Search

This research introduces TrajBERT-DSSM, a novel methodology for predicting vessel destinations, to address the growing importance of understanding vessel trajectory in marine transportation. This approach uses AIS records to compare the trajectories of the travelling vessel with the historical one, focusing on spatiotemporal correlation, geometrical attributes, and movement patterns. We define similarity as the likelihood that two trajectories have the same destination. The destination of the most similar past trajectory is predicted to be the destination port of the travelling vessel. TrajBERT-DSSM uniquely combines geohash, a hierarchical geocoding system, and TrajBERT, a contextual embedding model, with the Deep Semantic Similarity Model (DSSM) to measure trajectory similarity and predict destinations. Geohash encodes vessel coordinates into strings, maintaining spatial proximity, while TrajBERT interprets these codes to produce high-dimensional vector embeddings of trajectories. The effectiveness of TrajBERT-DSSM was affirmed through comparative studies against eleven advanced methods. The experimental results demonstrate that TrajBERT-DSSM achieves the best prediction accuracy and stability on 11,555 testing trajectories. • Leverage deep bidirectional representations to predict their destinations. • Contextual embedding model integrates domain knowledge of vessel voyage • Comparative studies validate the proposed model outperforms eleven state-of-the-arts. • The prediction model is without restrictions on regions, time ranges, etc. [ABSTRACT FROM AUTHOR]

A phase field model (PFM) for simulating multi-mode brittle fractures in transversely isotropic rock is proposed. In the model, we employ a novel decomposition form of strain energy consisting of tensile, tensile–shear and compressive–shear components to improve the fracture driving force, and the positive–negative decomposition forms of elastic energy are considered as an approximate extension of the spectral decomposition method for transversely isotropic materials. We also introduce the tensor form of the critical energy release rate and Lamé constants, which breaks through the limitations of traditional PFMs that cannot well reproduce fracture anisotropy and multi-mode brittle fracture in transversely isotropic rock, i.e. the tensile, shear, and mixed-mode fractures. A new mixed phase field model is established, which is solved in COMSOL by employing a staggered scheme to solve the phase field, displacement field, and historical field independently. Subsequently, the accuracy and effectiveness of the proposed model are comprehensively verified by a series of 2D and 3D numerical examples. The results show that the proposed model agrees well with the existing model in capturing tensile and tensile–shear fractures. Compressive–shear fractures can be additionally captured well in the compressive state. • Novel phase field model for simulating multi-mode rock fractures. • Driving force based on a new strain energy decomposition. • Fracture energy anisotropy, deformation anisotropy, and shear resistance are considered. [ABSTRACT FROM AUTHOR]

• Create a database from more than 141 million AIS records. • Propose a random forest-based model to measure the similarity between trajectories. • Develop a decision strategy for vessel destination prediction. Shipping is one of the major transportation approaches around the world. With the growing demands for global shipping service, vessel destination prediction has shown its significant role in improving the efficiency of decision making in industry and ensuring a safe and efficient maritime traffic environment. Currently, most vessel destination prediction methods focus on regional destination prediction, which has restrictions on destinations and regions. Thus, this paper proposes a general AIS (Automatic Identification System) data-driven model for vessel destination prediction. In this random forest-based model, the similarity between the vessel's traveling and historical trajectories are measured and utilized to predict the destination. The destination of the historical trajectory, which shares the highest similarity with the traveling trajectory, is predicted as the vessel's destination. The method is different from previous work which used maritime records as input to predict the destination. In our method, a historical trajectory database was generated from more than 141 million AIS records, which covers 534,824 traveling patterns between ports and more than 5.9 million historical trajectories. Comparative studies were carried out to validate the performance of the proposed model, where eleven state-of-the-art trajectories similarity measurement methods combined with two different decision strategies were implemented and compared. The experimental results demonstrate that the proposed model combined with the port frequency-based decision strategy achieves the best prediction accuracy on 35,937 testing trajectories. [ABSTRACT FROM AUTHOR]

