Multi-scale Datasets Research Articles

The triassic formations of Kuwait: From geological concepts to numerical forward stratigraphic modelling applications

The Triassic exploration campaign onshore Kuwait initiated during the 1980s was mainly based on the evaluation of regional 2D seismic data. To date, few scattered wells have penetrated this succession reaching the Minjur, Jilh and Sudair mixed clastic and carbonate formations. The latter reservoirs are expected to hold potential gas and condensates that are yet to be proven. This communication aims at deciphering the spatial and temporal evolution of sedimentary architectures and associated Triassic depositional environments and facies along Kuwait. An integrated G&G assessment utilizing multi-disciplinary and multi-scale dataset allowed to identify the major petroleum system elements and develop sequential conceptual depositional models. The proposed models are then tested using numerical process-based forward stratigraphic applications. The seismic interpretation points to dominantly south to north oriented progradations centered on the Dibdibah low with thicknesses reaching around 1300 ft in the Sudair. In addition, progradations oriented from East to West highlight the location of the basin margins around the Kuwait Arch and its northern extension. These prograding tendencies are also identified in the overlying Jilh Formation. Onlapping reflectors point to transgressional systems related to flooding events that are expected to deposit marine fine-grained carbonatesand thus act as potential seals. Minjur Formation on the other hand, shows thick siliciclastic deposits along the southern sector of the Kuwait Arch. These fluvio-deltaic deposits are 150 feet thick and thin towards the west. Stratigraphic modelling allowed to test and constrain the paleobathymetric evolution of the system. The latter ranges between 0 and 20 m (inner ramp) and 20–40 m (Middle ramp) and reaches up to 80 m in the deeper sectors of the basin. The resultant facies at 3rd- 4th stratigraphic order highlights the evolution of lithologies and facies with regards to the various system tracts leading to a better understanding of the timing and spatial extent of mixed carbonate systems, evaporites and siliciclastics. Future works will focus on a higher resolution mapping of geobodies and prospects using integrated shared earth approaches including seismic multi-attribute assessment and 3D inversion and characterization driven by novel artificial intelligence workflows.

A New Dataset and Comparative Study for Aphid Cluster Detection and Segmentation in Sorghum Fields.

Aphid infestations are one of the primary causes of extensive damage to wheat and sorghum fields and are one of the most common vectors for plant viruses, resulting in significant agricultural yield losses. To address this problem, farmers often employ the inefficient use of harmful chemical pesticides that have negative health and environmental impacts. As a result, a large amount of pesticide is wasted on areas without significant pest infestation. This brings to attention the urgent need for an intelligent autonomous system that can locate and spray sufficiently large infestations selectively within the complex crop canopies. We have developed a large multi-scale dataset for aphid cluster detection and segmentation, collected from actual sorghum fields and meticulously annotated to include clusters of aphids. Our dataset comprises a total of 54,742 image patches, showcasing a variety of viewpoints, diverse lighting conditions, and multiple scales, highlighting its effectiveness for real-world applications. In this study, we trained and evaluated four real-time semantic segmentation models and three object detection models specifically for aphid cluster segmentation and detection. Considering the balance between accuracy and efficiency, Fast-SCNN delivered the most effective segmentation results, achieving 80.46% mean precision, 81.21% mean recall, and 91.66 frames per second (FPS). For object detection, RT-DETR exhibited the best overall performance with a 61.63% mean average precision (mAP), 92.6% mean recall, and 72.55 on an NVIDIA V100 GPU. Our experiments further indicate that aphid cluster segmentation is more suitable for assessing aphid infestations than using detection models.

