Acoustic impedance inversion by feedback artificial neural network

Baddari, Kamel; Djarfour, Noureddine; Aı̈fa, Tahar; Ferahtia, Jalal

doi:10.1016/j.petrol.2009.09.012

Cited by 22 publications

(7 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Kuzma and Rector (2004) implement nonlinear AVO inversion aided by a support vector machine to yield the nonlinear inverse operator. Reflectivity inversion, or impedance inversion, has also been implemented using machine learning methods such as neural networks (Röth and Tarantola, 1994;Calderon-Macias et al, 2000;Baddari et al, 2010), support vector machines (Rojo-Álvarez et al, 2008), and Bayesian learning (Ji et al, 2008;Yuan and Wang, 2013). The machine learning methods are proven to be effective when non-linearity or noise exist in the data.…”

Section: Yuji Kim 1 and Nori Nakatamentioning

confidence: 99%

Geophysical inversion versus machine learning in inverse problems

2018

View full text Add to dashboard Cite

Geophysical inversion and machine learning both provide solutions for inverse problems in which we estimate model parameters from observations. Geophysical inversions such as impedance inversion, amplitude-variation-with-offset inversion, and traveltime tomography are commonly used in the industry to yield physical properties from measured seismic data. Machine learning, a data-driven approach, has become popular during the past decades and is useful for solving such inverse problems. An advantage of machine learning methods is that they can be implemented without knowledge of physical equations or theories. The challenges of machine learning lie in acquiring enough training data and selecting relevant parameters, which are essential in obtaining a good quality model. In this study, we compare geophysical inversion and machine learning approaches in solving inverse problems and show similarities and differences of these approaches in a mathematical form and numerical tests. Both methods aid in solving ill-posed and nonlinear problems and use similar optimization techniques. We take reflectivity inversion as an example of the inverse problem. We apply geophysical inversion based on the least-squares method and artificial neural networks as a machine learning approach to solve reflectivity inversion using 2D synthetic data sets and 3D field data sets. A neural network with multiple hidden layers successfully generates the nonlinear mapping function to predict reflectivity. For this inverse problem, we test different L1 regularizers for both approaches. L1 regularization alleviates effects of noise in seismic traces and enhances sparsity, especially in the least-squares method. The 2D synthetic wedge model and field data examples show that neural networks yield high spatial resolution.

show abstract

Section: Yuji Kim 1 and Nori Nakatamentioning

confidence: 99%

Geophysical inversion versus machine learning in inverse problems

2018

View full text Add to dashboard Cite

show abstract

“…Geri beslemeli ağlarda bir sinirin çıkışı kendinden önceki katmanlarda bulunan sinirlerin girişine iletilmektedir [16]. Bu tip sinir ağlarının hafızaları dinamiktir ve çıkış hem anlık hem de önceki girişleri yansıtır.…”

Section: şEkil 2 Yapay Sinir Ağı Genel Yapısıunclassified

Sinter Makinesinin Örüntü Tanıma Tabanlı Otomatik Hız Kontrolü

Beskardes¹,

Ozdemir²

2017

deufmd

View full text Add to dashboard Cite

Özet: Sinter tesisleri, yüksek fırın kullanılarak demir çelik üretimi yapılan entegre demir çelik fabrikalarının önemli bir parçasıdır. Sinter tesislerinin ana bölümlerinden biri olan sinter makinesinde toz cevherin sinterleşmesi sağlanarak yüksek fırında kullanılmaya elverişli hale getirilir. Bu çalışmada, sinter makinesinin verimini artırmak amacıyla, makine yönetiminin operatörlerden alınıp otomatik yapılması için bir örüntü tanıma sistemi tasarlanmıştır. Tasarlanan bu sistemde sınıflandırıcı olarak doğrusal ayırma sınıflandırıcısı, destek vektör makineleri ve yapay sinir ağı kullanılmıştır. Bu sınıflandırıcılar 4212 adetlik bağımsız veri kümesi üzerinde test edilerek birbiri ile karşılaştırılmıştır. Automatically Sinter Machine Speed Control With Pattern RecognitionBased System Support Vector MachineAbstract: Sinter plant is one of the important unit of the integrated plants, which made iron and steel production via blast furnace. In the sinter machine, which is the main part of the sinter plant fine ore has been provided sintering and rendered suitable for using in the blast furnace. In this study, aimed to increasing sinter machine production a pattern recognition system has been developed for automatic control instead of operator's control. Linear discriminant classifier, support vector machines and artificial neural network is used in designed system. These classifiers were compared to each individual data set by testing over 4212 units. Sorumlu yazar: abeskardes@isdemir.com.tr

