Search

The forecasting of steel futures prices is important for the steel futures market, even for the steel industry. We propose a decomposition ensemble model that incorporates the Ensemble Empirical Mode Decomposition (EEMD), Long Short‐Term Memory (LSTM), Support Vector Regression (SVR), and Back Propagation (BP) neural network to forecast steel futures prices. The forecasting procedures are as follows: (1) The price data are initially decomposed into several relatively independent Intrinsic Mode Functions (IMFs) and a residue using EEMD. (2) The IMFs are then reconstructed as components representing short‐term, medium‐term, and long‐term frequencies via fine‐to‐coarse. (3) LSTM, SVR, and BP neural network are utilized to forecast the short‐term, medium‐term, and long‐term reconstructed components, respectively. (4) The prediction results for each component are simply added to the final prediction results. The accuracy of the proposed model is compared with several benchmark models by experiments and evaluated by some prediction evaluation indexes. The experimental results show that our model outperforms other models in terms of forecast accuracy, confirming its strong predictive capabilities. This study provides some suggestions for investment and decision making by participants in the steel futures market. It may promote the smooth operation of the steel futures market and shed some light on the operation of the steel industry. [ABSTRACT FROM AUTHOR]

Artificial Intelligence (AI) Machine Learning (ML) technologies, particularly Deep Learning (DL), have demonstrated significant potential in the interpretation of Remote Sensing (RS) imagery, covering tasks such as scene classification, object detection, land-cover/land-use classification, change detection, and multi-view stereo reconstruction. Large-scale training samples are essential for ML/DL models to achieve optimal performance. However, the current organization of training samples is ad-hoc and vendor-specific, lacking an integrated approach that can effectively manage training samples from different vendors to meet the demands of various RS AI tasks. This article proposes a solution to address these challenges by designing and implementing LuoJiaSET, a large-scale training sample database system for intelligent interpretation of RS imagery. LuoJiaSET accommodates over five million training samples, providing support for cross-dataset queries and serving as a comprehensive training data store for RS AI model training and calibration. It overcomes challenges related to label semantic categories, structural heterogeneity in label representation, and interoperable data access. [ABSTRACT FROM AUTHOR]

The article focuses on improving emotion generation in dialogue systems by considering personality traits, aiming to enhance user engagement and service quality. It mentions through the proposed task of Personality-affected Emotion Generation and the development of the Personality EmotionLines Dataset (PELD), the study addresses challenges in integrating personality and emotional factors while modeling personality as a transition weight, resulting in improved emotion generation performance.

Diagnosing individuals with autism spectrum disorder (ASD) accurately faces great challenges in clinical practice, primarily due to the data's high heterogeneity and limited sample size. To tackle this issue, the authors constructed a deep graph convolutional network (GCN) based on variable multi‐graph and multimodal data (VMM‐DGCN) for ASD diagnosis. Firstly, the functional connectivity matrix was constructed to extract primary features. Then, the authors constructed a variable multi‐graph construction strategy to capture the multi‐scale feature representations of each subject by utilising convolutional filters with varying kernel sizes. Furthermore, the authors brought the non‐imaging information into the feature representation at each scale and constructed multiple population graphs based on multimodal data by fully considering the correlation between subjects. After extracting the deeper features of population graphs using the deep GCN(DeepGCN), the authors fused the node features of multiple subgraphs to perform node classification tasks for typical control and ASD patients. The proposed algorithm was evaluated on the Autism Brain Imaging Data Exchange I (ABIDE I) dataset, achieving an accuracy of 91.62% and an area under the curve value of 95.74%. These results demonstrated its outstanding performance compared to other ASD diagnostic algorithms. [ABSTRACT FROM AUTHOR]

This article addresses the problem statement and objective by exploring the necessity, scope, and execution of digital transformation in the oil and gas industry from a psychological perspective. It highlights the cognitive barriers faced by non-ICT professionals, which are often overlooked in traditional approaches. The study integrates case studies and empirical evidence from a mixed-methods approach, including qualitative interviews with industry experts and quantitative surveys among employees, to provide a comprehensive understanding of the transformation process. The research emphasizes the integration of psychological theories with practical digital transformation strategies, illustrating key obstacles and solutions. By adopting a holistic approach that incorporates both technological advancements and psychological insights, the study aims to enhance the effectiveness and sustainability of digital transformation efforts. Major contributions include identifying cognitive barriers, proposing solutions to overcome these barriers, and presenting real-world applications that demonstrate the successful merging of psychological insights and digital strategies in the oil and gas sector. [ABSTRACT FROM AUTHOR]

