### Recent Submissions

• #### IntraTomo: Self-supervised Learning-based Tomography via Sinogram Synthesis and Prediction

(IEEE, 2021-09-10) [Conference Paper]
We propose IntraTomo, a powerful framework that combines the benefits of learning-based and model-based approaches for solving highly ill-posed inverse problems, in the Computed Tomography (CT) context. IntraTomo is composed of two core modules: a novel sinogram prediction module and a geometry refinement module, which are applied iteratively. In the first module, the unknown density field is represented as a continuous and differentiable function, parameterized by a deep neural network. This network is learned, in a self-supervised fashion, from the incomplete or/and degraded input sinogram. After getting estimated through the sinogram prediction module, the density field is consistently refined in the second module using local and non-local geometrical priors. With these two core modules, we show that IntraTomo significantly outperforms existing approaches on several ill-posed inverse problems, such as limited angle tomography with a range of 45 degrees, sparse view tomographic reconstruction with as few as eight views, or super-resolution tomography with eight times increased resolution. The experiments on simulated and real data show that our approach can achieve results of unprecedented quality.
• #### EM-Based 2D Corrosion Azimuthal Imaging using Physics Informed Machine Learning PIML

(SPE, 2021-09-07) [Conference Paper]
In the wake of today's industrial revolution, many advanced technologies and techniques have been developed to address the complex challenges in well integrity evaluation. One of the most prominent innovations is the integration of physics-based data science for robust downhole measurements. This paper introduces a promising breakthrough in electromagnetism-based corrosion imaging using physics informed machine learning (PIML), tested and validated on the cross-sections of real metal casings/tubing with defects of various sizes, locations, and spacing. Unlike existing electromagnetism-based inspection tools, where only circumferential average metal thickness is measured, this research investigates the artificial intelligence (AI)-assisted interpretation of a unique arrangement of electromagnetic (EM) sensors. This facilitates the development of a novel solution for through-tubing corrosion imaging that enhances defect detection with pixel-level accuracy. The developed framework incorporates a finite-difference time-domain (FDTD)-based EM forward solver and an artificial neural network (ANN), namely the long short-term memory recurrent neural network (LSTM-RNN). The ANN is trained using the results generated from the FDTD solver, which simulates sensor readings for different scenarios of defects. The integration of the array EM-sensor responses and an ANN enabled generalizable and accurate measurements of metal loss percentage across various experimental defects. It also enabled the precise predictions of the defects’ aperture sizes, numbers, and locations in 360-degree coverage. Results were plotted in customized 2D heat-maps for any desired cross-section of the test casings. Further analysis of different techniques demonstrated that the LSTM-RNN could show higher precision and robustness compared to regular dense NNs, especially in the case of multiple defects. The LSTM-RNN is validated using additional data from simulated and experimental data. The results show reliable predictions even with limited training data. The model accurately predicted defects of larger and smaller sizes that were intentionally excluded from the training data to demonstrate generalizability. This highlights a major advance toward corrosion imaging behind tubing. This novel technique paves the way for the use of similar concepts on other sensors in multiple barriers imaging. Further work includes improvement to the sensor package and ANNs by adding a third dimension to the imaging capabilities to produce 3D images of defects on casings.
• #### A Real-Time Fiber Optical System for Wellbore Monitoring: A Johan Sverdrup Case Study

