scholarly journals Productivity Prediction of Fractured Horizontal Well in Shale Gas Reservoirs with Machine Learning Algorithms

2021 ◽  
Vol 11 (24) ◽  
pp. 12064
Author(s):  
Tianyu Wang ◽  
Qisheng Wang ◽  
Jing Shi ◽  
Wenhong Zhang ◽  
Wenxi Ren ◽  
...  
Keyword(s):  
Neural Network ◽  
Relative Error ◽  
Shale Gas ◽  
Gas Production ◽  
Average Relative Error ◽  
Maximum Relative Error ◽  
Gas Reservoirs ◽  
Lstm Network ◽  
Relative Errors ◽  
Productivity Prediction

Predicting shale gas production under different geological and fracturing conditions in the fractured shale gas reservoirs is the foundation of optimizing the fracturing parameters, which is crucial to effectively exploit shale gas. We present a multi-layer perceptron (MLP) network and a long short-term memory (LSTM) network to predict shale gas production, both of which can quickly and accurately forecast gas production. The prediction performances of the networks are comprehensively evaluated and compared. The results show that the MLP network can predict shale gas production by geological and fracturing reservoir parameters. The average relative error of the MLP neural network is 2.85%, and the maximum relative error is 12.9%, which can meet the demand of engineering shale gas productivity prediction. The LSTM network can predict shale gas production through historical production under the constraints of geological and fracturing reservoir parameters. The average relative error of the LSTM neural network is 0.68%, and the maximum relative error is 3.08%, which can reliably predict shale gas production. There is a slight deviation between the predicted results of the MLP model and the true values in the first 10 days. This is because the daily production decreases rapidly during the early production stage, and the production data change greatly. The largest relative errors of LSTM in this work on the 10th, 100th, and 1000th day are 0.95%, 0.73%, and 1.85%, respectively, which are far lower than the relative errors of the MLP predictions. The research results can provide a fast and effective mean for shale gas productivity prediction.

scholarly journals Investigation of the Main Factors During Shale-gas Production Using Grey Relational Analysis

2016 ◽  
Vol 9 (1) ◽  
pp. 207-215 ◽  
Author(s):  
Hongling Zhang ◽  
Jing Wang ◽  
Haiyong Zhang
Keyword(s):  
Shale Gas ◽  
Gas Production ◽  
Gas Reservoirs ◽  
Fracture Conductivity ◽  
Matrix Permeability ◽  
Shale Gas Reservoirs ◽  
Main Factors ◽  
Grey Relational ◽  
The Impact ◽  
Half Length

Shale gas is one of the primary types of unconventional reservoirs to be exploited in search for long-lasting resources. Production from shale gas reservoirs requires horizontal drilling with hydraulic fracturing to achieve the most economic production. However, plenty of parameters (e.g., fracture conductivity, fracture spacing, half-length, matrix permeability, and porosity,etc) have high uncertainty that may cause unexpected high cost. Therefore, to develop an efficient and practical method for quantifying uncertainty and optimizing shale-gas production is highly desirable. This paper focuses on analyzing the main factors during gas production, including petro-physical parameters, hydraulic fracture parameters, and work conditions on shale-gas production performances. Firstly, numerous key parameters of shale-gas production from the fourteen best-known shale gas reservoirs in the United States are selected through the correlation analysis. Secondly, a grey relational grade method is used to quantitatively estimate the potential of developing target shale gas reservoirs as well as the impact ranking of these factors. Analyses on production data of many shale-gas reservoirs indicate that the recovery efficiencies are highly correlated with the major parameters predicted by the new method. Among all main factors, the impact ranking of major factors, from more important to less important, is matrix permeability, fracture conductivity, fracture density of hydraulic fracturing, reservoir pressure, total organic content (TOC), fracture half-length, adsorbed gas, reservoir thickness, reservoir depth, and clay content. This work can provide significant insights into quantifying the evaluation of the development potential of shale gas reservoirs, the influence degree of main factors, and optimization of shale gas production.

scholarly journals An Analytical Method for Parameter Interpretation of Fracture Networks in Shale Gas Reservoirs considering Uneven Support of Fractures

Geofluids ◽  
2021 ◽  
Vol 2021 ◽  
pp. 1-15
Author(s):  
Shijun Huang ◽  
Jiaojiao Zhang ◽  
Sidong Fang ◽  
Xifeng Wang
Keyword(s):  
Data Analysis ◽  
Shale Gas ◽  
Analytical Method ◽  
Material Balance ◽  
Gas Production ◽  
Production Data ◽  
Analysis Method ◽  
Gas Reservoirs ◽  
Data Analysis Method ◽  
Parameter Interpretation