Robust multi-scale time series prediction for building carbon emissions with explainable deep learning

Accurate prediction dramatically enhances the effectiveness of carbon emission control, making it a valuable tool for achieving decarbonization goals. In this work, we propose a robust explainable encoder-decoder deep learning method for predicting multi-scale short-term carbon emissions and feature importance analysis to predict buildings’ emission patterns accurately. In the encoder, the initial inputs are transformed into a sequence to capture the time series and spatial information, where the model predicts the building emissions with an attention mechanism in the decoder. The hybrid model is trained and validated with a multi-scale dataset collected from a university. Results reveal that (1) After considering the buildings’ interactions, the R2 values regarding individual building emissions predictions are 0.943, 0.938, 0.941, and 0.943, respectively, climbing to 0.952, 0.943, 0.943, and 0.950. (2) Benefiting from the proposed model’s structure, the constructed model can accurately predict the community’s emissions with the R2 equal to 0.952 and keep excellent robustness after adding the white Gaussian noise data with all R2 are above 0.919. (3) The historical hourly emissions, outdoor humidity and time of day are considered the most essential factors responsible for the building emissions. The novelty of this work lies in the multi-scale building emissions prediction that is conducted by the developed hybrid model simultaneously considering the impact of environmental factors and building interactions, allowing us to expand the model’s application scenarios and provide valuable references for decarbonization decision-making.

Semantic segmentation of urban land classes using a multi-scale dataset

ABSTRACT The use of remote sensing imagery for land cover and land use classification has made significant advancements in recent years. However, it becomes particularly challenging to enhance the semantic representation of high-resolution networks while dealing with uneven land categories and merging multi-scale data without compromising the accuracy of semantic segmentation. To tackle this challenge, this paper presents a novel method for classifying high-resolution remote sensing images based on a deep neural network that performs semantic segmentation of urban construction lands into five categories: vegetation, water, buildings, roads, and bare soil. The network incorporates a U-shaped high-resolution neural network and the advanced high-resolution network (HRNet) framework. The parallel storage of feature maps with different resolutions enables the exchange of information between them. The data pre-processing module addresses the issue of data imbalance in the semantic segmentation of urban construction lands, resulting in an increase in Intersection over Union (IoU) values for different land types by 3.75%-12.01%. Additionally, a target context representation module is introduced to enhance the feature representation of pixels by calculating the relationship between pixels and multiple target regions. Moreover, a polarization attention mechanism is proposed to extract the characteristics of geographical objects in all directions and achieve a stronger semantic representation. This method provides a novel approach to accurately and effectively extract information on construction lands and advance the development of monitoring algorithms for urban construction lands. To validate the proposed U-HRNet-OCR+PSA network, a comparative analysis was conducted with six classical networks, including DeepLabv3+, PSPNet, U-Net, U-Net++, HRNet, and HRNet-OCR, as well as the relatively new ViT-adapter-L, Oneformer and InternImage-H. The experiments demonstrate that the U-HRNet-OCR+PSA network achieves higher accuracy compared to the aforementioned networks. Specifically, the corresponding IoU values for the buildings, roads, vegetation, bare soil, and water in the multi-scale dataset are 89.79%, 90.05%, 94.89%, 85.91%, and 88.36%, respectively.

Scale-space effect and scale hybridization in image intelligent recognition of geological discontinuities on rock slopes

Geological discontinuity (GD) plays a pivotal role in determining the catastrophic mechanical failure of jointed rock masses. Accurate and efficient acquisition of GD networks is essential for characterizing and understanding the progressive damage mechanisms of slopes based on monitoring image data. Inspired by recent advances in computer vision, deep learning (DL) models have been widely utilized for image-based fracture identification. The multi-scale characteristics, image resolution and annotation quality of images will cause a scale-space effect (SSE) that makes features indistinguishable from noise, directly affecting the accuracy. However, this effect has not received adequate attention. Herein, we try to address this gap by collecting slope images at various proportional scales and constructing multi-scale datasets using image processing techniques. Next, we quantify the intensity of feature signals using metrics such as peak signal-to-noise ratio (PSNR) and structural similarity (SSIM). Combining these metrics with the scale-space theory, we investigate the influence of the SSE on the differentiation of multi-scale features and the accuracy of recognition. It is found that augmenting the image's detail capacity does not always yield benefits for vision-based recognition models. In light of these observations, we propose a scale hybridization approach based on the diffusion mechanism of scale-space representation. The results show that scale hybridization strengthens the tolerance of multi-scale feature recognition under complex environmental noise interference and significantly enhances the recognition accuracy of GD. It also facilitates the objective understanding, description and analysis of the rock behavior and stability of slopes from the perspective of image data.

