A Comprehensive Machine Learning Approach for Quantitatively Analyzing Development Performance and Optimization for a Heterogeneous Carbonate Reservoir in Middle East

2021 ◽  
Author(s):  
Ruijie Huang ◽  
Chenji Wei ◽  
Baohua Wang ◽  
Baozhu Li ◽  
Jian Yang ◽  
...  
Keyword(s):  
Machine Learning ◽  
Principal Component ◽  
Optimal Number ◽  
Gas Production ◽  
Carbonate Reservoir ◽  
Multidimensional Data ◽  
Well Performance ◽  
Reference Case ◽  
Evaluation Approach ◽  
Water Cut

Abstract Compared with conventional reservoir, the development efficiency of the carbonate reservoir is lower, because of the strong heterogeneity and complicated reservoir structure. How to accurately and quantitatively analyze development performance is critical to understand challenges faced, and to propose optimization plans to improve recovery. In the study, we develop a workflow to evaluate similarities and difference of well performance based on Machine Learning methods. A comprehensive Machine Learning evaluation approach for well performance is established by utilizing Principal Component Analysis (PCA) in combination with K-Means clustering. The multidimensional dataset used for analysis consists of over 15 years dynamic surveillance data of producers and static geology parameters of formation, such as oil/water/gas production, GOR, water cut (WC), porosity, permeability, thickness, and depth. This approach divides multidimensional data into several clusters by PCA and K-Means, and quantitatively evaluate the well performance based on clustering results. The approach is successfully developed to visualize (dis)similarities among dynamic and static data of heterogeneous carbonate reservoir, the optimal number of clusters of 27-dimension data is 4. This method provides a systematic framework for visually and quantitatively analyzing and evaluating the development performance of production wells. Reservoir engineers can efficiently propose targeted optimization measures based on the analysis results. This paper offers a reference case for well performance clustering and quantitative analysis and proposing optimization plans that will help engineers make better decision in similar situation.

scholarly journals Prediction of Hydraulic Fractured Well Performance Using Empirical Correlation and Machine Learning

2021 ◽  
Vol 44 (2) ◽  
pp. 131-145
Author(s):  
Kamal Hamzah ◽  
Amega Yasutra ◽  
Dedy Irawan
Keyword(s):  
Machine Learning ◽  
Hydraulic Fracturing ◽  
Petroleum Industry ◽  
Prediction Method ◽  
Empirical Correlation ◽  
Well Performance ◽  
Data Limitations ◽  
Medium Permeability ◽  
Water Cut ◽  
Fractured Well

Hydraulic fracturing has been established as one of production enhancement methods in the petroleum industry. This method is proven to increase productivity and reserves in low permeability reservoirs, while in medium permeability, it accelerates production without affecting well reserves. However, production result looks scattered and appears to have no direct correlation to individual parameters. It also tend to have a decreasing trend, hence the success ratio needs to be increased. Hydraulic fracturing in the South Sumatra area has been implemented since 2002 and there is plenty of data that can be analyzed to resolve the relationship between actual production with reservoir parameters and fracturing treatment. Empirical correlation approach and machine learning (ML) methods are both used to evaluate this relationship. Concept of Darcy's equation is utilized as basis for the empirical correlation on the actual data. The ML method is then applied to provide better predictions both for production rate and water cut. This method has also been developed to solve data limitations so that the prediction method can be used for all wells. Empirical correlation can gives an R2 of 0.67, while ML can gives a better R2 that is close to 0.80. Furthermore, this prediction method can be used for well candidate selection means.

