Freeze-thaw action has caused many engineering geology disasters in cold regions, such as frost cracking of tunnel and pavement, rockfall of rock slope, falling off of building materials. The ...freeze-thaw damage of porous rock is mainly induced by the frost heaving pressure due to pore water freezing. Although some elastic crystallization theories have been proposed to model the freezing process of pore water, the long-term frost heave characteristics under cyclic freeze-thaw are rarely studied. The ice crystallization and frost heave theory under freeze-thaw have been developed in the elastoplastic regime in this study. A novel elastoplastic model has been proposed to estimate the long-term frost heave and frost damage under freeze-thaw. This model has considered the influence of the pore size, yield stress of matrix and degree of water saturation on the development of the frost heaving pressure in pores. During freezing, a plastic region will produce inside the matrix around the pore. This plastic region gradually expands outwards with the increase of this pressure. During thawing, the elastic deformation will be recovered but a considerable residual plastic strain is accumulated. However, the growth rate of the residual plastic strain will decrease under freeze-thaw due to the production of new empty voids. Therefore, the freezing of pore water in porous rock is an elastoplastic problem. Besides, the water saturation and mechanical properties of the porous rock are very important to quantify the frost heave. This study provides a better understanding of the frost damage of the porous rock and a reference for the stability of rock engineering under freeze-thaw in cold regions.
•The elastoplastic mechanical behavior of brittle porous media during ice crystallization process•A novel elastoplastic model of frost deformation under freeze-thaw•The variable of plastic region and frost heaving pressure under freeze-thaw•The influence of water saturation and new voids on the residual strain under freeze-thaw
Multi-source social sensing data provide new opportunities to identify urban functions from the perspective of human activity. The information embedded in multi-source data typically needs to be ...fused to obtain a comprehensive view of urban functions. Although multi-view clustering has been successfully used to fuse multi-source social sensing data, the adaptive determination of fusion weights for high-dimensional and noisy multi-source social sensing data remains challenging. Therefore, this study proposes an adaptive weighted multi-view subspace clustering (AWMSC) method. First, we use two neural networks to map multi-source data into a common latent representation and multiple specific latent representations, which serve as the query vector and input vectors of the attention mechanism, respectively. Then, the weight of each type of data is calculated based on the attention mechanism. Finally, the specific latent representations of the multi-source data are weighted and fused into a shared subspace representation, which is used as the input of the spectral clustering algorithm to obtain clustering results. AWMSC is applied to identify urban functional zones in Beijing using bus transactions, taxi trajectories, and points of interest datasets. The results show that AWMSC outperforms the typical single-view, weighted-average, and representative multi-view methods. AWMSC can obtain a comprehensive understanding of urban functional zones which may help government departments make more accurate strategic decisions.
Due to inter-modal effects hidden in multi-modalities and the impact of weak modalities on multi-modal entity alignment, a Multi-modal Entity Alignment Method with Inter-modal Enhancement (MEAIE) is ...proposed. This method introduces a unique modality called numerical modality in the modal aspect and applies a numerical feature encoder to encode it. In the feature embedding stage, this paper utilizes visual features to enhance entity relation representation and influence entity attribute weight distribution. Then, this paper introduces attention layers and contrastive learning to strengthen inter-modal effects and mitigate the impact of weak modalities. In order to evaluate the performance of the proposed method, experiments are conducted on three public datasets: FB15K, DB15K, and YG15K. By combining the datasets in pairs, compared with the current state-of-the-art multi-modal entity alignment models, the proposed model achieves a 2% and 3% improvement in Top-1 Hit Rate(Hit@1) and Mean Reciprocal Rank (MRR), demonstrating its feasibility and effectiveness.
For multiple transmitter-receiver pairs communication in a frequency-selective environment, typical power allocation method is the Iterative-Waterfilling (IW) algorithm. Main drawback of IW is its ...poor convergence performance, including low convergence probability and slow convergence speed in certain scenarios, which lead to high computational load. Large-scale network significantly magnifies the above drawback by lowering the convergence probability and convergence speed, which is difficult to satisfy real-time requirements. In this work, we propose a power allocation scheme based on convolutional neural network (CNN). The design of loss function takes into account the Sum Rate (SR) of all users. The output layer of the CNN model is replaced by several Softmax blocks, and the output of each Softmax block is the ratio of the transmission power of each user on the sub-carrier to the total power. Numerical studies show the advantages of our proposed scheme over IW: with the constraint of not lowering SR, there is no convergence problem and the computational load is significantly reduced.
The current mainstream computed tomography (CT) reconstruction methods based on deep learning usually need to fix the scanning geometry and dose level, which significantly aggravates the training ...costs and requires more training data for real clinical applications. In this paper, we propose a parameter-dependent framework (PDF) that trains a reconstruction network with data originating from multiple alternative geometries and dose levels simultaneously. In the proposed PDF, the geometry and dose level are parameterized and fed into two multilayer perceptrons (MLPs). The outputs of the MLPs are used to modulate the feature maps of the CT reconstruction network, which condition the network outputs on different geometries and dose levels. The experiments show that our proposed method can obtain competitive performance compared to the original network trained with either specific or mixed geometry and dose level, which can efficiently save extra training costs for multiple geometries and dose levels.