Application of a Web-Based Landslide Mitigation Information System (Sensor-Ku) in Kalongan Village Semarang, Central Java

Background and Objectives: Kalongan is one of the villages in Ungaran District with a potential landslide risk of 2% of its territory. Landslides occur along roadways and in residential areas, endangering the environment and damaging the infrastructure. A web-based early warning system application is the Landslide Early Warning System (Sensor-Ku) which is applied at the research location. This study intends to detect and provide real-time information on the likelihood of landslides, specifically to the people of Kalongan Village.
 Methods: This research creates a web-based monitoring and early warning system, as well as its implementation on an experimental microservice architecture, to provide landslide risk mitigation information to Kalongan residents.
 Results: According to the research, the landslide early warning system (LEWS) provides a solution by integrating many modules, such as recording dynamic variability, handling multi-scale data sets, extracting important trigger information about the onset of potential landslides, and broadcasting multi-level data. warning. This research also combines a real-time landslide early warning system. 
 Conclusion: The Landslide Early Warning System (LEWS) Program is effective in reducing the risk of landslides. Some of the issues found when running the program include the costs of maintaining the landslide early warning equipment, which must be managed by the Kalongan village community.

Fracture Network Analysis of Karstified Subis Limestone Build-Up in Niah, Sarawak

Understanding complex carbonate fracture networks and karstification at various geological scales is challenging, especially with limited multi-scale datasets. This paper aims to reduce uncertainty in the fracture architecture of Central Luconia karstified reservoirs by narrowing observational gaps between seismic and well data by using the discrete fracture models of exposed limestone outcrops as analogues for the subsurface carbonate reservoir. An outcrop-based fracture network characterisation of a near-surface paleo-karst at Subis Limestone combined with lineament analysis was conducted to extract fracture parameters. The karst structure was first delineated using a digital elevation map and outcrop examination. Then, topology analysis was performed, following the creation of two-dimensional discrete fracture models. Two main fracture sets oriented northeast–southwest and northwest–southeast and 79 potential dolines were identified. Fracture intersections, northeast–southwest major orientations, and drainage systems highly influenced the karst features. The Subis Limestone fracture model revealed that the highest number of fractures and total length of fractures were concentrated in the northern part of the Subis Limestone build-up (X: 250–350, Y: 150–250) and became denser towards the northwest direction of the outcrop (X: 600–800). The fractures in the Subis paleo-karsts appear isolated, with I-nodes ranging from 0.74 to 0.94. Hence, it is crucial to incorporate matrix porosity into multiple scales of fracture network modelling to improve upscaling and the modelling of fracture–vug networks, as well as to minimise the underestimation of discrete fracture networks in fractured and karstified limestone.

Seismic interpretation and sequence stratigraphic analysis of the Bahariya Formation in the South Umbaraka oilfields (Western Desert, Egypt): insights into reservoir distribution, architecture, and evaluation