MACHINE LEARNING FOR PRODUCTIVITY PREDICTION IN HETEROGENEOUS CARBONATE GAS RESERVOIRS, CENTRAL SICHUAN BASIN, CHINA

2021 ◽  
Author(s):  
Zuoan Zhao ◽  
◽  
Dali Wang ◽  
Keyword(s):  
Machine Learning ◽  
Gamma Ray ◽  
Pore Space ◽  
Gas Production ◽  
Carbonate Reservoir ◽  
Production Rates ◽  
Gas Reservoirs ◽  
Data Set ◽  
Productivity Prediction ◽  
Gamma Ray Log

An approach of machine learning was used to evaluate and predict the production of the heterogeneous carbonate gas reservoirs in the horizontal development wells of the late Precambrian Dengying Formation. The present data set of machine learning consists of gamma ray log, laterolog, high-resolution electrical image logs, and production rate data. The previous data set acquired the conventional openhole logs, including gamma ray log, neutron-density log, sonic log, laterolog, and dipole acoustic log. The challenge in the previous data set was that the training process for machine learning was not convergent. It was most likely that the conventional log responses did not fully correspond to the productivity of the heterogenous carbonate gas reservoirs. Forty-one wells associated with the present data set were used to set up the training sample data set for the machine learning to the productivity prediction of the carbonate gas reservoirs. The data set construction includes log depth shift, calibrated image log creation, classification of reservoir types from core and carbonate reservoir heterogeneity variables extraction from image logs. Core observation and core laboratory analysis indicate that the pore space of the carbonate gas reservoirs mainly consists of vugs, caves, and fractures. However, the vugs and caves are selectively developed and randomly distributed both laterally and vertically. This represents a complex heterogeneous carbonate reservoir in which the vugs and caves are key contributor to the total pore space of the carbonate gas reservoir. The attributes of the vugs and caves can be extracted from the electrical image logs, including connectedness, surface proportion, size, and thickness of vug, and cave zones. Six horizontal development wells were used to validate the machine learning approach. The predicted gas production rates in the four wells separately were 700,000 m3/d, 2,000,000 m3/d, 800,000 m3/d, 300,000 m3/d, 1,100,000 m3/d and 1,180,000 m3/d, and the respective actual gas production rates are 1,019,790 m3/d, 1,820,000 m3/d, 800,000 m3/d, 396,000 m3/d , 1,700,000 m3/d, and 1,411,900 m3/d. The machine learning workflow and approach provided satisfactory results in the six horizontal wells. Subsequently, the electrical image logs have run in the standard logging program in the more than 50 horizontal development wells.

Prediction of Hydraulic Fractured Well Performance Using Empirical Correlation and Machine Learning

2021 ◽  
Vol 44 (2) ◽  
pp. 141-152
Author(s):  
Kamal Hamzah ◽  
Amega Yasutra ◽  
Dedy Irawan
Keyword(s):  
Machine Learning ◽  
Hydraulic Fracturing ◽  
Petroleum Industry ◽  
Prediction Method ◽  
Empirical Correlation ◽  
Well Performance ◽  
Data Limitations ◽  
Medium Permeability ◽  
Water Cut ◽  
Fractured Well

Hydraulic fracturing has been established as one of production enhancement methods in the petroleum industry. This method is proven to increase productivity and reserves in low permeability reservoirs, while in medium permeability, it accelerates production without affecting well reserves. However, production result looks scattered and appears to have no direct correlation to individual parameters. It also tend to have a decreasing trend, hence the success ratio needs to be increased. Hydraulic fracturing in the South Sumatra area has been implemented since 2002 and there is plenty of data that can be analyzed to resolve the relationship between actual production with reservoir parameters and fracturing treatment. Empirical correlation approach and machine learning (ML) methods are both used to evaluate this relationship. Concept of Darcy's equation is utilized as basis for the empirical correlation on the actual data. The ML method is then applied to provide better predictions both for production rate and water cut. This method has also been developed to solve data limitations so that the prediction method can be used for all wells. Empirical correlation can gives an R2 of 0.67, while ML can gives a better R2 that is close to 0.80. Furthermore, this prediction method can be used for well candidate selection means.