Low-dose computed tomography (LDCT) scans, which can effectively alleviate the radiation problem, will degrade the imaging quality. In this paper, we propose a novel LDCT reconstruction network that ...unrolls the iterative scheme and performs in both image and manifold spaces. Because patch manifolds of medical images have low-dimensional structures, we can build graphs from the manifolds. Then, we simultaneously leverage the spatial convolution to extract the local pixel-level features from the images and incorporate the graph convolution to analyze the nonlocal topological features in manifold space. The experiments show that our proposed method outperforms both the quantitative and qualitative aspects of state-of-the-art methods. In addition, aided by a projection loss component, our proposed method also demonstrates superior performance for semi-supervised learning. The network can remove most noise while maintaining the details of only 10% (40 slices) of the training data labeled.
•A multi-task attention-LSTM load forecasting model is proposed.•Adaptive combo-task-wise loss function is proposed to optimize our model.•MTAL proves to be having good accuracy and robustness for ...load forecasting.
With the rapid development of power markets, smart grids and large-scale renewable energy generation, it is crucial to be able to accurately predict both reactive and active power loads. In this paper, we propose a novel multi-task load forecasting method for predicting both active and reactive power simultaneously. The long short-term memory (LSTM) architecture is employed in the backbone prediction model, supported by an attention mechanism to prevent performance deterioration. Considering the latent dynamic correlations between the reactive and active loads of a substation, multi-task regression based on hard parameter sharing is adopted to treat the forecasting of both types of loads as parallel subtasks. Meanwhile, we design an adaptive combined-task-wise loss function to optimize the proposed multi-task load forecasting model to avoid biasing of the final model for any subtask. We compare our multi-task attention-LSTM (MTAL) model with other popular single-task load forecasting models and achieve superior accuracy on both subtasks. The results indicate that the proposed method is robust and reliable for practical applications in power systems.
Optical coherence tomography (OCT) has become a very promising diagnostic method in clinical practice, especially for ophthalmic diseases. However, speckle noise and low sampling rates have ...intensively reduced the quality of OCT images, which prevents the development of OCT-assisted diagnosis. Therefore, we propose a generative adversarial network-based approach (named SDSR-OCT) to simultaneously denoise and super-resolve OCT images. Moreover, we trained three different super-resolution models with different upscale factors (2× , 4× and 8×) to adapt to the corresponding downsampling rates. We also quantitatively and qualitatively compared our proposed method with some well-known algorithms. The experimental results show that our approach can effectively suppress speckle noise and can super-resolve OCT images at different scales.
The presence of water has a great influence on the strength and deformation of clay-bearing sandstone. Most of the natural rocks suffer water softening of different saturations. In this study, due to ...the importance of the clay-bearing rocks in geoengineering practice, red sandstone was chosen to investigate the effect of water contents on the UCS (uniaxial compressive strength), TCS (triaxial compressive strength), BTS (Brazilian tensile strength), PLS (point load strength), and DSS (direct shear strength). The UCS and TCS are linearly decreased as the water saturation increases. When the water saturation increases from 0 to 100%, the UCS and TCS have decreased by about 52.8% and 29.3%, respectively. The BTS, PLS, and DSS are exponentially decreased with increasing water saturation, which have a reduction of about 45.5%, 66.3%, and 49%, respectively. Four typical deformation stages are observed during compression for this red sandstone samples, including an obvious compaction stage. The water saturation does not have any effect on the ratio of the strain on the transition points of these stages to the peak strain, as well as the stress. Besides, a rapid increase of the AE count occurs at the plastic yield point although the cumulative AE count has an obvious decrease trend with increasing the water saturation. Then, considering the effect of water saturation, a complete constitutive model has been proposed to well predict the stress-strain relationship of the red sandstone. The water’s weakening effect on the clay-bearing red sandstone has also been deeply discussed and developed.
Due to its noninvasive character, optical coherence tomography (OCT) has become a popular diagnostic method in clinical settings. However, the low-coherence interferometric imaging procedure is ...inevitably contaminated by heavy speckle noise, which impairs both visual quality and diagnosis of various ocular diseases. Although deep learning has been applied for image denoising and achieved promising results, the lack of well-registered clean and noisy image pairs makes it impractical for supervised learning-based approaches to achieve satisfactory OCT image denoising results. In this paper, we propose an unsupervised OCT image speckle reduction algorithm that does not rely on well-registered image pairs. Specifically, by employing the ideas of disentangled representation and generative adversarial network, the proposed method first disentangles the noisy image into content and noise spaces by corresponding encoders. Then, the generator is used to predict the denoised OCT image with the extracted content features. In addition, the noise patches cropped from the noisy image are utilized to facilitate more accurate disentanglement. Extensive experiments have been conducted, and the results suggest that our proposed method is superior to the classic methods and demonstrates competitive performance to several recently proposed learning-based approaches in both quantitative and qualitative aspects. Code is available at: https://github.com/tsmotlp/DRGAN-OCT .
PDF