(SPE, 2021-09-07) [Conference Paper]
Fiber Optic (FO) sensing capabilities for downhole monitoring include, among other techniques, Distributed Temperature Sensing (DTS) and Distributed Acoustic Sensing (DAS). The appeal of DTS and DAS data is based on its high temporal and spatial sampling, allowing for very fine localization of processes in a wellbore. Furthermore, the broad frequency spectrum that especially DAS data is acquired with, enables observations, ranging from more continuous effects like oil flow, to more distinct effects like opening and closing of valves. Due to the high data volume of hundreds of Gb per well per hour, DAS data has traditionally been acquired acquisition-based, where data is recorded for a limited amount of time and processed at a later point in time. This limits the decision-making capability based on this data as reacting to events is only possible long after the event occurred. Equinor has addressed these decision-making shortcomings by building a real-time streaming solution for transferring, processing, and interpretation of its FO data at the Johan Sverdrup field in the North Sea. The streaming solution for FO data consists of offshore interrogators streaming raw DAS and DTS data via a dedicated bandwidth to an onshore processing cluster. There, DAS data is transformed into FO feature data, e.g., Frequency Band Energies, which are heavily decimated versions of the raw data; allowing insight extraction, while significantly reducing data volumes. DTS and DAS FO feature data are then streamed to a custom-made, cloud-based visualization and integration platform. This cloud-based platform allows efficient inspection of large data sets, control and evaluation of applications based on these data, and sharing of FO data within the Johan Sverdrup asset. During the last year, this FO data streaming pipeline has processed several tens of PB of FO data, monitoring a range of well operations and processes. Qualitatively, the benefits and potential of the real-time data acquisitions have been illustrated by providing a greater understanding of current well conditions and processes. Alongside the FO data pipeline, multiple prototype applications have been developed for automated monitoring of Gas Lift Valves, Safety Valve operations, Gas Lift rate estimation, and monitoring production start-up, all providing insights in real-time. For certain use cases, such as monitoring production start-up, the FO data provides a previously non-existent monitoring solution. In this paper, we will discuss in detail the FO data pipeline architecture from-platform-to-cloud, illustrate several data examples, and discuss the way-forward for "real-time" FO data analytics.
• #### Dual-band generative learning for low-frequency extrapolation in seismic land data

(Society of Exploration Geophysicists, 2021-09-01) [Conference Paper]
The presence of low-frequency energy in seismic data can help mitigate cycle-skipping problems in full-waveform inversion. Unfortunately, the generation and recording of low-frequency signals in seismic exploration remains a non-trivial task. Extrapolation of missing low-frequency content in field data might be addressed in a data-driven framework. In particular, deep learning models trained on synthetic data could be used for inference on the field data. Such an implementation of switching application domains remains challenging. We, therefore, propose the concept of generative dual-band learning to facilitate the knowledge transfer between synthetic and field seismic data applications of low-frequency data extrapolation. We first explain the two-step procedure for training a generative adversarial network (GAN) that extrapolates low frequencies. Then, we describe the workflow for synthetic dataset generation. Finally, we explore the feasibility of the dual-band learning concept on real near-surface land data acquired in Saudi Arabia. The presence of low-frequency energy in seismic data can help mitigate cycle-skipping problems in full-waveform inversion. Unfortunately, the generation and recording of low-frequency signals in seismic exploration remains a non-trivial task. Extrapolation of missing low-frequency content in field data might be addressed in a data-driven framework. In particular, deep learning models trained on synthetic data could be used for inference on the field data. Such an implementation of switching application domains remains challenging. We, therefore, propose the concept of generative dual-band learning to facilitate the knowledge transfer between synthetic and field seismic data applications of low-frequency data extrapolation. We first explain the two-step procedure for training a generative adversarial network (GAN) that extrapolates low frequencies. Then, we describe the workflow for synthetic dataset generation. Finally, we explore the feasibility of the dual-band learning concept on real near-surface land data acquired in Saudi Arabia.
• #### Traveltime computation using a supervised learning approach

(Society of Exploration Geophysicists, 2021-09-01) [Conference Paper]
In real-time microseismic monitoring, the ability to efﬁcientlycompute source-receiver traveltimes can help in signiﬁcantlyspeeding up the model calibration and hypocenter determina-tion processes, thus ensuring timely information about the sub-surface fractures for use in effective decision making. Here,we present a supervised-learning based traveltime computationapproach for layered 1D velocity models. First, we generatenumerous synthetic traveltime examples from a combinationof source locations and layered subsurface models, coveringa broad range of realistic P-wave velocities (2500–5000 m/s).Next, we train a multi-layered feed-forward neural network us-ing the training set containing source locations and velocitiesas input and traveltimes as corresponding labels. By doing so,we aim for a neural-network model that is trained only onceand can be applied to a wide range of subsurface velocitiesas well as source-receiver positions to predict fast and accu-rate traveltimes. We apply the trained model on numeroustest examples to validate the accuracy and speed of the pro-posed method. Based on the comparisons with acoustic ﬁnite-difference modeling and a ray-shooting method, we show thatthe trained model can provide faster and reasonably accuratetraveltimes for any realistic model scenario within the trainedvelocity range.
• #### High-dimensional wavefield solutions based on neural network functions

