Mahshad Samnejad, PhD

Asset owners are required by the government to carry out regular inspection surveys to ensure the integrity of all pressure-containing equipment. Conventionally, such operations are performed manually by teams of trained inspectors through visual examinations on-site or remotely. We present an integrated framework for automating the entire process of pipeline inspection without any need for human intervention, pipeline function interruption, or equipment destruction by unleashing the power of… Show more

Asset owners are required by the government to carry out regular inspection surveys to ensure the integrity of all pressure-containing equipment. Conventionally, such operations are performed manually by teams of trained inspectors through visual examinations on-site or remotely. We present an integrated framework for automating the entire process of pipeline inspection without any need for human intervention, pipeline function interruption, or equipment destruction by unleashing the power of the state-of-the-art digital technologies, including deep learning, computer vision, and cloud storage and computing.
Decades of survey videos captured by remotely operated vehicles (ROVs) and drones are broken into frames with optical character recognition (OCR)-extracted time/location stamps, which are annotated by our inspectors to precisely delineate the boundaries of the damaged areas and their specific categories. After extracting images from videos, augmenting the image data, the training set is fed into a deep convolutional neural network architecture equipped with instance segmentation layers for object detection. Trained models are then deployed on the cloud acting as intelligent inspectors for future surveys, allowing also to balance the trade-off between the inference accuracy and performance speed, being crucial to real time usage of the software.
After a pipe segment subject to damage is identified to be imposing an integrity risk, it will automatically be raised as a flag into our linked maintenance infrastructure along with the corresponding spatiotemporal information of the event to take the necessary maintenance, repair, or replacement actions.
The proposed endeavor outperforms common industry practices, as it not only reduces the asset operational costs by eliminating the need for the labor-intensive manual diagnostic inspections, but also improves the hazard mitigation plan by providing accurate risk assessments in shorter time spans. Show less

In drilling operations, the rheological properties of the drilling fluid (‘mud’) are measured at rig site at ambient temperature using sensors in real-time. These measurements, however, are required to be reported at the API standard temperature (120/150 oF). As the rheological properties of the drilling fluids vary significantly with temperature, it is essential to calibrate sensor measurements at the ambient temperature to the API standard. Previous attempts in the literature to build… Show more

In drilling operations, the rheological properties of the drilling fluid (‘mud’) are measured at rig site at ambient temperature using sensors in real-time. These measurements, however, are required to be reported at the API standard temperature (120/150 oF). As the rheological properties of the drilling fluids vary significantly with temperature, it is essential to calibrate sensor measurements at the ambient temperature to the API standard. Previous attempts in the literature to build data-driven frameworks for predicting drilling fluids behavior demonstrate limited success due to restrained data access, neglect of the physics, and/or use of improper algorithms, such as neural networks which are shown to perform poorly despite their popularity.In this work, we develop a digital twin to calibrate the rig site rheology measurements for obtaining theAPI standard properties by exploring the use of a set of promising machine learning algorithms. A dataset composed of various drilling fluids composition with rheological measurements at both rig site and API conditions is collected. Our results demonstrate that the ensemble algorithm outperform other commonly used methods, such as regularized regression, polynomial regression, and neural networks. The optimized integrative model is deployed on a platform at rig for use at real-time drilling operations. Show less

The rise in earthquake activity attributed to anthropogenic causes stresses the need for hazard mitigation decision-making frameworks that are based upon realistic models giving reliable predictions.

A set of recent efforts have addressed the problem of modeling injection-induced fault slip using poroelastic coupled schemes, often failing to accurately predict fluid movement and pore pressure perturbations in the evolving fault damage zone, proven to significantly affect and be affected… Show more

The rise in earthquake activity attributed to anthropogenic causes stresses the need for hazard mitigation decision-making frameworks that are based upon realistic models giving reliable predictions.

