Combining classification and regression for improving permeability estimations from 1 H NMR relaxation data

Plastino, Alexandre; Gonçalves, Eduardo Corrêa; Silva, Pablo Nascimento da; Carneiro, Giovanna; Azeredo, Rodrigo Bagueira de Vasconcellos

doi:10.1016/j.jappgeo.2017.09.003

Cited by 14 publications

(4 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This is the major reason why the petroleum industry has embraced the application of computational intelligence techniques, particularly in reservoir characterization, as a result of its inexpensive approach in achieving reliable values [4]. Most studies [5][6][7][8][9][10][11][12][13][14] have reported the capability of various forms and combination of computational intelligence methods, including artificial neural network (ANN), in generating porosity and permeability values. However, most of these computational intelligence methods (including ANN) are slow in generating results mainly due to the iterative tuning of the models' user-defined parameters and the steepest-gradient training algorithms utilized.…”

Section: Introductionmentioning

confidence: 99%

Investigating the Predictive Performance of Gaussian Process Regression in Evaluating Reservoir Porosity and Permeability

et al. 2018

View full text Add to dashboard Cite

In this paper, a new predictive model based on Gaussian process regression (GPR) that does not require iterative tuning of user-defined model parameters has been proposed to determine reservoir porosity and permeability. For this purpose, the capability of GPR was appraised statistically for predicting porosity and permeability of the southern basin of the South Yellow Sea using petrophysical well log data. Generally, the performance of GPR is deeply reliant on the type covariance function utilized. Therefore, to obtain the optimal GPR model, five different kernel functions were tested. The resulting optimal GPR model consisted of the exponential covariance function, which produced the highest correlation coefficient (R) of 0.85 and the least root mean square error (RMSE) of 0.037 and 6.47 for porosity and permeability, respectively. Comparison was further made with benchmark methods involving a back propagation neural network (BPNN), generalized regression neural network (GRNN), and radial basis function neural network (RBFNN). The statistical findings revealed that the proposed GPR is a powerful technique and can be used as a supplement to the widely used artificial neural network methods. In terms of computational speed, the GPR technique was computationally faster than the BPNN, GRNN, and RBFNN methods in estimating reservoir porosity and permeability.

show abstract

Section: Introductionmentioning

confidence: 99%

Investigating the Predictive Performance of Gaussian Process Regression in Evaluating Reservoir Porosity and Permeability

et al. 2018

View full text Add to dashboard Cite

show abstract

“…These tools involved a series of records which illustrate the fluid behavior and nature of the reservoir during the production and injection procedures of an oil formation. Figure 2 shows the production logging tools components schematically (Frooqnia et al 2011;Grayson et al 2002;Hoffman and Narr 2012;Li and Zhao 2014;Plastino et al 2017;Sullivan 2007). One of the chief aims of production logging tools is to analyze and investigate the borehole performances like dynamic or static situation of a production well, measure the amount of productivity and injectivity index of zones or layers of a field, monitor the borehole inefficiencies by interpreting obtained logs, diagnose the effectiveness of stimulation or completion processes, and measure the physical condition of a well (Aghli et al 2016;Al-Mulhim et al 2015).…”

Section: Production Logging Toolsmentioning

confidence: 99%

Integrated production logging tools approach for convenient experimental individual layer permeability measurements in a multi-layered fractured reservoir

Davarpanah

Mirshekari

Behbahani³

et al. 2018

J Petrol Explor Prod Technol

104

View full text Add to dashboard Cite

Appropriate estimation of permeability is considered as one of the significant concerns of petroleum industries. Due to the growing demand for hydrocarbon fossil fuels in numerous industries, Petroleum Engineers always try to provide holistic and sustainable solutions to measure this parameter more accurately and to calculate the proper original oil in place (OOIP) and initial reserve. Hence, this accuracy estimation helps engineers whether the production and exploration operations are profitable or not and it might virtually eliminate the unnecessary expenditures. The term production logging tools (henceforth; PLT) involve a wide variety of measurement tools and many sensors. It, too, carries interpretation tools which evaluate the formation properties, in respect of the way PLTs would analyze the formation fluid movements inside and outside of the wellbore and subsequently estimate the production flow rate for each layer. On the other hand, it gives production and completion engineers the chance to investigate the appropriate efficiency of production and perforation processes to organize the remediation methodologies or preplan proper designing for modifying completion procedures which have based on the production logging tools interpretation. The purpose of this comprehensive research is to compare two different techniques (PLT and core analysis) of permeability measurement in a six-layered fractured reservoir and subsequently normalize each parameter to obtain the proper estimation. As a result, according to the evaluation of each technique, the amount of permeability in the layers 1, 2, 3, and 5 is relatively close to each other. Furthermore, regarding higher expenses of core analysis tests and the reliability of PLTs according to the results of this paper in the four out of six individual layers, Emeraude software by utilizing PLT interpretation could be a substitution and preferable methodology instead of core analysis measurements.

show abstract

“…Regardless of the models used for the estimation of permeability from NMR measurements, current approaches summarize all information contained in the relaxation distribution into one variable, without considering the local pore structure variations of adjoined regions in rock samples (Rios et al, ). While the mentioned equations work well for rock cores with a simple interior structure, permeabilities may be estimated less accurately for rock cores with a heterogeneous structure, especially if regions of high porosity are poorly connected to each other (Plastino et al, ). In this situation, a single‐step modeling to obtain a global permeability value using equations and may not be unsatisfactory to reflect the local permeability variation in samples, which has been found in carbonate rock, mudstone, and gas shale (Ge et al, ; Josh et al, ; Lala & El‐Sayed, ; Tan et al, ).…”

Section: Introductionmentioning

confidence: 99%

Permeability Profiling of Rock Cores Using a Novel Spatially Resolved NMR Relaxometry Method: Preliminary Results From Sandstone and Limestone

et al. 2019

View full text Add to dashboard Cite

Permeability characterizes how saturating fluids penetrate and flow through porous rocks. It is an important quantity that can be used to model the behavior of reservoir rocks and to optimize underground resource exploration. Permeability can be characterized via nuclear magnetic resonance (NMR) relaxometry techniques by utilizing the yielded pore volume and pore length scales. However, for heterogeneous rocks a simple estimation of a permeability value based on the existing models may be inadequate. In this work, a new relaxation technique is applied to rock cores in conjunction with magnetic resonance imaging schemes. This allows the interior structures of studied rock cores to be recorded along a chosen sample axis. Based on the spatially resolved relaxation maps obtained, a factor of local connectivity was introduced for the first time and calculated by using the correlation degree between adjacent relaxation time distributions originating from neighboring image slices. Consequently, a permeability profile was estimated by consecutively considering the local porosity, the spatially resolved relaxation time distribution, and the connectivity factor. Experimental results prove that permeability profiles trace the heterogeneity of rock samples. Furthermore, averaged permeabilities of rock cores were calculated while taking into account local connectivity. The values obtained approach better to brine-permeability measurements as compared to data processing disregarding connectivity.

show abstract

Combining classification and regression for improving permeability estimations from 1 H NMR relaxation data

Cited by 14 publications

References 22 publications

Investigating the Predictive Performance of Gaussian Process Regression in Evaluating Reservoir Porosity and Permeability

Investigating the Predictive Performance of Gaussian Process Regression in Evaluating Reservoir Porosity and Permeability

Integrated production logging tools approach for convenient experimental individual layer permeability measurements in a multi-layered fractured reservoir

Permeability Profiling of Rock Cores Using a Novel Spatially Resolved NMR Relaxometry Method: Preliminary Results From Sandstone and Limestone

Contact Info

Product

Resources

About