(Society of Exploration Geophysicists, 2021-09-01) [Conference Paper]
Waveﬁeld solutions are critical for applications ranging from imaging to full waveform inversion. These wave-ﬁelds are often large, especially for 3D media, and multiple p o int sources, like Green’s functions. A recently introduced framework based on neu ra l networks admit-ting functional solutions to partial differential equations(PDEs) o↵ers the opportunity to solve the Helmholtz equation with a neural network (NN) model. The input to such an NN is a location in space and the output are the real and imaginary parts of the scattered waveﬁeldat that location, thus, acting like a function. The net-work is trained on random input points in space and a variance of the Helmh o lt z equation for the scatteredwaveﬁeld is used as the loss function to update the network parameters. In spite of the methods ﬂexibility, like handling irregular surfaces and complex media, and its potential for velocity model building, the cost of training the network far exceeds that of numerical solutions. Relying on the network’s ability to learn waveﬁeld features, we extend the dimension of this NN function to learn the waveﬁeld for many sources and frequencies, simultaneously. We show, in this preliminary study, that reasonable waveﬁeld solutions can be predicted using smaller networks. This includes waveﬁelds for frequencies not within the training range. The new NN function has the potential to eﬃciently represent the waveﬁeld as a function of location in space, as well as source location and frequency.
• #### Self-supervised learning for random noise suppression in seismic data

(Society of Exploration Geophysicists, 2021-09-01) [Conference Paper]
A staunch companion to seismic signals, noise consistently hinders processing and interpretation of seismic data. Borrowing ideas from the field of computer vision, we propose the use of self-supervised deep learning for the task of random noise suppression. These techniques require no clean training data and therefore remove any requirement of pre-cleaning of field data or the generation of realistic synthetic datasets for training purposes. Through the use of blind-spot networks, we show that self-supervised Noise2Void (N2V) procedure can be adapted to the seismic context, and trained solely on noisy data. An initial validation performed on a synthetic dataset corrupted by additive, white, Gaussian noise confirms that N2V can be trained to accurately separate the correlated seismic signal from the uncorrelated noise. Furthermore, when correlated and random noise are both present in the data, whilst the model cannot remove the majority of the correlated noise, a portion of it is suppressed alongside the random noise. Finally, the network is validated on a field dataset that is heavily contaminated with strong random noise caused by the surface conditions. The N2V denoising approach is shown to drastically reduce the random noise in the data. Through these examples, we have validated the effectiveness of blind-spot networks on highly oscillating signals, such as seismic data. This pave the way for the application of other self-supervised procedures to seismic data that go beyond the assumption of statistically independent noise.
• #### Misfit functions based on differentiable dynamic time warping for waveform inversion

(Society of Exploration Geophysicists, 2021-09-01) [Conference Paper]
Misfit functions based on differentiable dynamic time warping (DTW) have demonstrated an excellent performance in various sequence-related tasks. We introduce this concept in the context of waveform inversion and discuss a fast algorithm to calculate the first and second derivatives of such misfits. The DTW distance is originally calculated by solving a series of min sorting problems, thus it is not differentiable. The fundamental idea of differentiable DTW is replacing the min operator with its smooth relaxation. It is then straightforward to solve the derivatives following the differentiation rules, which results in both the differentiable misfit measurements and warping path. This path weights the traveltime mismatch more than its amplitude counterpart. We can construct a penalization term based on this warping path. The penalized misfit function is adaptable to measure traveltime and amplitude separately. Numerical examples show the flexibility and the performance of the proposed misfit function in mitigating local minima issues and retrieving the long-wavelength model features
• #### Seismic wavefield processing with deep preconditioners