The integration of multiscale datasets, including seismic, well-logs and stratigraphy, is providing a precise picture about reservoir evaluation and spatial distribution. This study focuses on the seismic interpretation and sequence stratigraphic analysis of the Bahariya Formation in the South Umbaraka oilfields (Selkit, Khepri, and Sethos), located in the Western Desert of Egypt. Seismic interpretations reveal various structural closures, including anticlinal domes and faulted anticlines in different fields. The study area is divided into northern and southern blocks separated by a major fault, and structural cross sections provide valuable information on reservoir zones regarding lateral extension, thickness variation, and fault distribution. The Bahariya Formation is subdivided into two depositional sequences: Lower Bahariya sequence (BAH.SQ-1) and Upper Bahariya sequence (BAH.SQ-2) with distinct wireline-log characteristics, seismic reflections, and lithofacies variations. BAH.SQ-1 comprises mixed siliciclastic-dominated units with some carbonate units, while BAH.SQ-2 consists of mixed siliciclastic and carbonate units. Multiple sequence boundaries, depositional surfaces, and systems tracts are identified within BAH.SQ-1 and BAH.SQ-2, indicating changes in sedimentary environments and sea-level fluctuations. The petrophysical evaluation reveals promising reservoir zones in both the Lower and Upper Bahariya sequences. However, the Lower Bahariya sequence exhibits superior reservoir characteristics in terms of net-pay thickness, shale volume, effective porosity, and water saturation. Particularly, the sandstone-dominated LST-1 within the Lower Bahariya Formation is identified as a favorable reservoir zone. Conversely, the Upper Bahariya Formation has limited reservoir potential. This study underscores the significance of sequence stratigraphy and reservoir architecture, emphasizing the role of faulting, folding, and depositional environments in the distribution and quality of reservoir zones. The findings contribute valuable insights into the structural and stratigraphic characteristics of the Bahariya Formation, enhancing understanding of hydrocarbon plays and reservoir distribution in the studied area.

A fast [formula omitted] time hybrid clustering algorithm using the circumference proximity based merging technique for diversified datasets

Clustering has been widely employed for extracting intrinsic groups because of its low reliance on domain knowledge. Though several clustering techniques have been developed in the literature, the majority of them become inefficient due to their dependence on user-defined parameters or inability to cluster multi-scale datasets. To handle these issues, several hybrid clustering algorithms which combine the advantages of partitional, hierarchical, and graph-based clustering techniques have been developed. Hybrid clustering algorithms first partition the data into smaller clusters and then merge them into genuine clusters. However, most of them incur a quadratic time complexity which limits their application for large datasets. The proposed work presents a two-step fast hybrid clustering algorithm based on partitions and efficient merging to reduce the computational cost by maintaining the clustering quality. In the first step, the dispersion of the data is considered to produce balanced partitions. Then graph circumference proximity-based merging technique is proposed to merge the sub-clusters. The overall computational complexity of the algorithm is O(NlgN) where N is the number of data points, ignoring the dimension. To the best of our knowledge, this is the fastest known graph neighborhood-based hybrid clustering algorithm. Experimental results on various diversified datasets exhibit a significant improvement in the running time as well as cluster quality and robustness against noise.

Structural transformations in single-crystalline AgPd nanoalloys from multiscale deep potential molecular dynamics.

AgPd nanoalloys often undergo structural evolution during catalytic reactions; the mechanism underlying such restructuring remains largely unknown due to the use of oversimplified interatomic potentials in simulations. Herein, a deep-learning potential is developed for AgPd nanoalloys based on a multiscale dataset spanning from nanoclusters to bulk configurations, exhibits precise predictions of mechanical properties and formation energies with near-density functional theory accuracy, calculates the surface energies closer to experimental values compared to those obtained by Gupta potentials, and is applied to investigate the shape reconstruction of single-crystalline AgPd nanoalloys from cuboctahedron (Oh) to icosahedron (Ih) geometries. The Oh to Ih shape restructuring is thermodynamically favorable and occurs at 11 and 92ps for Pd55@Ag254 and Ag147@Pd162 nanoalloys, respectively. During the shape reconstruction of Pd@Ag nanoalloys, concurrent surface restructuring of the (100) facet and internal multi-twinned phase change are observed with collaborative displacive characters. The presence of vacancies can influence the final product and reconstructing rate of Pd@Ag core-shell nanoalloys. The Ag outward diffusion on Ag@Pd nanoalloys is more pronounced in Ih geometry compared to Oh geometry and can be further accelerated by the Oh to Ih deformation. The deformation of single-crystalline Pd@Ag nanoalloys is characterized by a displacive transformation involving the collaborative displacement of a large number of atoms, distinguishing it from the diffusion-coupled transformation of Ag@Pd nanoalloys.

