Artificial Intelligent Approaches in Petroleum Geosciences by Constantin Cranganu, Henri Luchian, Mihaela Elena Breaban

By Constantin Cranganu, Henri Luchian, Mihaela Elena Breaban

This ebook offers numerous clever techniques for tackling and fixing demanding useful difficulties dealing with these within the petroleum geosciences and petroleum undefined. Written through skilled teachers, this e-book bargains cutting-edge operating examples and gives the reader with publicity to the most recent advancements within the box of clever tools utilized to grease and gasoline study, exploration and construction. It additionally analyzes the strengths and weaknesses of every approach awarded utilizing benchmarking, when additionally emphasizing crucial parameters resembling robustness, accuracy, velocity of convergence, desktop time, overlearning and the function of normalization. The clever techniques offered comprise synthetic neural networks, fuzzy good judgment, lively studying procedure, genetic algorithms and aid vector machines, among others.

Integration, dealing with info of substantial measurement and uncertainty, and working with chance administration are between the most important matters in petroleum geosciences. the issues we need to resolve during this area have gotten too complicated to depend upon a unmarried self-discipline for powerful recommendations and the prices linked to bad predictions (e.g. dry holes) raise. hence, there's a have to identify a brand new process aimed toward right integration of disciplines (such as petroleum engineering, geology, geophysics and geochemistry), facts fusion, possibility relief and uncertainty administration. those clever strategies can be utilized for uncertainty research, danger evaluation, facts fusion and mining, information research and interpretation, and information discovery, from different information similar to 3-D seismic, geological information, good logging, and construction information. This publication is meant for petroleum scientists, info miners, facts scientists and execs and post-graduate scholars interested in petroleum industry.

Show description

Read Online or Download Artificial Intelligent Approaches in Petroleum Geosciences PDF

Best mineralogy books

Röntgen-Pulverdiffraktometrie: Rechnergestützte Auswertung, Phasenanalyse und Strukturbestimmung (German Edition)

Die Beugung von Röntgenstrahlen wird heute in vielen Labors zur schnellen, sicheren und zerstörungsfreien Identifikation von Festkörperproben benutzt. Der Einsatz von Kleinrechnern (PCs) und von schnelleren Detektoren hat in der Röntgenpulverdiffraktometrie zu einer Renaissance sowohl in der Phasenanalyse von Gemischen als auch in der Strukturverfeinerung aus Pulverdaten geführt.

Chemical Elements in the Environment: Factsheets for the Geochemist and Environmental Scientist

How huge is the ordinary edition in focus of many of the components in numerous media? How do the oft-cited "World commonplace concentrations" in numerous media examine with real analytical information? How low a detection restrict do i have to reach if i need to examine for a component in soils, sediments, water or vegetation?

Isotopic Landscapes in Bioarchaeology

This paintings takes a severe examine the present inspiration of isotopic landscapes ("isoscapes") in bioarchaeology and its program in destiny study. It particularly addresses the learn strength of cremated unearths, a a little overlooked bioarchaeological substrate, ensuing essentially from the inherent osteological demanding situations and complicated mineralogy linked to it.

Additional resources for Artificial Intelligent Approaches in Petroleum Geosciences

Example text

Model=TRUE) and is used to generate distribution probabilities for each of the 12 entries of the test set by writing > pred_p <- predict(svm,testIris,type = "probabilities") Intelligent Data Analysis Techniques … 21 Note the use of the parameter kernel = “vanilladot”. We will explain later the use of kernels. 968759363 Note that in each case, one of the numbers strongly dominates the others, a consequence of the linear separability of this data set. Alternatively, a prediction that returns directly the class of various objects can be generated by pred <- predict(svm,testIris,type="response") and generates > pred [1] setosa versicolor [9] versicolor virginica [17] virginica setosa versicolor versicolor virginica virginica setosa setosa setosa versicolor versicolor versicolor virginica virginica virginica Levels: setosa versicolor virginica.

If the data are “almost” linearly separable, a separation hyperplane exists such that the majority (but not all) of the positive examples inhabit the positive halfspace of the hyperplane and the majority (but not all) of the negative examples inhabit the negative half-space. In this case, we shall seek a “separating hyperplane” that separates the two classes with the smallest error. This is achieved by assigning to each object xi in the data set a slack variable ni , where ni >0, by relaxing Inequalities 2 and 3 as 20 D.

Ann Mo Bot Gard 23:457–509 Bishop CM (2007) Pattern recognition and machine learning. Springer, New York Blumer A, Ehrenfeucht A, Haussler D, Warmuth MK (1989) Learnability and the vapnikchervonenkis dimension. J ACM 36(4):929–965 Breiman L, Friedman JH, Olshen RO, Stone CS (1998) Classification and regression trees. Chapman and Hall, Boca Raton (reprint edition) Cortes C, Vapnik V (1995) Support-vector networks. Mach Learn 20(3):273–297 50 D. Simovici Cristianini N, Shawe-Taylor J (2000) Support vector machines and other kernel-based learning methods.

Download PDF sample

Rated 4.01 of 5 – based on 23 votes