(Society of Exploration Geophysicists, 2021-09-01) [Conference Paper]
In the last decade, seismic wavefield processing has begun to rely more heavily on the solution of wave-equation-based inverse problems. Especially when dealing with unfavourable data acquisition conditions (e.g., poor, regular or irregular sampling of sources and/or receivers), the underlying inverse problem is generally very ill-posed; sparsity promoting inversion coupled with fixed-basis sparsifying transforms has become the de-facto approach for many processing algorithms. Motivated by the ability of deep neural networks to identify compact representations of N-dimensional vector spaces, we propose to learn a mapping between the input seismic data and a latent manifold by means of an Autoencoder. The trained decoder is subsequently used as a nonlinear preconditioner for the inverse problem we wish to solve. Using joint deghosting and data reconstruction as an example, we show that nonlinear learned transforms outperform fixed-basis transforms and enable faster convergence to the sought solution (i.e, fewer applications of the forward and adjoint operators are required).
• #### Tomographic deconvolution of reflection tomograms

(Society of Exploration Geophysicists, 2021-09-01) [Conference Paper]
We present a tomographic deconvolution procedure for highresolution imaging of velocity anomalies between reflecting interfaces. The key idea is to first invert reflection or transmission traveltimes for the background velocity model. A convolutional neural network (CNN) model is then trained to estimate the inverse to the blurred tomogram consisting of small scatterers in the background velocity model. We call this CNN a tomographic deconvolution operator because it deconvolves the blurring artifacts in traveltime slowness tomograms. This procedure is similar to that of migration deconvolution which deconvolves the migration butterfly artifacts in migration images. Results with synthetic examples show the effectiveness of this procedure in significantly sharpening the tomographic images of small scatterers.
• #### Target-oriented time-lapse elastic full-waveform inversion assisted by deep learning with prior information

(Society of Exploration Geophysicists, 2021-09-01) [Conference Paper]
Time-lapse (TL) monitoring of the elastic property changes in the reservoir of interest is important for optimizing the reservoir interpretation and development plan. Given that elastic full-waveform inversion (EFWI) provides quantitative estimations of the elastic properties (Vp and Vs), its application to time-lapse elastic data is of considerable interest. For practical applications in reservoir monitoring, we need EFWI to provide high-resolution reservoir information at a reasonable cost. Thus, we develop an elastic redatuming technique to provide the required virtual elastic data for a target-oriented inversion, thus improving the computational efficiency by focusing our full-band inversion on the target zone. To improve the inversion resolution, we combine the well information and seismic data in the proposed time-lapse inversion approach using a regularized objective function. To derive the required prior model, we train a deep neural network (DNN) to learn the connection between the seismic estimation and the facies interpreted from well logs. We then apply the trained network to the target inversion domain to predict a prior model. Given the prior model, we perform another time-lapse inversion. We fit the simulated data difference for the virtual survey to the redatumed one from the surface recording and fit the model changes to the predicted prior model. The numerical results demonstrate that the proposed method enables the recovery of the time-lapse changes effectively in the target zone by incorporating the learned model changes from well logs.
• #### A Graph-based Approach for Trajectory Similarity Computation in Spatial Networks

(ACM, 2021-08-14) [Conference Paper]
Trajectory similarity computation is an essential operation in many applications of spatial data analysis. In this paper, we study the problem of trajectory similarity computation over spatial network, where the real distances between objects are reflected by the network distance. Unlike previous studies which learn the representation of trajectories in Euclidean space, it requires to capture not only the sequence information of the trajectory but also the structure of spatial network. To this end, we propose GTS, a brand new framework that can jointly learn both factors so as to accurately compute the similarity. It first learns the representation of each point-of-interest (POI) in the road network along with the trajectory information. This is realized by incorporating the distances between POIs and trajectory in the random walk over the spatial network as well as the loss function. Then the trajectory representation is learned by a Graph Neural Network model to identify neighboring POIs within the same trajectory, together with an LSTM model to capture the sequence information in the trajectory. We conduct comprehensive evaluation on several real world datasets. The experimental results demonstrate that our model substantially outperforms all existing approaches.
• #### Socially-Aware Self-Supervised Tri-Training for Recommendation