A Novel Approach for Mining Spatiotemporal Explicit and Implicit Information in Multiscale Spatiotemporal Data

In the era of big data, a significant volume of spatiotemporal data exists in a multiscale format, describing diverse phenomena in the objective world across different spatial and temporal scales. While existing methods focus on analyzing the features and connections of spatiotemporal data at various scales, they often overlook the consideration of uncertainty in spatiotemporal information within the context of multiscale meaning. To effectively harness the potential of spatiotemporal data, it becomes crucial to capture the fuzzy spatiotemporal information inherent in multiscale datasets. This paper proposes a novel multiscale spatiotemporal correlation method that accounts for and quantifies the uncertainty of spatiotemporal information. Spatiotemporal information is categorized into two types, explicit information and implicit information, based on respective levels of uncertainty. The method employs spatiotemporal cubes to interpret the spatiotemporal items within the data, followed by the introduction of a benchmark scale to determine the certainty of each spatiotemporal item based on its range and topological relationships. Subsequently, spatiotemporal confidence and correlation index are proposed to gauge the significance of geographical elements and their interrelationships. To validate the proposed method, a multiscale spatiotemporal transaction dataset is generated and utilized in the experiment. The experimental results demonstrate that the proposed method effectively captures spatiotemporal implicit information and enables better utilization of multiscale spatiotemporal data. Notably, the importance of each object of study varies when analyzed using different benchmark scales, providing valuable insights for professionals to identify novel objects and associations worthy of consideration. The obtained results can be used to construct spatiotemporal knowledge graphs.

Implication of the micro- and lithofacies types on the quality of a gas-bearing deltaic reservoir in the Nile Delta, Egypt

The Messinian Qawasim sequence, as one of the most important reservoirs in the Egyptian Nile Delta, represents a typical gas-bearing deltaic reservoir sequence. It aims at delineating the implication of the litho- and microfacies associations on the reservoir characteristics of the different stages of the fluvial deltas including the prodelta, proximal/distal delta front, and the delta plain depositional sequences. The studied reservoir sequence was divided into two units; upper and lower clastic units. The petrophysical properties of these two units were studied on the borehole scale using the gamma-ray, caliper, sonic, density, neutron, and resistivity logs to estimate the reservoir parameters including the total and effective porosities, water saturation, shale volume, and the net-pay thickness. For more details, they are also examined on the plug-scale using the core dataset including helium porosity, density, permeability, and fluids saturations, where the flow zone indicator, the reservoir potentiality index, the reservoir quality index, and the average reservoir pore radius were then estimated. The entire sequence is divided into five reservoir rock types (RRT1-RRT5) where, the best quality is assigned to the first RRT (the upper reservoir unit), and the lowest quality is assigned to the RRT5 of the lower reservoir unit. Based on core description and the petrographical studies five lithofacies and four microfacies have been identified. The lithofacies are (1) laminated mudstones/siltstones, (2) ripple laminated sandstones, (3) cross-laminated sandstones, (4) cross-bedded sandstones, and (5) pebbly massive sandstones. These lithofacies are primarily composed of four microfacies; sublithic arenites, subarkose arenite, glauconitic quartz wacke, and sandy mudstone/siltstone. Based on this study, the pebbly massive and the cross-bedded sandstones of the upper unit which is composed of the sublithic and subarkose arenites (RRT1-RRT2) have the best reservoir quality. On the other hand, the lowest quality is assigned to the RRT5 (sandy mudstone/siltstone microfacies) which is represented on the macro scale by the laminated mudstones/siltstones lithofacies. The integration between multi-scale datasets (core-well, petrography, well logs, and seismic) gives a precise picture of the deltaic Qawasim reservoir rock units in the Nile Delta. This workflow has never been applied to the deltaic system in the onshore Nile delta and North Africa. Thereby, this study is considered a standard case study for the deltaic sequences and its proposed workflow is applicable to the Nile Delta elsewhere and worldwide for similar reservoir sequences.

