Shaly Sand Analysis using Density-Neutron Porosities from a Cased-hole Pulsed Neutron System

Odom, Richard B.; Paul, Pavlakos; Diocee, Satwant; Bailey, S. Lawrence; Zander, Dennis; Gillespie, John W.

doi:10.2523/55641-ms

Cited by 4 publications

(3 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…(3) For each habitat, calculate its mean square error by relative parameters based on formulas (9). And the basic rule of fitness function is the better performance maintains the smaller value of MSE.…”

Section: Scientific Programmingmentioning

confidence: 99%

“…While the two-layered FNN is the most popular neural network used in practical applications, it is not suitable for solving nonlinear problems [7,8]. The Multilayer Perceptron (MLP) [9,10], a feedforward neural network with one or more hidden layers between the input and the output layers, is more successful in dealing with nonlinear problems such as pattern classification, big data prediction, and function approximation. Previous research [11] shows that MLPs with one hidden layer are able to approximate any continuous or discontinuous function.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Developing a Novel Hybrid Biogeography-Based Optimization Algorithm for Multilayer Perceptron Training under Big Data Challenge

Chen

et al. 2018

Scientific Programming

View full text Add to dashboard Cite

A Multilayer Perceptron (MLP) is a feedforward neural network model consisting of one or more hidden layers between the input and output layers. MLPs have been successfully applied to solve a wide range of problems in the fields of neuroscience, computational linguistics, and parallel distributed processing. While MLPs are highly successful in solving problems which are not linearly separable, two of the biggest challenges in their development and application are the local-minima problem and the problem of slow convergence under big data challenge. In order to tackle these problems, this study proposes a Hybrid Chaotic Biogeography-Based Optimization (HCBBO) algorithm for training MLPs for big data analysis and processing. Four benchmark datasets are employed to investigate the effectiveness of HCBBO in training MLPs. The accuracy of the results and the convergence of HCBBO are compared to three well-known heuristic algorithms: (a) Biogeography-Based Optimization (BBO), (b) Particle Swarm Optimization (PSO), and (c) Genetic Algorithms (GA). The experimental results show that training MLPs by using HCBBO is better than the other three heuristic learning approaches for big data processing.

show abstract

Section: Scientific Programmingmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Developing a Novel Hybrid Biogeography-Based Optimization Algorithm for Multilayer Perceptron Training under Big Data Challenge

Chen

et al. 2018

Scientific Programming

View full text Add to dashboard Cite

show abstract

“…For a MLP, the expression for actual output found from input is provided in (1). The weights of the MLP are changed by BP so that the error E of equation (2), which computes the sum of the squared errors between the actual and desired outputs, decreases.…”

Section: Discussion Of Back-propagationmentioning

confidence: 99%

Analysis and improvement of multiple optimal learning factors for feed-forward networks

Jesudhas

Manry

Rawat

et al. 2011

The 2011 International Joint Conference on Neural Networks

View full text Add to dashboard Cite

are analyzed. The use of optimal diagonal transformation matrices on the net function vector is proved to be equivalent to training the MLP using multiple optimal learning factors (MOLF). A method for linearly compressing large ill-conditioned MOLF Hessian matrices into smaller wellconditioned ones is developed. This compression approach is shown to be equivalent to using several hidden units per learning factor. The technique is extended to large networks. In simulations, the proposed algorithm performs almost as well as the Levenberg Marquardt (LM) algorithm with the computational complexity of a first order training algorithm.

show abstract

Examples of Cased Reservoir Analysis in the Ventura Basin, California

Odom¹,

March²

2000

SPE/AAPG Western Regional Meeting

View full text Add to dashboard Cite

The oil fields of the Ventura Basin are very mature with many wells drilled before the advent of modern logging tools. Despite the years of development, there remain opportunities for cost-effective production, given the great thickness of the productive intervals and the dynamic-nature of the depositional and tectonic environment that created them. Cased-reservoir analysis in existing wells is an efficient tool to target productive intervals. The well-logs and analysis examples utilize pulsed-neutron technologies. The cased-hole neutron and density porosities from the pulsed-neutron system provide the porosity information and describe gas-bearing intervals. Water flooding and gas injection has changed many of the original fluid properties. In the log analysis examples, Carbon/Oxygen logging describes the water saturations, and Sigma logging helps in the discrimination of the sands from the mudstones. Introduction Pulsed-neutron technologies offer engineers and geoscientists analysis tools for describing reservoirs in existing cased wells. Two log analysis examples show the applications of these technologies to the fields of the Ventura Basin. The low salinity of the reservoir water and secondary water flooding require Carbon/Oxygen logging to measure the water saturation. In the log examples the water saturation is measured using CATO, a normalized ratio of the oxygen channel to the total counts.1 The lengthy productive intervals in this area (e.g. 1000's of feet) require some practical limits be set on sampling times. In the log analysis examples, a single six-foot per minute pass acquired the entire data set, with a single repeat pass to check tool performance and repeatability. Repeats over zones of interest can be merged to minimize the Carbon/Oxygen measurement statistics. Several recent publications describe the pulsed-neutron porosities2 and the development of a density porosity3,4 based on scattering of inelastic gamma rays. The cased-hole porosities (neutron porosity referenced as RPHI, density porosity referenced as IPHI) and the clay volume are used to derive the effective porosity PHIE. The effective porosity is the total porosity (from the cross-plot) minus the clay bound porosity; thus it can be considered the porosity available for fluid movement. Sigma, the capture cross-section, is also recorded. Although the contrast of oil to water is limited by the low water salinity, it is still useful for clay volume and gas saturation. Geological setting Figure 1 is a map of the oil fields and major structural features in the western portion of the Basin. The ancient Ventura Basin was a trough-like deep-water basin, with the long axis in the current east-west direction5. This depocenter entrapped thousands of feet of sediment as the basin floor subsided during Pliocene time. Facies analysis suggest that the sands are turbidites deposited as elongated lenses, spread along the axis of the basin by westerly currents. The source of the detrital material was from uplifted Paleocene and Miocene sediments north of the Red Mountain and San Cayetano faults. The sands in the analysis examples are arkosic with angular feldspar and quartz grains. Organic detritus, such as coal and wood fragments, have been noted. Intervals of mud and silt debris flows and deposition of marine shales overlap the turbidite sands. Folding and faulting of these Pliocene sediments have created traps for the accumulation of hydrocarbons. Common reservoir compartments are anticlines and blocks truncated by faults. On a smaller scale, the reservoirs have stratgraphic compartments from features such as clay and calcite seals. Further, the grading and pinching-out of the lenticular sand bodies create depositional reservoir compartments.

show abstract

Shaly Sand Analysis using Density-Neutron Porosities from a Cased-hole Pulsed Neutron System

Cited by 4 publications

References 0 publications

Developing a Novel Hybrid Biogeography-Based Optimization Algorithm for Multilayer Perceptron Training under Big Data Challenge

Developing a Novel Hybrid Biogeography-Based Optimization Algorithm for Multilayer Perceptron Training under Big Data Challenge

Analysis and improvement of multiple optimal learning factors for feed-forward networks

Examples of Cased Reservoir Analysis in the Ventura Basin, California

Contact Info

Product

Resources

About