3D Prestack Full Waveform Inversion

Vigh, Denes; Starr, E. William; Dingwall, Kenneth

doi:10.4043/otc-19897-ms

Cited by 3 publications

(4 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…where the search direction hi is determined by conjugate gradients (Vigh & Starr, 2008;Gong et al, 2008):…”

Section: Fwi As An Optimization Problemmentioning

confidence: 99%

“…Because the gradient calculation for full waveform inversion is similar to the process of reverse time migration (RTM) (Tarantola & Valette, 1982;Pratt, 1999), a straightforward approach is to perform FWI using an RTM engine. Vigh & Starr (2008) note that the advantages of implementing FWI in the time domain include increased parallelism and reduced memory requirements, thereby making FWI more applicable to large 3D models and data sets. In the examples shown in this report, we used RTM and implemented FWI in the time domain.…”

Section: Fwi As An Optimization Problemmentioning

confidence: 99%

“…In step (iii), we then compute a step length α0 that determines how much to change our velocity model in this first iteration. We compute the step length using a quadratic line search algorithm and search in a direction defined by conjugate gradients (Vigh & Starr, 2008;Gong et al, 2008). This line search requires computation of at least 2 synthetic data sets.…”

Section: Synthetic Examplementioning

confidence: 99%

See 2 more Smart Citations

Full waveform inversion with image‐guided gradient

Hale

Meng

et al. 2010

SEG Technical Program Expanded Abstracts 2010

View full text Add to dashboard Cite

show abstract

“…where the search direction hi is determined by conjugate gradients (Vigh & Starr, 2008;Gong et al, 2008):…”

Section: Fwi As An Optimization Problemmentioning

confidence: 99%

Section: Fwi As An Optimization Problemmentioning

confidence: 99%

Section: Synthetic Examplementioning

confidence: 99%

See 1 more Smart Citation

Full waveform inversion with image‐guided gradient

Hale

Meng

et al. 2010

SEG Technical Program Expanded Abstracts 2010

View full text Add to dashboard Cite

show abstract

“…Potentially, the FWI (Full-Waveform Inversion) methodology can generate high-resolution velocity models (Vigh & Starr, 2008), which are currently appropriate for the RTM (Reverse Time Migration) technique. However, according to Vigh et al (2009), one of the main challenges for the FWI techniques is to produce (or to reproduce) a good initial velocity model to generate the seismic images with geological confidence regarding the subsurface geology.…”

Section: Introductionmentioning

confidence: 99%

Improving Pre-Salt Reservoirs Seismic Images When Considering the Stratified Evaporites Insertion in the Initial Model for the Velocity Updating Processes Prior to the Seismic Migration