The heavy metal contamination like Cr(VI) has been increased by human activities and that threats the ecosystem health of mangrove areas. Bioindicator is an emerging tool in the environmental contamination assessment. The objective of this study was to investigate the Geloina erosa response mechanisms and sensitivities of several biomarkers in the Cr(VI) exposure and identify the G. erosa capability of being used as heavy metals bioindicator. In this study, G. erosa was exposed to 100 μmol·L−1 Cr(VI) for 48 h. After transcriptome sequencing, a total of 134,817 unigenes were obtained, including 12,555 up-regulated and 18,829 down-regulated differentially expressed genes and were validated through quantitative real-time PCR. In addition, a total of 12,185 SSRs and 1,428,214 candidate SNPs were identified from all the G. erosa transcriptome libraries. Histopathology of the gill indicated the Cr(VI) exposure induced damage of the organ leading to its immunization, detoxification or apoptosis reactions. Among eight genes of the selected biomarkers, Calm, HSP70, CYP450, ATG5, TLR2, MYD88 and CASP8 were up-regulated, while TLR4 was down-regulated in response to the Cr(VI) exposure. Unlabelled Image • Histological changes in Cr(VI) exposed gill were observed. • It is the first transcriptome analysis of Geloina erosa gills in response to Cr(VI). • 134817 contigs were obtained from four libraries, and 12555 up-regulated genes and 18829 down-regulated genes were identified. • A total of 10016 SSRs and 1428214 candidate SNP loci were identified in the gill transcriptome of Geloina erosa. • Eight genes of the selected biomarkers, Calm, HSP70, CYP450, ATG5, TLR2, MYD88 and CASP8 were identified [ABSTRACT FROM AUTHOR]

A novel method using multilateral wells to perform oil shale in-situ conversion process is proposed in this paper. This method constructs radial branches in upper and lower oil shale formation as injection and production wells. Hot fluids are injected from the injection wells, and pyrolyzed oil and gas are extracted by production wells. In this study, a 3D transient model coupling fluid flow, heat transfer and chemical process is established and implemented on COMSOL Multiphysics platform to investigate the oil shale in-situ conversion process. The temperature field, production characteristics and energy performance are characterized. Sensitivity of oil shale properties and operational parameters are analyzed. Influences of multilateral-well arrangements are studied. The simulation results indicate that the products are strongly dependent to oil shale temperature. The specific heat capacity, injection fluid temperature and injection mass flow rate can significantly influence production performance, while thermal conductivity has negligible effect. Multilateral wells with 5 branches, 60° branch angle and 40 m branch length show the best production performance among the computational cases. This study provides comprehensive insights and suggestions for the application of multilateral wells in oil shale in-situ conversion process. • A novel oil shale in-situ conversion method using multilateral wells is proposed. • A 3D model coupling fluid flow, heat transfer and chemical process is established. • The temperature field and production performance are analyzed comprehensively. • Effects of oil shale properties and operational parameters are discussed. • Influences of multilateral-well arrangements are studied. [ABSTRACT FROM AUTHOR]

Benzoic acid is a representative allelochemical in soils affected by continuous cropping obstacles, and it is also a typical aromatic pollutant, causing a direct or indirect inhibition on crop growth. In this study, a strain B28 that was isolated from the rhizosphere of peanut by using benzoic acid as the sole carbon source and sequenced. Strain B28 contained a single circular chromosome, the complete genome size was 4,132,583 bp, the average GC was 43.89%, with 3928 predicted genes, and based on phylogenomic analyses, B28 was designated as Bacillus halotolerans. The strain B28 with outstanding abilities to biodegrade benzoic acid, enhance plant growth and control soil-borne diseases, which were the key to the phytoprobiotic functions. Genome mining revealed the potential genomic properties associated with plant growth promotion, such as aromatic xenobiotics biodegradation, IAA synthesis, biofilm formation, chemotaxis motility and secondary metabolite synthesis. The effects of strain B28 on the growth of Arachis hypogaea L. (peanut) was evaluated, and the physiological parameters of peanut such as fresh weight, dry weight, plant height and root length were significantly higher after inoculating strain B28. Overall, our present results indicate that Bacillus halotolerans B28 is a potential PGPR strain for benzoic acid bioremediation and organic agriculture application. • The bioremediation of benzoic acid in environment is being studied. • Novel benzoic acid degrading Bacillus halotolerans strain B28 was isolated and sequenced. • High-efficiency plant growth-promoting rhizobacteria with the function of degrading benzoic acid, promoting plant growth and control soil-borne diseases. • Abundant genes putatively responsible biodegradation and phytoprobiotic were determined. [ABSTRACT FROM AUTHOR]