(ACM, 2021-08-14) [Conference Paper]
Self-supervised learning (SSL), which can automatically generate ground-truth samples from raw data, holds vast potential to improve recommender systems. Most existing SSL-based methods perturb the raw data graph with uniform node/edge dropout to generate new data views and then conduct the self-discrimination based contrastive learning over different views to learn generalizable representations. Under this scheme, only a bijective mapping is built between nodes in two different views, which means that the self-supervision signals from other nodes are being neglected. Due to the widely observed homophily in recommender systems, we argue that the supervisory signals from other nodes are also highly likely to benefit the representation learning for recommendation. To capture these signals, a general socially-aware SSL framework that integrates tri-training is proposed in this paper. Technically, our framework first augments the user data views with the user social information. And then under the regime of tri-training for multi-view encoding, the framework builds three graph encoders (one for recommendation) upon the augmented views and iteratively improves each encoder with self-supervision signals from other users, generated by the other two encoders. Since the tri-training operates on the augmented views of the same data sources for self-supervision signals, we name it self-supervised tri-training. Extensive experiments on multiple real-world datasets consistently validate the effectiveness of the self-supervised tri-training framework for improving recommendation. The code is released at https://github.com/Coder-Yu/QRec.
• #### An empirical analysis of the progress in wireless communication generations

(ACM, 2021-08-09) [Conference Paper]
The controversy and argument on the usefulness of the physical layer (PHY) academic research for wireless communications are long-standing since the cellular communication paradigm gets to its maturity. In particular, researchers suspect that the performance improvement in cellular communications is primarily attributable to the increases in telecommunication infrastructure and radio spectrum instead of the PHY academic research, whereas concrete evidence is lacking. To respond to this controversy from an objective perspective, we employ econometric approaches to quantify the contributions of the PHY academic research and other performance determinants. Through empirical analysis and the quantitative evidence obtained, albeit preliminary, we shed light on the following issues: 1) what determines the cross-national differences in cellular network performance; 2) to what extent the PHY academic research and other factors affect cellular network performance; 3) what suggestions we can obtain from the data analysis for the stakeholders of the PHY research.
• #### An Effective Wind Power Prediction using Latent Regression Models

(IEEE, 2021-08-02) [Conference Paper]
Wind power is considered one of the most promising renewable energies. Efficient prediction of wind power will support in efficiently integrating wind power in the power grid. However, the major challenge in wind power is its high fluctuation and intermittent nature, making it challenging to predict. This paper investigated and compared the performance of two commonly latent variable regression methods, namely principal component regression (PCR) and partial least squares regression (PLSR), for predicting wind power. Actual measurements recorded every 10 minutes from an actual wind turbine are used to demonstrate the prediction precision of the investigated techniques. The result showed that the prediction performances of PCR and PLSR are relatively comparable. The investigated models in this study can represent a helpful tool for model-based anomaly detection in wind turbines.
• #### Skewness effects on the turbulence structure in a high-speed compressible and multi-component inert mixing layers

(American Institute of Aeronautics and Astronautics, 2021-07-28) [Conference Paper]
This work presents the analysis of the effects of the misalignment angle between two asymptotic streams of fluid, $\zeta$, whose interaction leads to a turbulent mixing region. In fact, spatially evolving mixing layers may see their turbulent structure statistics altered in the presence of the skew angle $\zeta$. The investigation is conducted by analyzing a new set of direct numerical simulations of spatially-developing compressible non-reactive hydrogen--air shear layers. To assess the effects associated to misalignment angle, the turbulent structure statistics of a skewed configuration with $\zeta=15^{\circ}$ are compared to the reference case where no skewness is introduced. The analysis of the mixing layer time-averaged statistics reveals the ability of the skewness to accelerate the inlet structures growth which, consequently, yields to a substantial enhancement of the mixing efficiency.
• #### In Silico and In Vitro Experiments on Chevron Nozzles with Enhanced Momentum Thrust using Streamtube Expansion Waves

