the tight gas reservoir production prediction is investigated and compared. To further improve the performance of the sequence learning method, the hyperparameters in the sequence learning methods are optimized through a particle swarm optimization algorithm, which can greatly simplify the optimization process of the neural network model in an automated manner. Results show that the optimized GRU and RNN models have more compact neural network structures than the LSTM model and that the GRU is more efficiently trained. The predictive performance of LSTM and GRU is similar, and both are better than the RNN and the decline curve analysis model and thus can be used to predict tight gas production.

Highlights

With the increasing demand for natural gas resources to empower economic development, tight gas reservoirs have become an important source of energy and have been discovered in major basins to produce oil and gas worldwide
It has been found that such a reservoir has a threshold pressure gradient [9,10]; the permeability in a tight gas reservoir may change with pressure and exhibits a stress-sensitive effect during the production process [11,12]; the proppant embedment issues in the hydraulic fractures of a tight reservoir may affect the gas production [13]; the temperature and pressure may affect the imbibition recovery for tight or shale gas reservoir [14]; the tight gas production may be seriously reduced by water blockage [15–17]; the dispersed distribution of kerogen within matrices may affect the production evaluation [18]; sulfur precipitation and reservoir pressure-sensitive effects may affect the permeability, porosity and formation pressure [19]; the micro-scale flow mechanism, such as Knudsen diffusion, slippage effect, and adsorption, can be difficult to describe quantitatively [20,21]
The results showed that the normal distribution transformation (NDT) model can significantly improve the classification accuracy of the C4.5 algorithm

Summary

Introduction

With the increasing demand for natural gas resources to empower economic development, tight gas reservoirs have become an important source of energy and have been discovered in major basins to produce oil and gas worldwide. It has been found that such a reservoir has a threshold pressure gradient [9,10]; the permeability in a tight gas reservoir may change with pressure and exhibits a stress-sensitive effect during the production process [11,12]; the proppant embedment issues in the hydraulic fractures of a tight reservoir may affect the gas production [13]; the temperature and pressure may affect the imbibition recovery for tight or shale gas reservoir [14]; the tight gas production may be seriously reduced by water blockage [15–17]; the dispersed distribution of kerogen within matrices may affect the production evaluation [18]; sulfur precipitation and reservoir pressure-sensitive effects may affect the permeability, porosity and formation pressure [19]; the micro-scale flow mechanism, such as Knudsen diffusion, slippage effect, and adsorption, can be difficult to describe quantitatively [20,21] All these complexities associated with tight gas reservoirs make it difficult to build an accurate mathematical model to predict gas flow in tight gas reservoirs. The Duong model was built to estimate the EUR for unconventional reservoirs when the fracture flow is the most significant flow regime, such as in tight and shale gas wells [27]

Objectives

Methods

Results

Conclusion