Maul¹,

Santos²,

Silva³

et al. 2019

Braz J Geophys

View full text Add to dashboard Cite

ABSTRACT. Structurally complex areas, such as the pre-salt section in the offshore Santos Basin, SE Brazil, is a challenge to represent the geology using seismic images. One of the main causes of the observed imaging problems is the evaporitic section and its considerations about velocities used for seismic migration purposes. Some authors consider set to this section an almost constant value (close to 4,500 m/s) which approximately represents the halite velocity, the most abundant mineral in this salt formation. Others, over these models, apply the tomographic inversion or FWI schemes giving to the velocity model the mathematical support to build confident seismic images. We believe in the importance to build starting velocity models reflecting the existing geological features prior to applying the tomographic/FWI updating. In this sense, we propose the insertion of the so-called stratifications within the evaporitic section using an adaptation of the model-based seismic inversion technique. Following this new velocity model including the stratification, we suggest tomographic iterations update, or FWI, to add to the geological constrains of the model the needed mathematical convergence. Finally, in this work, we performed the seismic migration with and without inserting these geological features in the initial velocity model and compared the results.Keywords: evaporitic section, stratifications, velocity model, seismic migration, seismic image.RESUMO. Em áreas estruturalmente complexas, como na seção pré-sal da Bacia offshore de Santos, região SE do Brasil, é um desafio representar a geologia utilizando imagens sísmicas. Uma das principais causas dos problemas observados está nas considerações sobre a seção evaporítica e suas velocidades com propósito de migração sísmica. Alguns autores consideram esta seção como tendo velocidades aproximadamente constantes (próximas de 4.500 m/s), o que representa aproximadamente o comportamento da halita, o mineral mais abundante nesta seção. Outros, sobre este modelo aplicam a atualização por inversão tomográfica ou FWI para dar ao modelo de velocidades o suporte matemático necessário para construir imagens sísmicas confiáveis. Nós acreditamos na importância de construir modelos iniciais de velocidades que reflitam as características geológicas existentes antes de aplicar esta atualização tomográfica/FWI mencionada. Neste sentido, propomos a inserção das denominadas estratificações dentro da seção evaporítica, utilizando uma adaptação da técnica de inversão sísmica model-based. Seguindo este novo modelo incluindo as estratificações, sugerimos a atualização por iterações tomográficas, ou FWI, para adicionar ao controle geológico do modelo a convergência matemática necessária. Finalmente, neste trabalho, nós realizamos a migração com e sem a inserção destas características geológicas no modelo inicial de velocidades e comparamos os resultados.Palavras-chave: seção evaporítica, estratificações, modelo de velocidade, migração sísmica, imagem sísmica.

show abstract

Next Generation Broadband Land Acquisition

Baeten

Hwang

Walk

2014

Abu Dhabi International Petroleum Exhibition and Conference

View full text Add to dashboard Cite

The acquisition of broadband full azimuth surveys in a desert environment requires the application of several new technologies. These technologies can be divided in three main fields: the seismic receiver, the seismic source and the processing algorithms. We will discuss these new technologies, and show that the developments towards high channel counts and low frequencies are intimately related. Hence, an integrated approach is required to make optimum use of new emerging technologies. We also show some applications of the technology, demonstrating the value that these novel acquisition techniques can provide.Source -Desert surveys almost exclusively use Vibroseis as a source. Over the last two years, the use of low frequencies in onshore seismic acquisition and processing has become more and more a routine practice. However, the emission of low frequencies by seismic vibrators is far from trivial and required a number of new technological developments and field tests. We will discuss the developments in vibrator technology, the limitations that are still in existence today and possible future developments.Receiver -Current developments in land seismic acquisition show a trend towards higher channel counts and increased bandwidth. In addition, the increase in use of single sensors provides potential for acquiring higher resolution data, and improved signal preservation at the higher frequencies.For broadband desert surveys, to significantly improve data fidelity, data sampling needs to be much improved by much increasing the channel count and including very low frequencies down to 1Hz. Shell is investing in the development of broad band land sensors that records to very low frequencies (Ͻ 1Hz) and are cost-effectively scalable to 1 million channels. Single, light weighted, sensor systems will be much cheaper and faster to deploy, with less staff and lower HSE footprint. We will describe the status and report on recent field trial in the Middle East.Both the full azimuth aspect of the surveys and the low frequency content require the application of dedicated processing algorithms. These processes vary from a geophone amplitude-and phase correction early in the processing chain, to Full Waveform Inversion and processing techniques that account for azimuthal variations at later stages in the processing. Applications of broadband full azimuth technologies will be shown, demonstrating the benefits that a bandwidth extension can provide to the seismic data in terms of increased resolution and structural definition.

show abstract

3D Prestack Full Waveform Inversion

Cited by 3 publications

References 0 publications

Full waveform inversion with image‐guided gradient

Full waveform inversion with image‐guided gradient

Improving Pre-Salt Reservoirs Seismic Images When Considering the Stratified Evaporites Insertion in the Initial Model for the Velocity Updating Processes Prior to the Seismic Migration

Next Generation Broadband Land Acquisition

Contact Info

Product

Resources

About