Deep learning for enhancing multi-source reverse time migration

<p>Reverse time migration (RTM) is a technique used to obtain high-resolution images of underground reflectors; however, this method is computationally intensive when dealing with large amounts of seismic data. Multi-source RTM can significantly reduce the computational cost by processing multiple shots simultaneously. However, multi-source-based methods frequently result in crosstalk artifacts in the migrated images, causing serious interference in the imaging signals. Plane-wave migration, as a mainstream multi-source method, can yield migrated images with plane waves in different angles by implementing phase encoding of the source and receiver wavefields; however, this method frequently requires a trade-off between computational efficiency and imaging quality. We propose a method based on deep learning for removing crosstalk artifacts and enhancing the image quality of plane-wave migration images. We designed a convolutional neural network that accepts an input of seven plane-wave images at different angles and outputs a clear and enhanced image. We built 505 1024×256 velocity models, and employed each of them using plane-wave migration to produce raw images at 0°, ±20°, ±40°, and ±60° as input of the network. Labels are high-resolution images computed from the corresponding reflectivity models by convolving with a Ricker wavelet. Random sub-images with a size of 512×128 were used for training the network. Numerical examples demonstrated the effectiveness of the trained network in crosstalk removal and imaging enhancement. The proposed method is superior to both the conventional RTM and plane-wave RTM (PWRTM) in imaging resolution. Moreover, the proposed method requires only seven migrations, significantly improving the computational efficiency. In the numerical examples, the processing time required by our method was approximately 1.6% and 10% of that required by RTM and PWRTM, respectively.</p>

Deep learning for enhancing multi-source reverse time migration

<p>Reverse time migration (RTM) is a technique used to obtain high-resolution images of underground reflectors; however, this method is computationally intensive when dealing with large amounts of seismic data. Multi-source RTM can significantly reduce the computational cost by processing multiple shots simultaneously. However, multi-source-based methods frequently result in crosstalk artifacts in the migrated images, causing serious interference in the imaging signals. Plane-wave migration, as a mainstream multi-source method, can yield migrated images with plane waves in different angles by implementing phase encoding of the source and receiver wavefields; however, this method frequently requires a trade-off between computational efficiency and imaging quality. We propose a method based on deep learning for removing crosstalk artifacts and enhancing the image quality of plane-wave migration images. We designed a convolutional neural network that accepts an input of seven plane-wave images at different angles and outputs a clear and enhanced image. We built 505 1024×256 velocity models, and employed each of them using plane-wave migration to produce raw images at 0°, ±20°, ±40°, and ±60° as input of the network. Labels are high-resolution images computed from the corresponding reflectivity models by convolving with a Ricker wavelet. Random sub-images with a size of 512×128 were used for training the network. Numerical examples demonstrated the effectiveness of the trained network in crosstalk removal and imaging enhancement. The proposed method is superior to both the conventional RTM and plane-wave RTM (PWRTM) in imaging resolution. Moreover, the proposed method requires only seven migrations, significantly improving the computational efficiency. In the numerical examples, the processing time required by our method was approximately 1.6% and 10% of that required by RTM and PWRTM, respectively.</p>

An optimised one-way wave migration method for complex media with arbitrarily varying velocity based on eigen-decomposition

The classical one-way generalised screen propagator (GSP) and Fourier finite-difference (FFD) schemes have limitations in imaging large angles in complex media with substantial lateral variations in wave velocity. Some improvements to the classical one-way wave scheme have been proposed with optimised methods. However, the performance of these methods in imaging complex media remains unsatisfying. To overcome this issue, a new strategy for wavefield extrapolation based on the eigenvalue and eigenvector decomposition of the Helmholtz operator is presented herein. In this method, the square root operator is calculated after the decomposition of the Helmholtz operator at the product of the eigenvalues and eigenvectors. Then, Euler transformation is applied using the best polynomial approximation of the trigonometric function based on the infinite norm, and the propagator for one-way wave migration is calculated. According to this scheme, a one-way operator can be computed more accurately with a lower-order expansion. The imaging performance of this scheme was compared with that of the classical GSP, FFD and the recently developed full-wave-equation depth migration (FWDM) schemes. The impulse responses in media with arbitrary velocity inhomogeneity demonstrate that the proposed migration scheme performs better at large angles than the classical GSP scheme. The wavefronts calculated in the dipping and salt dome models illustrate that this scheme can provide a precise wavefield calculation. The applications of the Marmousi model further demonstrate that the proposed approach can achieve better-migrated results in imaging small-scale and complex structures, especially in media with steep-dipping faults.

Managing Pipeline Integrity and Dynamic Free Spans on Mobile Seabed in the Southern North Sea

This paper describes the continuous improvement efforts to manage the integrity status of the Southern North Sea subsea pipeline system in the context of free spanning. The dynamic free-spanning threat is typically attributed to a mobile seabed. Current and wave action are constantly moving and eroding sediment by means of sand wave migration and scouring. It can lead to a fluctuation in span characteristics with respect to span length, span height and location over time. It makes pipeline integrity demonstration and spans remediation challenges. Focus areas include (1) identifying regions where operational pipelines are susceptible to critical span formation (2) understanding the broader context of seabed mobility, supported by several years of multibeam echo sound and met ocean data (3) risk-ranking & criticality of span formation (4) developing simplified calculation tool that allows fatigue damage to be estimated and accumulated for every location along the pipeline, conservatively (5) optimising and incorporating risk/event-based survey requirements (6) identification of suitable remediation solutions and developing a decision flow chart to facilitate selection of fit for purpose remediation solutions, with respect to span configuration and the surrounding seabed features. The outcome has improved the robustness of span management, reduced “reactive” span remediation activities, and allowed application of sound technical theory to allocate pipeline traffic light integrity status regarding the observed free spans.