Energy load forecasting is critical for sustainable building development and management. Although the energy data could be collected through Internet of Things (IoT) systems, it is a big challenge to train a large-scale machine learning model due to data isolation. Since the building energy data could reveal confidential information such as user behaviors and building operations, the privacy regulations would not allow central service to collect distributed data from data owners directly. This paper designs a secure federated data analytics system for forecasting community buildings' energy data load. A novel adaptive weight federated learning algorithm is proposed to handle the system faults frequently happening during networking operations. Moreover, a new deep learning model is re-invented to improve energy load forecasting performance. The experiments of the system are performed on an actual university campus dataset, and the results show the new federated algorithm improves the load forecasting accuracy and achieves the best load forecasting result. The new deep learning model improves the forecasting accuracy by almost 10% on error reduction under the same federated learning settings. To evaluate the load forecasting model's practical usefulness, an anomaly prediction pipeline is designed through the combination of gaussian mixture model and load forecasting model, which reveals the system's effectiveness at building energy management that 92% F1 score with 97% accuracy is achieved by the best model. • Develop an adaptive federated learning algorithm for community-level building energy load forecasting. • Re-invent a new deep learning load forecasting model through a combination of existing methodologies. • Design an anomaly prediction pipeline through a combination of load forecasting models and unsupervised GMM. • Evaluate the practical usefulness of load forecasting and anomaly prediction through extensive experiments. [ABSTRACT FROM AUTHOR]

With the miniaturization of the light-emitting diodes (LEDs), it is a challenge to detect the defective LED-chip with neural networks. Convolution neural networks (CNNs) can capture local information, but the deeper or wider layer with a higher computational cost is required to achieve a stronger model capacity. Additionally, the vision transformer (ViT) is able to extract global information at the cost of large datasets, extensive data augmentation and computational complexity. To solve the aforementioned issues, this paper proposes a local–global lightweight ViT model based on Convmixer for mini/micro-LED-chip defect recognition, which combines the advantages of CNNs and ViT to capture both detailed and global information and is suitable for resource-constrained embedded or mobile devices. A fine-coarse-grained convolutional block is proposed to capture different sizes of features in the spatial dimension with depth-wise convolution. Then, the multi-mode and cross-channel information is extracted by the point-wise convolution. Finally, a position attention module embedded with the long-range relationship is presented to highlight the important patches and captures global information. Besides, the structure of the proposed model is plain so that the detailed information is preserved. Experiments show that the proposed model achieves a higher accuracy of 96.17% with fewer parameters and of 0.038M and a lower computational cost of 0.138G, which outperforms the state-of-the-art. [ABSTRACT FROM AUTHOR]

• Annealing promotes to form more Co3+ and oxygen vacancies on the LaCoO 3-δ surface. • Triple-phase boundaries act as the active sites for electrocatalysis. • The nanoporous Ag-LaCoO 3-δ catalyst shows improved ORR performance. Substantial progress has been made in developing perovskite-based catalysts for oxygen reduction reaction (ORR). This article reported an ORR electrocatalyst composed of nanoporous Ag (NP-Ag) and LaCoO 3-δ (LC) perovskite oxide and designed to take advantage of the bi-continuous NP structure and high electrocatalytic performance of perovskite oxides. The triple-phase boundaries (TPBs) formed by NP-Ag and LC provided active sites for the electrocatalytic reactions. Further annealing promoted the interaction between NP-Ag and LC, which induced the formation of more Co3+ and surface oxygen vacancies, as proved by multiple characterizations. When the mass ratio of LC and NP-Ag was 1:2 (LC/Ag-1/2-A), the catalyst showed the best ORR performance with a relatively high half-wave potential, and only attenuated by 4% after 5,000 cycles of accelerated durability testing. Besides, it delivered significant bifunctionality towards both ORR and OER, bearing comparison with the reported perovskite-based electrocatalysts. This work demonstrated a novel route for extending the application of NP metals and facilely designing efficient bifunctional electrocatalysts. By mixing and annealing with nanoporous Ag, the charge transfer and surface chemical conditions of LaCoO 3-δ perovskite is modified to boost the ORR performance. [Display omitted] [ABSTRACT FROM AUTHOR]