Artificial Intelligence for Infill Well Placement and Design Optimization in Multi-layered/stacked Reservoirs Under Subsurface Uncertainty

2021 ◽  
Author(s):  
Shi Su ◽  
Ralf Schulze-Riegert ◽  
Hussein Mustapha ◽  
Philipp Lang ◽  
Chakib Kada Kloucha
Keyword(s):  
Artificial Intelligence ◽  
Machine Learning ◽  
Supervised Machine Learning ◽  
Machine Learning Techniques ◽  
Multiple Model ◽  
Well Performance ◽  
Reference Case ◽  
Well Placement ◽  
Learning Techniques ◽  
Economic Constraints

Abstract Effective well placement and design planning accounts for subsurface uncertainties to estimate production and economic outcomes. Reservoir modelling and simulation workflows build on ensemble approaches to manage uncertainties for production forecasting. Ensemble generation and interpretation requires a higher degree of automation analytics and artificial intelligence for fast value extraction and decision support. This work develops practical intelligent workflow steps for a robust infill well placement and design scenario in multi-layered/stacked reservoirs under uncertainty. Potential well targets are classified by an opportunity index defined by a combination of rock and hydrocarbon flow properties as well as connected volumes above a minimum economic volume. Unsupervised learning techniques are applied to automate the search for alternative target areas, so-called hotspot regions. Supervised machine/learning models are used to predict infill well performance based on simulated and/or past production experience. A stochastic evaluation including all ensemble cases is used to capture uncertainty. Vertical, deviated, horizontal and multilateral wells are proposed to optimally target single or connect to multiple hotspot regions under technical and economic constraints. A structured workflow design is applied to a multi-layered/stacked reservoir model. Subsurface uncertainties are described and captured by multiple model realizations, which are constrained in areas of historical wells. An infill well program for a multi-layered/stacked reservoir is defined for incremental production increase under economic constraints. This work shows how robust well location and design builds on the full ensemble of cases with a high degree of automation using analytics and machine-learning techniques. Both production and economic targets are calculated and compared to a reference case for robust solution verification and probability of success. In conclusion, an overall reservoir-driven field development strategy is required for efficient execution. However, automation is well applicable to repetitive workflow steps which includes hotspot search in an ensemble of validated reservoir models. This work presents an integrated, intelligent solution for informed decision making on infill drilling locations and refined well design. Higher degree of automation with embedded intelligence are discussed from case generation to hotspot identification. Aspects of model calibration in a producing field environment are addressed.

Classification of Observations through Combination of the Dimension Reduction and the Cluster Analysis

2017 ◽  
Vol 7 (8) ◽  
pp. 30
Author(s):  
Hyeuk Kim
Keyword(s):  
Machine Learning ◽  
Principal Component Analysis ◽  
Cluster Analysis ◽  
Unsupervised Learning ◽  
Principal Component ◽  
Component Analysis ◽  
Baseball Players ◽  
Partitioning Around Medoids ◽  
Different Characteristics

Unsupervised learning in machine learning divides data into several groups. The observations in the same group have similar characteristics and the observations in the different groups have the different characteristics. In the paper, we classify data by partitioning around medoids which have some advantages over the k-means clustering. We apply it to baseball players in Korea Baseball League. We also apply the principal component analysis to data and draw the graph using two components for axis. We interpret the meaning of the clustering graphically through the procedure. The combination of the partitioning around medoids and the principal component analysis can be used to any other data and the approach makes us to figure out the characteristics easily.