Design across multi-scale datasets by developing a novel approach to 3DGANs

The development of Generative Adversarial Networks (GANs) has accelerated the research of Artificial Intelligence (AI) in architecture as a generative tool. However, since their initial invention, many versions have been developed that only focus on 2D image datasets for training and images as output. The current state of 3DGAN research has yielded promising results. However, these contributions focus primarily on building mass, extrusion of 2D plans, or the overall shape of objects. In comparison, our newly developed 3DGAN approach, using fully spatial building datasets, demonstrates that unprecedented interconnections across different scales are possible resulting in unconventional spatial configurations. Unlike a traditional design process, based on analyzing only a few precedents (typology) according to the task, by collaborating with the machine we can draw on a significantly wider variety of buildings across multiple typologies. In addition, the dataset was extended beyond the scale of complete buildings and involved building components that define space. Thus, our results achieve a high spatial diversity. A detailed analysis of the results also revealed new hybrid architectural elements illustrating that the machine continued the interconnections of scale since elements were not explicitly part of the dataset, becoming a true design collaborator.

A lightweight neural network compatible with high-correlation broadband optical filters in computational spectroscopy

Computational spectrometer is powerful for portable in situ applications. To achieve better signal to noise ratio (SNR), broadband optical filters are more commonly introduced. As for conventional sparse signals reconstruction algorithms like gradient projection, its performance relies on the low correlation property of optical filters, which poses severe challenge for filters design and fabrication. In this study, we propose a lightweight neural network(NN) named HBOF-Net that can be compatible with high-correlation broadband optical filters to renovate the situation. The spectra reconstruction performs very well under a group of high correlated optical filters and multiscale training datasets. Firstly, we propose a group of various structure design of photonic crystal(PC) slabs and select every 16 to construct a spectrometer megapixel randomly. Then we extract different percentage of spectra from CAVE and ICVL dataset to build up multiscale training dataset groups. Afterwards, the training of the network has been carried out and the testing curve indicate there is no over fitting. A total of 16 trainings have been carried out and the results have been compared and discussed. Lastly, the experiment demonstrates that the mean square error(MSE) of the reconstructed spectra by HBOF-Net can reach 1E-6 by maximum correlation coefficient of 0.93 under multi-scale datasets which is 90%, 10%, 5%, and 3% of the entire dataset. This study successfully prove that NN can be compatible with high correlated optical filters, which is a significant leap in the development of computational spectrometers. It provides a good reference for computational spectrometers based on broadband and high-correlation optical filters.

Plumbing systems and associated seafloor fluid seepages in deep-water Nigeria: Factors controlling their architecture and cyclic evolution

A range of plumbing systems has been identified in deep-water Nigeria by interpreting multi-scale seismic data sets (reprocessed 3D exploration seismic reflection data and 2D very high-resolution near-bottom seismic reflection data) as well as sedimentological and geochemical data obtained from a long sediment core recovered by the seafloor drill rig MARUM-MeBo70. The plumbing systems are characterized by diverse fault networks and lie above tectonic features including thrust/fold structures and strike-slip faults linked to gravity-driven deformation caused by underlying over-pressured shales. The plumbing systems are associated with fluid seeping structures at the seafloor such as pockmarks, mud volcanoes and seafloor zonations colonized by living benthic macrofauna typical of active fluid flow. The comparison of seismic stratigraphic sequences with climate and environmental proxies determined for sediments including lithology, element ratios such as Ca/Fe and Zr/Rb, sedimentation rates, and planktonic foraminifera δ18O records shows a control of short-term (0.1–0.4 Myr) and long-term (around 1 Myr) sedimentary cycles. During short-term fluctuations, the recurrence of fluid seeping structures is controlled by the lithology (coarse-grained versus fine-grained sediments) linked to glacial-interglacial fluctuations and a monsoon regime. Conditions favorable for fluid flow are related to the deposition of coarse-grained layers, glacial periods, sea-level lowstands, and low sedimentation rates. We hypothesize that cyclic gravity-driven sediment deformations controlled by glacial-interglacial and long-term variations in accommodation and sedimentation, have led to the cyclic evolution of the plumbing systems and associated seeping structures since the Plio-Pleistocene.