(American Institute of Aeronautics and Astronautics, 2021-07-28) [Conference Paper]
Noise regulations around the airport and rocket launching stations due to the environmental concern have made jet noise a crucial problem in the present day aeroacoustics research aiming to mitigate the undesirable jarring sound to avoid health hazards. The present paper is a continuation of our previous connected paper (AIAA 2019-4070). In this manuscript comprehensive in silico and in vitro studies on the jet acoustic characterization of chevron nozzles for its geometry optimization have been carried out. In silico studies have been carried out using validated steady 3d, double precision, density-based implicit, SST k-ɷ turbulence model with heat transfer effect. In this study, the fully implicit finite volume scheme of compressible, Navier–Stokes equations are employed. As a part of the code verification and calibration the numerically predicted 3d boundary layer blockage at the Sanal flow choking conditions for a channel flow is verified with the closed form analytical model (V.R.S.Kumar et al., Nature Scientific Reports, 2021) and found good agreement with the benchmark data. Comprehensive in silico and in vitro experiments reveal that the suitable selection of the number of lobes in a chevron nozzle with aerodynamically designed tapered tip creating streamtube expansion wave is a meaningful objective for achieving the jet noise reduction coupled with enhanced momentum thrust to the aircraft and rockets. In our case study, we found that the chevron lobes facilitated with tapered-tip with 80 angle could enhance the momentum thrust by 6 % and reduce the acoustic power level by 5 % while comparing with the base model. We concluded that the chevron cap with tapered tip creating streamtube expansion waves (V.R.S.Kumar et al., Physics of Fluids, 2021, doi: 10.1063/5.0040440) at the exit of the chevron nozzle is a profitable option for attenuation of the acoustic power level with momentum thrust benefits. This paper is a pointer towards for the geometry optimization of environmental friendly chevron nozzles with tapered tip for generating streamtube expansion waves for improving the momentum thrust while attenuating the acoustic power level for the future orbital and sub-orbital launch vehicle applications without sacrificing the mission demanding thrust-time requirements.
• #### @What does a virus actually look like?

(ACM, 2021-07-27) [Image]
This is the first time you can see real (flash-frozen) coronavirus in 3D.
• #### A Multilayer Perceptron-based Carotid-to-Femoral Pulse Wave Velocity Estimation using PPG Signal

(IEEE, 2021-07-27) [Conference Paper]
Cardiovascular diseases (CVDs) are the primary cause of death in the world. The development of easy-to-use and non-invasive monitoring CVDs’ diagnosis methods is crucial. Among the key parameters in the cardiovascular system is arterial stiffness. An increase in arterial stiffness is considered a primary risk factor for CVDs. Although arterial stiffness can be assessed non-invasively by measuring the carotid-to-femoral pulse wave velocity (cf−PWV), which is considered as a gold standard for arterial stiffness measurement, the clinical process of assessing this parameter is very intrusive and complicated. This paper investigated the potential of estimating (cf−PWV) from distal photoplethysmogram (PPG) waveforms using regression technique based on a multilayer perceptron. Functionally, PPG offers a simple, reliable, low-cost technique to measure blood volume change and hence assess cardiovascular function. In this work, we identify and select features from the timing fiducial points-based PPG, its first, second, and third derivative waveforms. The in-silico validation shows promising results and satisfactory accuracy. It demonstrates good estimation performances with an R<sup>2</sup> (correlation coefficient) around 0.95 and MAPE (mean absolute percentage error) less than 2.22% based on features extracted from PPG at the brachial artery level, an R<sup>2</sup> around 0.98 and MAPE less than 1.71% based on features extracted from PPG at the radial artery level and R<sup>2</sup> around 0.97 and MAPE less than 1.88% based on features extracted from PPG at the digital artery level.
• #### A Comparative Study of Different Human Skin Impedance Models

(IEEE, 2021-07-27) [Conference Paper]
This paper presents a comparative study of different human skin impedance models using meta-heuristic optimization Flower Pollination Algorithm (FPA). Four different models: Montague, Tregear, Lykken, and modified Montague skin models were used. There are two categories of electrical modelling; the RC-based model and the CPE-based model. The RC-based model focuses on skin stratification, while the CPE-based model focuses on the skin's biological properties. Six samples were taken as an experiment to acquire the skin impedance data to find the most accurate fitting model.