Analysis of the Bath Motion in the MM-SQC Dynamics Using Unsupervised Machine Learning Dimensionality Reduction Approaches: Principal Component Analysis

2020 ◽  
Author(s):  
Jiawei Peng ◽  
Yu Xie ◽  
Deping Hu ◽  
Zhenggang Lan
Keyword(s):  
Machine Learning ◽  
Principal Component Analysis ◽  
Collective Motion ◽  
Principal Component ◽  
Component Analysis ◽  
Nonadiabatic Dynamics ◽  
Trajectory Data ◽  
Unsupervised Machine Learning ◽  
Physical Knowledge ◽  
Vibronic Couplings

The system-plus-bath model is an important tool to understand nonadiabatic dynamics for large molecular systems. The understanding of the collective motion of a huge number of bath modes is essential to reveal their key roles in the overall dynamics. We apply the principal component analysis (PCA) to investigate the bath motion based on the massive data generated from the MM-SQC (symmetrical quasi-classical dynamics method based on the Meyer-Miller mapping Hamiltonian) nonadiabatic dynamics of the excited-state energy transfer dynamics of Frenkel-exciton model. The PCA method clearly clarifies that two types of bath modes, which either display the strong vibronic couplings or have the frequencies close to electronic transition, are very important to the nonadiabatic dynamics. These observations are fully consistent with the physical insights. This conclusion is obtained purely based on the PCA understanding of the trajectory data, without the large involvement of pre-defined physical knowledge. The results show that the PCA approach, one of the simplest unsupervised machine learning methods, is very powerful to analyze the complicated nonadiabatic dynamics in condensed phase involving many degrees of freedom.

Integrated Subsurface to Surface Modeling to Assess Reservoir Uncertainty Quantification to A Carbonate Reservoir

2020 ◽  
Author(s):  
A. Chaterine
Keyword(s):  
Uncertainty Quantification ◽  
Gas Production ◽  
Carbonate Reservoir ◽  
Production Volume ◽  
Production Forecast ◽  
Uncertainty Range ◽  
Field Development ◽  
Production Forecasting ◽  
Production Scheme ◽  
Rock Compressibility

This study accommodates subsurface uncertainties analysis and quantifies the effects on surface production volume to propose the optimal future field development. The problem of well productivity is sometimes only viewed from the surface components themselves, where in fact the subsurface component often has a significant effect on these production figures. In order to track the relationship between surface and subsurface, a model that integrates both must be created. The methods covered integrated asset modeling, probability forecasting, uncertainty quantification, sensitivity analysis, and optimization forecast. Subsurface uncertainties examined were : reservoir closure, regional segmentation, fluid contact, and SCAL properties. As the Integrated Asset Modeling is successfully conducted and a matched model is obtained for the gas-producing carbonate reservoir, highlights of the method are the following: 1) Up to ± 75% uncertainty range of reservoir parameters yields various production forecasting scenario using BHP control with the best case obtained is 335 BSCF of gas production and 254.4 MSTB of oil production, 2) SCAL properties and pseudo-faults are the most sensitive subsurface uncertainty that gives major impact to the production scheme, 3) EOS modeling and rock compressibility modeling must be evaluated seriously as those contribute significantly to condensate production and the field’s revenue, and 4) a proposed optimum production scenario for future development of the field with 151.6 BSCF gas and 414.4 MSTB oil that yields a total NPV of 218.7 MMUSD. The approach and methods implemented has been proven to result in more accurate production forecast and reduce the project cost as the effect of uncertainty reduction.

Application of Machine Learning in Animal Disease Analysis and Prediction

2020 ◽  
Vol 15 ◽  
Author(s):  
Shuwen Zhang ◽  
Qiang Su ◽  
Qin Chen
Keyword(s):  
Machine Learning ◽  
Unsupervised Learning ◽  
Supervised Learning ◽  
Clustering Algorithm ◽  
Principal Component ◽  
Support Vector ◽  
Animal Disease ◽  
Human Beings ◽  
Animal Diseases ◽  
Disease Analysis