Multi-scale geophysical characterization of microporous carbonate reservoirs utilizing machine learning techniques: An analog case study from an upper Jubaila formation outcrop, Saudi Arabia

Microporosity hosts a significant portion of the total hydrocarbon volume in the Middle Eastern carbonate reservoirs. An improved understanding of microcrystals' morphology that hosts micropores, their impact on reservoir properties, and their spatial distributions will contribute to a more accurate reservoir quality prediction. This study proposes a methodology that utilizes machine learning approaches for microporosity characterization and integrates a multi-scale dataset ranging from micrometer-scale SEM images to meter-scale seismic attributes to build reservoir porosity models. We tested the methodology on an outcrop in Riyadh, Saudi Arabia, which exposes the upper part of the Jubaila Formation equivalent to the lower part of the Arab-D reservoir. We acquired a 35 m-long core and a 600 m-long 2D seismic line behind the outcrop. Laboratory-scale petrophysical measurements, including porosity, permeability, acoustic velocity, bulk and grain density, and x-ray diffraction, were performed over 106 horizontal core plugs drilled from the core. We quantitatively characterized the morphology and microtextures of micrite crystals using SEM images by performing Random Forest classifications trained on SEM image features. We performed unsupervised classification using Self-Organizing Map (SOM) to all lab-measured properties for data clustering. We investigated potential correlations between SOM clusters with micrite morphology, which resulted in a predictable relationship following the granularity and sphericity of microcrystals with porosity, permeability, and acoustic velocity. It was possible to represent multiple lithofacies with a single log-linear porosity-permeability relationship and a single value of equivalent differential effective medium (DEM) aspect ratio in velocity-porosity space. The inter-relationship between micrite morphology and microporosity in the well was then propagated to reservoir-grid scale using inverse differential effective medium (DEM) of acoustic impedance from inverted seismic data. The methodology developed in this study thus provides a practical way to integrate key sub-grid scale micro-and macro-heterogeneities into reservoir scale property models.

Study on Intelligent Classification of Aging Heat-Resistant Materials.

High chromium martensitic heat-resistant steel is considered as a candidate material for pressure components of the next generation of incinerators of the subcritical level or above in China due to its excellent high-temperature and corrosion resistance, but in the long-term service, aging will significantly affect the service safety of materials. So, accurate identification of its aging state is important to enhance the safety of a power plant. In this paper, an automatic aging grading model of high chromium martensite heat-resistant steel based on the depth residual network is proposed according to different scales of metallographic data. A multiscale data set is constructed by image reduction to verify the accuracy of the model in identifying microstructure images of high chromium martensitic heat-resistant steel with different scales. The experimental results show that the model using multiscale data sets performs well, and then, through feature pyramid network model training, the accuracy rate is further improved, and a relatively good prediction accuracy model is obtained. The validity of the deep learning method for the classification of damage and aging of P91 steel with different scales is verified.

Geophysical assessment for the oil potentiality of the Abu Roash “G” reservoir in West Beni Suef Basin, Western Desert, Egypt