In shale gas reservoirs, the production data analysis method is widely used to invert reservoir and fracture parameter, and productivity prediction. Compared with numerical models and semianalytical models, which have high computational cost, the analytical model is mostly used in the production data analysis method to characterize the complex fracture network formed after fracturing. However, most of the current calculation models ignore the uneven support of fractures, and most of them use a single supported fracture model to describe the flow characteristics, which magnifies the role of supported fracture to a certain extent. Therefore, in this study, firstly, the fractures are divided into supported fractures and unsupported fractures. According to the near-well supported fractures and far-well unsupported fractures, the SRV zone is divided into outer SRV and inner SRV. The four areas are characterized by different seepage models, and the analytical solutions of the models are obtained by Laplace transform and inverse transform. Secondly, the material balance pseudotime is introduced to process the production data under the conditions of variable production and variable pressure. The double logarithmic curves of normalized production rate, rate integration, the derivative of the integration, and material balance pseudotime are established, and the parameters are interpreted by fitting the theoretical curve to the measured data. Then, the accuracy of the method is verified by comparison the parameter interpretation results with well test results, and the influence of parameters such as the half-length and permeability of supported and unsupported fractures on gas production is analyzed. Finally, the proposed method is applied to four field cases in southwest China. This paper mainly establishes an analytical method for parameter interpretation after hydraulic fracturing based on the production data analysis method considering the uneven support of fractures, which is of great significance for understanding the mechanism of fracturing stimulation, optimization of fracturing parameters, and gas production forecast.

Study Evaluates Fracturing Interference for Multiwell Pads in Shale Gas Reservoirs

2021 ◽  
Vol 73 (08) ◽  
pp. 67-68
Author(s):  
Chris Carpenter
Keyword(s):  
Hydraulic Fracturing ◽  
Shale Gas ◽  
Technical Conference ◽  
Gas Production ◽  
Gas Reservoirs ◽  
Well Drilling ◽  
Recovery Potential ◽  
Shale Gas Reservoirs ◽  
Well Spacing ◽  
Southwest Region

This article, written by JPT Technology Editor Chris Carpenter, contains highlights of paper SPE 201694, “Interwell Fracturing Interference Evaluation of Multiwell Pads in Shale Gas Reservoirs: A Case Study in WY Basin,” by Youwei He, SPE, Jianchun Guo, SPE, and Yong Tang, Southwest Petroleum University, et al., prepared for the 2020 SPE Annual Technical Conference and Exhibition, originally scheduled to be held in Denver, Colorado, 5–7 October. The paper has not been peer reviewed. The paper aims to determine the mechanisms of fracturing interference for multiwell pads in shale gas reservoirs and evaluate the effect of interwell fracturing interference on production. Field data of 56 shale gas wells in the WY Basin are applied to calculate the ratio of affected wells to newly fractured wells and understand its influence on gas production. The main controlling factors of fracturing interference are determined, and the interwell fracturing interacting types are presented. Production recovery potential for affected wells is analyzed, and suggestions for mitigating fracturing interference are proposed. Interwell Fracturing Interference Evaluation The WY shale play is in the southwest region of the Sichuan Basin, where shale gas reserves in the Wufeng-Longmaxi formation are estimated to be the highest in China. The reservoir has produced hydrocarbons since 2016. Infill well drilling and massive hydraulic fracturing operations have been applied in the basin. Each well pad usually is composed of six to eight multifractured horizontal wells (MFHWs). Well spacing within one pad, or the distance between adjacent well pads, is so small that fracture interference can occur easily between infill wells and parent wells. Fig. 1 shows the number of wells affected by in-fill well fracturing from 2016 to 2019 in the basin. As the number of newly drilled wells increased between 2017 and 2019, the number of wells affected by hydraulic fracturing has greatly increased. The number of wells experiencing fracturing interaction has reached 65 in the last 4 years at the time of writing.

scholarly journals Analysis of the Influence of Different Fracture Network Structures on the Production of Shale Gas Reservoirs

Geofluids ◽  
2020 ◽  
Vol 2020 ◽  
pp. 1-11
Author(s):  
Ming Yue ◽  
Xiaohe Huang ◽  
Fanmin He ◽  
Lianzhi Yang ◽  
Weiyao Zhu ◽  
...  
Keyword(s):  
Finite Element ◽  
Shale Gas ◽  
Flow Direction ◽  
Fracture Network ◽  
Gas Production ◽  
Low Flow ◽  
Gas Reservoirs ◽  
Gas Reservoir ◽  
Shale Gas Reservoir ◽  
Gas Seepage

Volume fracturing is a key technology in developing unconventional gas reservoirs that contain nano/micron pores. Different fracture structures exert significantly different effects on shale gas production, and a fracture structure can be learned only in a later part of detection. On the basis of a multiscale gas seepage model considering diffusion, slippage, and desorption effects, a three-dimensional finite element algorithm is developed. Two finite element models for different fracture structures for a shale gas reservoir in the Sichuan Basin are established and studied under the condition of equal fracture volumes. One is a tree-like fracture, and the other is a lattice-like fracture. Their effects on the production of a fracture network structure are studied. Numerical results show that under the same condition of equal volumes, the production of the tree-like fracture is higher than that of the lattice-like fracture in the early development period because the angle between fracture branches and the flow direction plays an important role in the seepage of shale gas. In the middle and later periods, owing to a low flow rate, the production of the two structures is nearly similar. Finally, the lattice-like fracture model is regarded as an example to analyze the factors of shale properties that influence shale gas production. The analysis shows that gas production increases along with the diffusion coefficient and matrix permeability. The increase in permeability leads to a larger increase in production, but the decrease in permeability leads to a smaller decrease in production, indicating that the contribution of shale gas production is mainly fracture. The findings of this study can help better understand the influence of different shapes of fractures on the production in a shale gas reservoir.