A new temperature fiber ring laser (FRL) sensor based on a cascaded Sagnac loops fiber structure is proposed and experimentally demonstrated. The optical FRL sensor consists of cascaded Sagnac loops by inserting two polarization-maintaining optical fibers (PMF) with slightly different lengths. PMF with length of 56 cm and 75 cm are used in Sagnac loops as a filter and sensing unit in laser cavity. The working principle of the sensor is based on the phase shift (θ) caused by birefringence between two principal polarization modes and enhance the sensitivity by constructing a Vernier-scale. In an appropriate temperature range (25 °C - 31 °C), the detection sensitivity of FRL sensor based on a cascaded Sagnac structure is significantly higher than other FRL sensors. Thanks to the laser sensing system, the sensitivity can be modulated by changing the free spectral range (FSR). The experimental results show that the temperature sensitivity of the cascade Sagnac structure sensor is - 4.031 nm / °C, which is five times higher than that of FRL sensor base on the single Sagnac structure.

The discovery of eclogite outcrops in the East Kunlun Orogen Belt (EKOB) has confirmed the existence of an Early Paleozoic HP-UHP metamorphic belt. However, the protoliths and metamorphic histories of widespread metabasites remain poorly constrained. We collected three types of metabasites from the central part of EKOB. We present an integrated study of petrography, whole-rock geochemistry, Sr-Nd isotopes, estimated P–T conditions, and zircon U-Pb isotope ages. The results show that amphibolites and retrograde eclogites have clockwise P–T paths with peak conditions of, respectively, 11–12 kbar and 675–695 °C, and 21.5–22.2 kbar and 715–750 °C. Zircon dating of metabasites from Dagele yields Late Ordovician (~449 Ma) to Early Silurian (~440 Ma) protolith ages and Early Devonian (~414 Ma) amphibolite facies metamorphic ages. Retrograde eclogites from east Nuomuhong have a protolith age of ~902 Ma and metamorphic ages of ~418 Ma, consistent with other eclogites from East Kunlun. Our data suggest that the protoliths of Dagele metabasites represent arc-type magmatism during the subduction of a small back-arc oceanic basin. Instead, the protoliths of retrograde eclogites are Neoproterozoic tholeiitic basalts emplaced into continental crust and subsequently deeply subducted. We develop a new model for Early Paleozoic subduction and collision in the East Kunlun region, emphasizing the role of 'dominant' and 'secondary' suture boundaries. This model helps explain the ages and metamorphic histories of the metabasites studied here and offers new perspectives on the evolution of the Proto-Tethys Ocean. [ABSTRACT FROM AUTHOR]

Reservoirs are essential hydraulic facilities for water resource allocation in Xinjiang. Since the 1950s, many reservoirs have been constructed for oasis water resource utilization in Xinjiang, enhancing the storage capacity of water resources. There are a few intact and open reservoir archives containing both geolocations and hydraulic attributes, such as the reservoir completion year, which can facilitate our understanding of the correlation between hydraulic engineering and oasis expansion. This paper mapped all the reservoirs of Xinjiang using Sentinel-2 MSI images from 2022. It associated their attributes with the reservoir's extent, such as the capacity, area, complete year, altitude, etc., by consulting historical almanac data to establish a full elemental dataset with both geographic and attribute information. Furthermore, the spatial variability and historical process of the reservoirs were discussed against geomorphic information and oasis evolution. The results showed that 804 reservoirs were mapped cumulatively in Xinjiang up to 2022, and 1960–1980 and 2005–2010 are the rapidly developed periods. The construction history of the reservoirs indicates that reservoirs' locations have the spatial tendency to shift from being in oasis plain areas to mountainous areas, and the newly built reservoirs showed a trend of miniaturization in area and maximization in volume. [ABSTRACT FROM AUTHOR]

Hydraulic fracturing is an essential technique for reservoir stimulation in the process of deep energy exploitation. Granite is composed of different rock-forming minerals and exhibits obvious heterogeneity at the mesoscale, which affects the strength and deformation characteristics of rocks and controls the damage and failure processes. Therefore, in this paper, based on the discrete element fluid-solid coupling algorithm and multiple parallel bond-grain based model (Multi Pb-GBM), a numerical model of a granite hydraulic fracturing test is established to study the evolution of hydraulic fractures in crystalline granite under different ground stress conditions. The main conclusions are as follows. The crack propagation of hydraulic fractures in granite is determined by the in situ stress state, crystal size, and mineral distribution, and the ground stress is the main controlling factor. The final fracture mode affects the maximum principal stress and shear stress, and the generation of cracks changes the distribution of the stress field. The hydraulic fracturing initiation pressure decreases with decreasing crystal size. The influence of the crystal size on the crack inclination angle is mainly reflected in local areas, and the general trend of the fissure dip angle distribution is along the direction of the maximum in situ stress. This study not only has important theoretical significance for clarifying the propagation mechanism of hydraulic fractures but also provides a theoretical basis for deep reservoir reconstruction and energy extraction. [ABSTRACT FROM AUTHOR]