Modeling extra-deep electromagnetic logs using a deep neural network

Alyaev, Sergey; Shahriari, Mostafa; Pardo, David; Omella, Ángel Javier; Larsen, David; Jahani, Nazanin; Suter, Erich

doi:10.1190/geo2020-0389.1

Cited by 20 publications

(11 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In these situations, the use of PDE solvers for forward modeling may not be a good choice because of the computational cost. Recently, it was shown that the DNN models could provide a good approximation of the electromagnetic geophysical logs [Alyaev et al, 2021, Shahriari et al, 2020a.…”

Section: Problem Descriptionmentioning

confidence: 99%

“…In this work, a DNN is utilized as the forward model during the real-time probabilistic inversion. We use the trained DNN described in Alyaev et al [2021]. The DNN model is trained on the dataset [ Alyaev et al, 2021] containing around seventy thousand samples, which are created using a high-fidelity physics-based simulator provided by the tool vendor [Sviridov et al, 2014].…”

Section: Problem Descriptionmentioning

confidence: 99%

“…We use the trained DNN described in Alyaev et al [2021]. The DNN model is trained on the dataset [ Alyaev et al, 2021] containing around seventy thousand samples, which are created using a high-fidelity physics-based simulator provided by the tool vendor [Sviridov et al, 2014]. The input for each sample is a 1D-layered environment as shown in Figure ( 1).…”

Section: Problem Descriptionmentioning

confidence: 99%

“…For example, Deep Learning (DL) has been used in an increasing number of physics-based computational problems like heat transfer [Tamaddon-Jahromi et al, 2020], stress concentration [Saha et al, 2021] and nano-fluids [Sheikholeslami et al, 2019]. Furthermore, DNNs showed relatively good approximation of the Maxwell's equations required for modeling extra deep electromagnetic (DeepEM) logging-whiledrilling (LWD) measurements [Alyaev et al, 2021].…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Probabilistic model-error assessment of deep learning proxies: an application to real-time inversion of borehole electromagnetic measurements

Rammay

Alyaev

Elsheikh

2022

Geophysical Journal International

Self Cite

View full text Add to dashboard Cite

The advent of fast sensing technologies allow for real-time model updates in many applications where the model parameters are uncertain. Once the observations are collected, Bayesian algorithms offer a pathway for real-time inversion (a.k.a. model parameters/inputs update) because of the flexibility of the Bayesian framework against non-uniqueness and uncertainties. However, Bayesian algorithms rely on the repeated evaluation of the computational models and deep learning based proxies can be useful to address this computational bottleneck. In this paper, we study the effects of the approximate nature of the deep learned models and associated model errors during the inversion of borehole electromagnetic measurements, which are usually obtained from logging while drilling (LWD). We rely on the iterative ensemble smoothers as an effective algorithm for real-time inversion due to its parallel nature and relatively low computational cost. The real-time inversion of electromagnetic measurements is used to determine the subsurface geology and properties, which are critical for real-time adjustments of the well trajectory (geosteering). The use of deep neural network as a forward model allows us to perform thousands of model evaluations within seconds, which is very useful to quantify uncertainties and non-uniqueness in real-time. While significant efforts are usually made to ensure the accuracy of the deep learning models, it is widely known that the deep neural networks can contain some type of model-error in the regions not covered by the training data, which are unknown and training specific. When the deep learning models are utilized during inversion of electromagnetic measurements, the effects of the model-errors could manifest themselves as a bias in the estimated input parameters and as a consequence might result in a low-quality geosteering decision. We present numerical results highlighting the challenges associated with the inversion of electromagnetic measurements while neglecting model-error. We further demonstrate the utility of a recently proposed flexible iterative ensemble smoother in reducing the effect of model-bias by capturing the unknown model-errors, thus improving the quality of the estimated subsurface properties for geosteering operation. Moreover, we describe a procedure for identifying inversion multi-modality and propose possible solutions to alleviate it in real-time.

show abstract

Section: Problem Descriptionmentioning

confidence: 99%

Section: Problem Descriptionmentioning

confidence: 99%

Section: Problem Descriptionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Probabilistic model-error assessment of deep learning proxies: an application to real-time inversion of borehole electromagnetic measurements

Rammay

Alyaev

Elsheikh

2022

Geophysical Journal International

Self Cite

View full text Add to dashboard Cite

show abstract

“…These traditional methods evaluate the inverse solution pointwise (i.e., for a given set of measurements), but they rarely provide a global representation of the inverse operator. To overcome this problem and approximate the full inverse function, it is possible to use Deep Learning (DL) methods (see, e.g., [9,10,11,12,13,14]), which allow to approximate complex mappings via a composition of linear and non-linear functions.…”

Section: Introductionmentioning

confidence: 99%