A set of recent efforts have addressed the problem of modeling injection-induced fault slip using poroelastic coupled schemes, often failing to accurately predict fluid movement and pore pressure perturbations in the evolving fault damage zone, proven to significantly affect and be affected by fault stresses and rupture.
We go beyond conventional poroelastic simulations by incorporating the nonlinear behavior of the subsurface material, which controls dominant inelastic deformation. We differentiate between fault rupture planes, damage zone, and host rock by defining a set of heterogeneous characteristic hydromechanical properties within each damage zone element, and predicting the evolution of the magnitude and extent of such anomalies in the vicinity of fault friction planes. We analyze pore pressure and stress state changes spatiotemporally by applying a continuum damage mechanics workflow to our computational simulation framework, which allows to integrate multi-scale physical processes of flow, deformation, and crack growth. We conclude with safe injection design implications in terms of well placement and flow rate under various in-situ conditions. Show less

Observational studies have long suggested that stresses generated from earthquakes can alter permeability, as frequently evidenced by higher levels of fluid outflow, as measured at the surface. Similarly, by means of tidal deconvolution and opportunely placed boreholes, more precise measurements of seismically induced permeability changes may be obtained. As evidenced from borehole observations, this process of permeability change appears to be a transient process, with measured permeability… Show more

Observational studies have long suggested that stresses generated from earthquakes can alter permeability, as frequently evidenced by higher levels of fluid outflow, as measured at the surface. Similarly, by means of tidal deconvolution and opportunely placed boreholes, more precise measurements of seismically induced permeability changes may be obtained. As evidenced from borehole observations, this process of permeability change appears to be a transient process, with measured permeability values regressing to a mean value after some time. What is less clear, however, is how such a temporary change in permeability might affect environments already containing critically stressed faults.

We seek to investigate the phenomenon of seismically induced permeability change initially from an observational perspective, taking account of cases where permeability appeared to have been altered by coseismic effects, and considering the potential influence from dynamic and static stresses respectively. Similarly, we seek to draw a connection from a conservation of energy perspective between seismic input and observed permeability change, along with a decay function fit to the observed value regression.
Lastly, we seek to fit the relations gained from our observational studies to guide a permeability update function within an existing, fully coupled, and fully implicit poroelastic model defined for heterogeneous reservoirs in the presence of production wells, injection wells, and critically stressed faults, modeled using the Mohr-Coulomb failure criterion. Through the use of this numerical model, we seek to investigate not only the potential for induced seismicity resulting solely from anthropocentric causes, but also the likelihood for prior seismicity serving as a catalyst for further events that might not be otherwise triggered from injection or production alone. Show less

The success of hydraulic fracturing jobs is often related to rock brittleness indices, which are taken as the sole impact factor determining fracturing results. Indeed, hydraulic fractures play a principal role in producing from low-permeability reservoirs; however, brittleness is not the only parameter contributing to productivity of unconventional resources. Under a variety of circumstances, brittleness indices are insufficient to explain rock fracability and permeability enhancement during… Show more

The success of hydraulic fracturing jobs is often related to rock brittleness indices, which are taken as the sole impact factor determining fracturing results. Indeed, hydraulic fractures play a principal role in producing from low-permeability reservoirs; however, brittleness is not the only parameter contributing to productivity of unconventional resources. Under a variety of circumstances, brittleness indices are insufficient to explain rock fracability and permeability enhancement during hydraulic stimulation. For better prediction and design, it is imperative to identify and understand other factors affecting fracture creation and propagation, and to build models that include the effect of these factors on flow enhancement. To numerically model permeability enhancement after injection, we can regard fractured rock as a damaged continuum, which allows simulation of the deformation and fracturing response of the reservoir using material constitutive laws for brittle and ductile regions. We outline a coupled flow- geomechanical simulation framework that fits into available reservoir simulation platforms and does not require pre-specified fracture paths. We develop the fracture growth mechanisms for the coupled simulation framework by analyzing the effect of rock properties and in-situ stresses on the fracture length at different injection pressures. Based on these mechanisms, we propose factors that quantify the success of hydraulic fracturing jobs beyond the simplified rock brittleness indices. Show less

In hydraulic fracturing operations, permeability is enhanced when fractures are created and/or stimulated by injecting a highly pressurized fluid. In addition, mechanical response of the rock changes because of permanent modifications in the structure and properties of the rock after failure. In order for engineers to accurately predict results of hydraulic stimulation projects, mathematically rigorous and numerically efficient models of fluid flow and geomechanical deformation in fractured… Show more