Abstract: Major animal diseases pose a great threat to animal husbandry and human beings. With the deepening of globalization and the abundance of data resources, the prediction and analysis of animal diseases by using big data are becoming more and more important. The focus of machine learning is to make computers learn how to learn from data and use the learned experience to analyze and predict. Firstly, this paper introduces the animal epidemic situation and machine learning. Then it briefly introduces the application of machine learning in animal disease analysis and prediction. Machine learning is mainly divided into supervised learning and unsupervised learning. Supervised learning includes support vector machines, naive bayes, decision trees, random forests, logistic regression, artificial neural networks, deep learning, and AdaBoost. Unsupervised learning has maximum expectation algorithm, principal component analysis hierarchical clustering algorithm and maxent. Through the discussion of this paper, people have a clearer concept of machine learning and understand its application prospect in animal diseases.

scholarly journals Transformer Oil Quality Assessment Using Random Forest with Feature Engineering

Energies ◽  
2021 ◽  
Vol 14 (7) ◽  
pp. 1809
Author(s):  
Mohammed El Amine Senoussaoui ◽  
Mostefa Brahami ◽  
Issouf Fofana
Keyword(s):  
Machine Learning ◽  
Random Forest ◽  
Oil Quality ◽  
Principal Component ◽  
Condition Assessment ◽  
Classification Performance ◽  
Transformer Oil ◽  
Classification Model ◽  
Insulation Degradation ◽  
Transformer Oils

Machine learning is widely used as a panacea in many engineering applications including the condition assessment of power transformers. Most statistics attribute the main cause of transformer failure to insulation degradation. Thus, a new, simple, and effective machine-learning approach was proposed to monitor the condition of transformer oils based on some aging indicators. The proposed approach was used to compare the performance of two machine-learning classifiers: J48 decision tree and random forest. The service-aged transformer oils were classified into four groups: the oils that can be maintained in service, the oils that should be reconditioned or filtered, the oils that should be reclaimed, and the oils that must be discarded. From the two algorithms, random forest exhibited a better performance and high accuracy with only a small amount of data. Good performance was achieved through not only the application of the proposed algorithm but also the approach of data preprocessing. Before feeding the classification model, the available data were transformed using the simple k-means method. Subsequently, the obtained data were filtered through correlation-based feature selection (CFsSubset). The resulting features were again retransformed by conducting the principal component analysis and were passed through the CFsSubset filter. The transformation and filtration of the data improved the classification performance of the adopted algorithms, especially random forest. Another advantage of the proposed method is the decrease in the number of the datasets required for the condition assessment of transformer oils, which is valuable for transformer condition monitoring.

scholarly journals Comparison of Capability of SAR and Optical Data in Mapping Forest above Ground Biomass Based on Machine Learning

2020 ◽  
Vol 5 (1) ◽  
pp. 13
Author(s):  
Negar Tavasoli ◽  
Hossein Arefi
Keyword(s):  
Machine Learning ◽  
Carbon Stock ◽  
Vegetation Indices ◽  
Forest Biomass ◽  
Principal Component ◽  
Optical Data ◽  
Above Ground Biomass ◽  
Ground Biomass ◽  
Texture Characteristics ◽  
Sentinel 2

Assessment of forest above ground biomass (AGB) is critical for managing forest and understanding the role of forest as source of carbon fluxes. Recently, satellite remote sensing products offer the chance to map forest biomass and carbon stock. The present study focuses on comparing the potential use of combination of ALOSPALSAR and Sentinel-1 SAR data, with Sentinel-2 optical data to estimate above ground biomass and carbon stock using Genetic-Random forest machine learning (GA-RF) algorithm. Polarimetric decompositions, texture characteristics and backscatter coefficients of ALOSPALSAR and Sentinel-1, and vegetation indices, tasseled cap, texture parameters and principal component analysis (PCA) of Sentinel-2 based on measured AGB samples were used to estimate biomass. The overall coefficient (R2) of AGB modelling using combination of ALOSPALSAR and Sentinel-1 data, and Sentinel-2 data were respectively 0.70 and 0.62. The result showed that Combining ALOSPALSAR and Sentinel-1 data to predict AGB by using GA-RF model performed better than Sentinel-2 data.

Sign in / Sign up

Export Citation Format

Share Document