The current study evaluates the hydrocarbon potentiality of the sandstone intervals of the Abu Roash “G” Member in Beni Suef Basin, Egypt. This evaluation has been performed through seismic interpretation for thirty seismic profiles and petrophysical appraisal for Fayoum-2X well (Fayoum oil field) and Yusif-4X well (Yusif oil field). The seismic interpretation presented the prevalence of the Lower Cretaceous NW-SE normal faults followed and linked with a major Late Cretaceous ENE strike-slip fault affected the Beni Suef Basin. This strike-slip fault is accompanied by a significant flower structure and releasing bend in the middle part of the studied area. The entrapment styles in both Fayoum and Yusif oil fields are considered by the existence of NW-SE trending horst closures bounded by the NW-SE faults. The petrophysical analysis exhibits good reservoir quality for the sandstones of Abu Roash “G” Member in the examined fields with low shale volume (5–20%); high effective porosity (15–25%), relatively low water saturation (35–55%) and pay thickness varies between 14 ft and 16 ft. Although the Abu Roash “G” Member exhibits lateral facies changes across the Beni Suef Basin, this study highlights the hydrocarbon potentiality of its entire sandstones for further exploration and development purposes in both Fayoum and Yusif oil fields. Core-based petrophysical analysis and detailed sedimentological analyses for the reservoir zones within Abu Roash “G” Member at Beni Suef Basin are highly recommended to emphasize the effect of the lateral facies change on the reservoir properties along the whole basin which will affect the development plans. The integration of multiscale datasets which includes seismic, different types of well logs, core is giving a precise picture about the reservoir units. Altogether are figuring out the detailed composition of reservoir sedimentology and quality by using the core information, investigate the hydrocarbon potentiality using the well logging evaluation, deducing their trapping mechanism, and controlling structures using the seismic datasets and the spatial distribution is define based on the correlation between different wells, depositional and structural conditions.

Using one-dimensional convolutional neural networks and data augmentation to predict thermal production in geothermal fields

Numerical simulation is the most common method to predict reservoir production temperatures during geothermal energy extraction. Considering the principle of numerical modeling, the numerical simulation establishment process requires a large amount of good exploration data. In addition, it is heavily influenced by subsurface heterogeneity. Also, despite the superior performance of deep learning models, sparse data is a critical challenge in the training process. Therefore, we propose a one-dimensional-convolutional neural network (1D-CNN) model and use data augmentation techniques to build a large-scale multiscale production temperature data set. The network learns the nonlinear relationship between boundary conditions and production temperature from the data set and reaches the production temperature prediction for a three-well geothermal system. The maximum difference in production temperature is 1.8181 °C and the generalization performance is improved by 59.6%. It is worth noting that the excellent generalization capability indicates that the data-driven concept behind the model is an easily interpretable one. As a new data processing concept, the “data-guided approach” is a key step in establishing a universal approach for application in the geothermal field.

Hierarchical Information Enhancing Detector for Remotely Sensed Object Detection

For the remote sensing object detection task, two-stage networks are widely used due to their high accuracy. These networks roughly predict the proposal regions containing potential objects. It is assumed in these methods that the sizes of these regions are close to that of the corresponding real object. However, this assumption is not always true. Consequently, the detector is affected by the size-unfitting proposal regions. In this letter, a hierarchical information enhancing detector (HIE-Det) is advocated to deal with this issue. First, the important semantic reinjection (ISR) module is proposed to mitigate the lack of object semantics caused by the size-unfitting problem. Compared with the normal detectors, the ISR module increases the proportion of information on objects and improves the effectiveness of the detection model. Second, the object boundary enhancing (OBE) module is proposed to improve the robustness of the regression. The OBE module introduces the convolutional branch stacking multigranularity grids for the same proposal region. Multiple granularity levels improve the robustness of the model to the different degrees of proposal size unfitting. Finally, to evaluate the effectiveness of the HIE-Det on multiscale datasets in a balanced and effective manner, we propose the scale-modulating scores (S-scores), i.e., scale-modulating average precision (sAP) and scale-modulating average recall (sAR). Compared with the other comprehensive scores, the S-scores are rid of the sample amounts and give priority to weaker indices. Implementing the proposed HIE-Det, S-scores {sAP, sAR} are, respectively, improved from {17.3%, 29.7%} to {34.8%, 43.2%}, reaching the state-of-the-art performance on the HRRSD dataset. These experiments verify the effectiveness of the proposed HIE-Det.