scholarly journals The Evaluation and Sensitivity of Decline Curve Modelling

Energies ◽  
2020 ◽  
Vol 13 (11) ◽  
pp. 2765
Author(s):  
Prinisha Manda ◽  
Diakanua Nkazi
Keyword(s):  
Shale Gas ◽  
Goodness Of Fit ◽  
Accuracy Assessment ◽  
Gas Production ◽  
Production Performance ◽  
Logistic Growth ◽  
Gas Reservoirs ◽  
Shale Gas Reservoirs ◽  
Decline Curve ◽  
Exponential Decline

The development of prediction tools for production performance and the lifespan of shale gas reservoirs has been a focus for petroleum engineers. Several decline curve models have been developed and compared with data from shale gas production. To accurately forecast the estimated ultimate recovery for shale gas reservoirs, consistent and accurate decline curve modelling is required. In this paper, the current decline curve models are evaluated using the goodness of fit as a measure of accuracy with field data. The evaluation found that there are advantages in using the current DCA models; however, they also have limitations associated with them that have to be addressed. Based on the accuracy assessment conducted on the different models, it appears that the Stretched Exponential Decline Model (SEDM) and Logistic Growth Model (LGM), followed by the Extended Exponential Decline Model (EEDM), the Power Law Exponential Model (PLE), the Doung’s Model, and lastly, the Arps Hyperbolic Decline Model, provide the best fit with production data.

scholarly journals Water-Gas Two-Phase Flow Behavior of Multi-Fractured Horizontal Wells in Shale Gas Reservoirs

Processes ◽  
2019 ◽  
Vol 7 (10) ◽  
pp. 664 ◽  
Author(s):  
Lei Li ◽  
Guanglong Sheng ◽  
Yuliang Su
Keyword(s):  
Organic Matter ◽  
Shale Gas ◽  
Flow Behavior ◽  
Horizontal Wells ◽  
Gas Production ◽  
Phase Flow ◽  
Gas Reservoirs ◽  
Two Phase ◽  
Fractured Horizontal Wells ◽  
Water Gas

Hydraulic fracturing is a necessary method to develop shale gas reservoirs effectively and economically. However, the flow behavior in multi-porosity fractured reservoirs is difficult to characterize by conventional methods. In this paper, combined with apparent porosity/permeability model of organic matter, inorganic matter and induced fractures, considering the water film in unstimulated reservoir volume (USRV) region water and bulk water in effectively stimulated reservoir volume (ESRV) region, a multi-media water-gas two-phase flow model was established. The finite difference is used to solve the model and the water-gas two-phase flow behavior of multi-fractured horizontal wells is obtained. Mass transfer between different-scale media, the effects of pore pressure on reservoirs and fluid properties at different production stages were considered in this model. The influence of the dynamic reservoir physical parameters on flow behavior and gas production in multi-fractured horizontal wells is studied. The results show that the properties of the total organic content (TOC) and the inherent porosity of the organic matter affect gas production after 40 days. With the gradual increase of production time, the gas production rate decreases rapidly compared with the water production rate, and the gas saturation in the inorganic matter of the ESRV region gradually decreases. The ignorance of stress sensitivity would cause the gas production increase, and the ignorance of organic matter shrinkage decrease the gas production gradually. The water film mainly affects gas production after 100 days, while the bulk water has a greater impact on gas production throughout the whole period. The research provides a new method to accurately describe the two-phase fluid flow behavior in different scale media of fractured shale gas reservoirs.

A New Method to Predict Gas Production Based on Fuzzy BP Artificial Neural Network

2014 ◽  
Vol 1044-1045 ◽  
pp. 688-691
Author(s):  
Ran Zhang ◽  
Jun Zhou ◽  
Cheng Yong Li
Keyword(s):  
Neural Network ◽  
Bp Neural Network ◽  
Evaluation Method ◽  
Comprehensive Evaluation ◽  
Fuzzy Comprehensive Evaluation ◽  
Gas Production ◽  
Gas Well ◽  
Well Productivity ◽  
Qualitative Factors ◽  
Productivity Prediction

BP neural network has been successfully used in the gas well productivity prediction, but as a result of neural network is sensitive to the number of input parameters, we had to ignore some factors that is less important to the gas well productivity. In addition, the existing various productivity prediction method cannot consider the influence of some important qualitative factors. This article integrated the advantages of fuzzy comprehensive evaluation and BP neural network, fuzzy comprehensive evaluation method is used to construct the BP neural network's input matrix, and BP neural network learning function is used to solve the connection weights, so as to achieve the aim of predicting gas production. This method not only can consider as many factors influence on gas well production, ut also can consider qualitative factors, so the forecast results of the new model are more realistically close to the actual production situation of reservoirs.

Sign in / Sign up

Export Citation Format

Share Document