In hydraulic fracturing operations, permeability is enhanced when fractures are created and/or stimulated by injecting a highly pressurized fluid. In addition, mechanical response of the rock changes because of permanent modifications in the structure and properties of the rock after failure. In order for engineers to accurately predict results of hydraulic stimulation projects, mathematically rigorous and numerically efficient models of fluid flow and geomechanical deformation in fractured porous media must be used in computer simulations. Some of the earlier approaches address the problem of fluid flow through fractured media with mathematical models that are either too simplistic or too expensive computationally and are not compatible with the available petroleum reservoir simulation platforms. In this work, a reservoir simulation framework is developed using a sequentially coupled numerical scheme of flow, deformation and poromechanical damage to study variations occurring in the fractured rock properties and state variablesas a result of hydraulic stimulation. We numerically simulate injection-induced permeability enhancement and plastic deformation as well as post-stimulation softening behavior of the rock by considering the stimulated rock as a mechanically damaged configuration, the properties of which are modeled using an effective continuum model. We study how the flow and mechanical properties of fractured rock, namely permeability and stiffness, change by virtue of hydraulic fracturing, and we investigate the dynamics of pressure distribution and stress state with time. The sequential nature of the proposed coupling framework lends itself to easy integration with reservoir simulation and prediction tools. Show less

Knowledge of the magnitude of in situ stresses is crucial to a broad range of applications, particularly drilling, well completion, and hydraulic fracturing design. Stress field characterization is attained by determination of magnitude and direction of three principal stresses: the vertical or overburden stress(Sv), the minimum horizontal principal stress (Shmin), and maximum horizontal principal stress (SHmax).While numerous direct borehole measurements, such as density logs and leak-off… Show more

Knowledge of the magnitude of in situ stresses is crucial to a broad range of applications, particularly drilling, well completion, and hydraulic fracturing design. Stress field characterization is attained by determination of magnitude and direction of three principal stresses: the vertical or overburden stress(Sv), the minimum horizontal principal stress (Shmin), and maximum horizontal principal stress (SHmax).While numerous direct borehole measurements, such as density logs and leak-off tests, provide reliable assessments of Sv and Shmin, estimating local SHmax remains a challenge.This paper presents a workflow that provides with an approach for estimation and constraining of the maximum horizontal principal stress in a carbonate reservoir from borehole sonic logs, by comparison to the isotropic stress state. Our study was restricted to cases with limited achievable data, namely, conventional logs, which is expected to be the circumstances for most practical applications.The results were compared to published regional stress reports, with goal of validating and understanding the variations of stresses at field scale. Those findings were in agreement with the estimations obtained from other acoustoelastic methods and the region fault regime in terms of the relative obtained magnitudes of the stresses. In addition to an increasing trend in the calculated stress magnitudes with depth, we observe local variations of stresses in this oilfield, that we suspected to be caused partially by the non-uniform distribution of production and injection activities.The novelty of proposed workflow is ability to give borehole estimates of the maximum horizontal principal stress magnitude for a carbonate field located in a regionally compressive zone, without the need for availability of costly measurements, and by taking advantage of techniques to fill in gaps of limited data, thus, allowing for integrative use of multiple data types that are commonly available. Show less

Understanding the spatial distribution of fracture is important for identification of reservoir zones with higher production potential. Furthermore, knowledge of distribution of natural fracture network could be helpful in the design of appropriate stimulation jobs such as multi-stage hydraulic fracturing. Azimuthal Amplitude Variation Versus Offset of seismic data (AzAVO) is a common measure of anisotropy which is often analyzed with the purpose of fracture detection and characterization… Show more

Understanding the spatial distribution of fracture is important for identification of reservoir zones with higher production potential. Furthermore, knowledge of distribution of natural fracture network could be helpful in the design of appropriate stimulation jobs such as multi-stage hydraulic fracturing. Azimuthal Amplitude Variation Versus Offset of seismic data (AzAVO) is a common measure of anisotropy which is often analyzed with the purpose of fracture detection and characterization. Determining fracture properties such as spatial distribution, orientation and whether they are open or close, from AzAVO data is an inverse problem. This could be addresses using model-based inversion techniques. Common methods for fracture density inversion from pre-stack gathers are illustrated by constructing a set of optimal basis functions through singular value decomposition. The general workflow of inverting for multiple seismic attributes is presented with a simple synthetic fracture density mapping case of multi-azimuth common-middle-point gathers. Further complementation of AVO anisotropy attributes with geological information and other anisotropic attributes is then considered with the purpose of better interpreting the attributes and reducing the level of uncertainty in such problems. Integrative approaches are investigated for combining AzAVO anisotropy with attributes such as travel-time anisotropy and shear-wave anisotropy and frequency absorption attributes. Show less