Subsea Pipeline Engineering Challenges in Sand Wave Area: The Lufeng Feed Project

Lufeng oilfields are located in the Pearl River Mouth Basin, South China Sea, where significant sand wave is located. The water depth of the area is 140 to 330 m. Sand waves are present around LF15-1. A study on the sand waves is required to assess the impact of the sand waves on the pipeline design. Due to its special seabed characteristic, it is challenging for the subsea pipeline engineering. This paper presents the Lufeng sand wave pipeline project on general basis. Collect and review available information including metocean, bathymetric data and soil data and carry out general morphological analysis for the project area including seabed erodibility assessment and analysis of sediment transport potential. Identify morphological features and bed forms in the project area and analyze characteristics of the sand waves. Sand wave migration and mobility are predicted and the pipeline route (least dredging/trenching and least free spans) is optimize considering on-bottom stability, in-place strength, global buckling and installation. Determine burial (dredging/trenching) requirements assuring pipeline stability/integrity. The main challenges faced are summarized, some preliminary results are also presented. Discussions about the solutions are also included, which may shed light to similar projects.

Pressure Sensing Technique for Observing Seabed Deformation Caused by Submarine Sand Wave Migration

Long-term, continuous in-situ observation of seabed deformation plays an important role in studying the mechanisms of sand wave migration and engineering early warning methods. Research on pressure sensing techniques has examined the possibility of using the temporal characteristics of the vertical deformation of the seafloor to identify important factors (e.g., wave height and migration rate) of submarine sand wave migration. Two pressure sensing tools were developed in this study to observe the seabed deformation caused by submarine sand wave migration (a fixed-depth total pressure recorder (TPRFD) and a surface synchronous bottom pressure recorder (BPRSS)), based on the principle that as a sand wave migrates under hydrodynamic forcing, the near-bottom water pressure, bottom pressure and total fixed pressure synchronously change with time. Laboratory flume experiments were performed, using natural sandy sediments taken from the beach of Qingdao, China, to better present and discuss the feasibility and limitations of using these two pressure sensing methods to acquire continuous observations of seabed deformation. The results illustrate that the proposed pressure sensor techniques can be effectively applied in reflecting elevation caused by submarine sand wave migration (the accuracy of the two methods in observing the experimental bed morphology was more than 90%). However, an unexpected step-like process of the change in sand wave height observed by BPRSS is presented to show that the sensor states can be easily disturbed by submarine environments, and thus throw the validity of BPRSS into question. Therefore, the TPRFD technique is more worthy of further study for observing submarine sand wave migration continuously and in real-time.

Sand Wave Migration and its Factors on Giant Sand Ridge in Taiwan Strait

&lt;p&gt;&amp;#160;&amp;#160;&amp;#160;&amp;#160;&amp;#160;&amp;#160;&amp;#160; Submarine sand waves are known to be induced by tidal currents and their migration has become an important issue since it may affect seafloor installations. In Taiwan Strait, widely spreading sand waves have been recognized on the Changyun Ridge, a tide-dominated giant sand ridge offshore western Taiwan. However, due to lacking of high-resolution and repeated geophysical surveys before, detailed characteristics and migrating features of the sand waves in Taiwan Strait were poorly understood. As new multibeam bathymetric and seismic data were collected repeatedly during 2016 - 2018 for offshore wind farm projects, we can now advance the understanding of sand wave characteristics and migration patterns in the study area. We apply a geostatistical analysis method on bathymetry data to reveal distribution and spatial characteristics of the sand waves, and estimate its migration pattern by using an updated spatial cross-correlation method. Then, sedimentary features, internal structures and thicknesses of sand waves are observed and estimated on high-resolution seismic profiles. Our results show that the study area is mostly superimposed by multi-scaled sandy rhythmic bed forms. However, the geomorphological and migrating characteristics of the sand waves are complicated. Their wavelengths range from 80 to 200 m, heights range from 1.5 to 8 m, and crests are generally oriented in the WNW-ESE direction. Obvious sand wave migration was detected from repeated high-resolution multi-beam data between 2016 and 2018, and migration distances can be up to ~150 m in 15 months. The average elevation change of the seafloor over the whole survey area is ~3.0 m, with a maximum value of 6.9 m. Moreover, the sand waves can migrate over 30 m with ~2.5 m elevation change in 2 months and migrate over 5 m with ~1 m elevation change in 15 days. The results also show that the orientation of wave movement can be reversed even within a small distance. By identifying the base of sand wave on seismic profiles, the thicknesses of sand waves are found ranging from 1 to 10 meters. The base of wave structure become slightly deeper from nearshore to offshore. Our results indicate that the thickness of sand waves increases with degree of asymmetry and migration rate. By bathymetric and reflection seismic data analyses, systematic spatial information of sand waves in the study area are established, and we suggest that not only tidal currents can affect sand wave migration patterns, but also wave structures and thicknesses play important roles in sand wave migrating processes and related geomorphological changes.&lt;/p&gt;