show abstract

“…Its applications in geophysics are described by Calderón-Macías et al (2000), van der Baan and Jutten (2000) and Poulton (2002). It is both implemented in seismic inversion (Röth and Tarantola, 1994;Langer et al, 1996;Baddari et al, 2010) and reservoir chacterization Saggaf et al, 2003;Leite and Vidal, 2011;Ahmadi et al, 2013;Mohamed et al, 2015).…”

Section: Brief Overview Of Conventional Algorithmsmentioning

confidence: 99%

Seismic inversion and uncertainty analysis using a transdimensional Markov chain Monte Carlo method

Zhu

Gibson

2016

SEG Technical Program Expanded Abstracts 2016

View full text Add to dashboard Cite

We use a transdimensional inversion algorithm, reversible jump MCMC (rjMCMC), in the seismic waveform inversion of post-stack and prestack data to characterize reservoir properties such as seismic wave velocity, density as well as impedance and then estimate uncertainty. Each seismic trace is inverted independently based on a layered earth model. The model dimensionality is defined as the number of the layers multiplied with the number of model parameters per layer. The rjMCMC is able to infer the number of model parameters from data itself by allowing it to vary in the iterative inversion process, converge to proper parameterization and prevent underparameterization and overparameterization. We also use rjMCMC to enhance uncertainty estimation since it can transdimensionally sample different model spaces of different dimensionalities and can prevent a biased sampling in only one space which may have a different dimensionality than that of the true model space. An ensemble of solutions from difference spaces can statistically reduce the bias for parameter estimation and uncertainty quantification. Inversion uncertainty is comprised of property uncertainty and location uncertainty. Our study revealed that the inversion uncertainty is correlated with the discontinuity of property in such a way that 1) a smaller discontinuity will induce a lower uncertainty in property at the discontinuity but also a higher uncertainty of the location of that discontinuity and 2) a larger discontinuity will induce a higher uncertainty in property at the discontinuity but also a higher ''certainty'' of the location of that discontinuity. Therefore, there is a trade-off between the property uncertainty and the location uncertainty. To our surprise, there is a lot of hidden information in the uncertainty result that we can actually take advantage of due to this trade-off effect. On the basis of our study ii using rjMCMC, we propose to use the inversion uncertainty as a novel attribute in an optimistic way to characterize the magnitude and the location of subsurface discontinuities and reflectors. important it is to estimate the uncertainty, and I used another method for two years but it turned out that method was unsuccessful for solving my research problems because it couldn't answer the question for uncertainty assessment. This is one of the biggest mistakes that I have ever made because I wasn't aware of the importance of my advisor's words and questions. Indeed, research is not always successful and smooth and I did encounter failures, but I didn't give up.

show abstract

Acoustic impedance inversion by feedback artificial neural network

Cited by 22 publications

References 17 publications

Geophysical inversion versus machine learning in inverse problems

Geophysical inversion versus machine learning in inverse problems

Sinter Makinesinin Örüntü Tanıma Tabanlı Otomatik Hız Kontrolü

Seismic inversion and uncertainty analysis using a transdimensional Markov chain Monte Carlo method

Contact Info

Product

Resources

About