de Hua Zhu

This paper presents a novel form of rectangular fiber-reinforced polymer (FRP)-concrete-high strength steel (HSS) hybrid columns referred to as rectangular double-tube concrete columns (DTCCs). These columns consist of a rectangular FRP tube filled with concrete and an elliptical HSS tube as internal reinforcement. The inner elliptical steel tube encloses most of the concrete in the rectangular section and therefore most of the concrete in the section is effectively confined. The high susceptibility of the HSS elliptical tube to local buckling is expected to be suppressed by the confined concrete, leading to full utilization of the high strength of HSS. This paper demonstrates the excellent performance of the new columns due to the above beneficial interactions between different components in the column through an experimental study. In the experimental study, twelve rectangular DTCCs covering different FRP tube thicknesses and aspect ratios were tested under axial compression. Finally, an existing stress–strain model for FRP-confined concrete in rectangular columns is modified to be applicable to the concrete in the new column by accounting for the confinement effect from the inner elliptical HSS tube. It is found that the modified model provides more accurate predictions on the stress–strain curves of confined concrete in the rectangular DTCCs than the original model.

We discuss the production of doubly strange \ensuremath{\Lambda}\ensuremath{\Lambda} hypernuclei in ${\mathrm{\ensuremath{\Xi}}}^{\mathrm{\ensuremath{-}}}$ atomic capture. The two-body ${\mathrm{\ensuremath{\Xi}}}^{\mathrm{\ensuremath{-}}}$${+}^{6}$Li${\ensuremath{\rightarrow}}_{\mathrm{\ensuremath{\Lambda}}\mathrm{\ensuremath{\Lambda}}}^{6}$He+n reaction is a relatively favorable case, with a yield of order 3%. For heavier p-shell targets, such as $^{14}\mathrm{N}$, the (1${\mathit{s}}_{\mathrm{\ensuremath{\Lambda}}}$${)}^{2}$ ground-state yields in the ${\mathrm{\ensuremath{\Xi}}}^{\mathrm{\ensuremath{-}}}$${+}^{\mathit{A}}$Z${\ensuremath{\rightarrow}}_{\mathrm{\ensuremath{\Lambda}}\mathrm{\ensuremath{\Lambda}}}^{\mathit{A}}$(Z-1)+n reaction are suppressed, and states of (1${\mathit{s}}_{\mathrm{\ensuremath{\Lambda}}}$1${\mathit{p}}_{\mathrm{\ensuremath{\Lambda}}}$) structure, coupled to an excited nuclear core, are preferentially populated.

Abstract In high-altitude nuclear detonations, the proportion of pulsed X-ray energy can exceed 70%, making it a specific monitoring signal for such events. These pulsed X-rays can be captured using a satellite-borne X-ray detector following atmospheric transmission. To quantitatively analyze the effects of different satellite detection altitudes, burst heights, and transmission angles on the physical processes of X-ray transport and energy fluence, we developed an atmospheric transmission algorithm for pulsed X-rays from high-altitude nuclear detonations based on scattering correction. The proposed method is an improvement over the traditional analytical method that only computes direct-transmission X-rays. The traditional analytical method exhibits a maximum relative error of 67.79% compared with the Monte Carlo method. Our improved method reduces this error to within 10% under the same conditions, even reaching 1% in certain scenarios. Moreover, its computation time is 48,000 times faster than that of the Monte Carlo method. These results have important theoretical significance and engineering application value for designing satellite-borne nuclear detonation pulsed X-ray detectors, inverting nuclear detonation source terms, and assessing ionospheric effects.

This paper considers a novel application of deep AUC maximization (DAM) for multi-instance learning (MIL), in which a single class label is assigned to a bag of instances (e.g., multiple 2D slices of a CT scan for a patient). We address a neglected yet non-negligible computational challenge of MIL in the context of DAM, i.e., bag size is too large to be loaded into {GPU} memory for backpropagation, which is required by the standard pooling methods of MIL. To tackle this challenge, we propose variance-reduced stochastic pooling methods in the spirit of stochastic optimization by formulating the loss function over the pooled prediction as a multi-level compositional function. By synthesizing techniques from stochastic compositional optimization and non-convex min-max optimization, we propose a unified and provable muli-instance DAM (MIDAM) algorithm with stochastic smoothed-max pooling or stochastic attention-based pooling, which only samples a few instances for each bag to compute a stochastic gradient estimator and to update the model parameter. We establish a similar convergence rate of the proposed MIDAM algorithm as the state-of-the-art DAM algorithms. Our extensive experiments on conventional MIL datasets and medical datasets demonstrate the superiority of our MIDAM algorithm.

Domestic waste classification is of great significance for building a beautiful living environment. Waste classification is a scientific management method for effective disposal of waste. Aiming at the problem of garbage pollution faced by human beings at present, a garbage classification algorithm based on deep learning is proposed.In this paper, we first build a garbage classification dataset, identify the categories of garbage items in the picture, and then build a garbage classification model. Through the comparison of different models, we test the accuracy of different models on the garbage classification results, and find a more effective and accurate garbage classification algorithm model.The experiment shows that this algorithm analyzes the existing model, preprocesses the data image, and then uses the image enhancement method to increase the robustness of the model used. It improves the model by adjusting the learning rate, label smoothing, and data normalization processing strategies. The optimized algorithm precision increases to 95%, and makes the model have better generalization ability.

Digital Financial Inclusion is a hot topic under the theme of China's economy today. Digital Financial Inclusion is an important supplement and extension of traditional Financial inclusion. Its core idea is to reduce the threshold of financial services and ease the financing difficulties of low-income groups, small and micro enterprises, and other vulnerable groups by providing low-cost and efficient financial services to the public. The empirical results of much literature show that digital Financial inclusion has narrowed the gap between the rich and the poor, which plays a vital role in the high-quality development of China's economy. Therefore, it is important to explore it's influencing factors and analyze the influencing mechanism. This study focuses on the three provinces in eastern China and applies the grounded theory approach. Utilizing data from the Peking University database, 17 factors influencing the level of digital inclusive finance are selected. PCA (principal component analysis) is used to reduce dimensions, and ANN-MLP (artificial neural network multi-layer perceptron) algorithm is used to explore the contribution factors of Digital Financial Inclusion, explore important factors, and put forward relevant suggestions. The significance of this study is that taking the three provinces in eastern China as an example, it explores the factors influencing the development of DFI in China, and makes an important contribution to the issue of how to promote the rapid development of digital Financial inclusion in China.

In high energy particle physics, accelerator- and detector-upgrades always go hand in hand. The instantaneous luminosity of the Large Hadron Collider will increase to up to L = 2×1034cm−2s−1 during Run 2 until 2023. In order to cope with such luminosities, the pixel detector of the CMS experiment has been replaced early 2017. The so-called CMS Pixel phase 1 upgrade detector consists of 1184 modules with new design. An important production step is the module qualification and calibration, ensuring their proper functionality within the detector. This paper summarizes the qualification and calibration tests and results of modules used in the innermost two detector layers with focus on methods using module-internal calibration signals. Extended characterizations on pixel level such as electronic noise and bump bond connectivity, optimization of operational parameters, sensor quality and thermal stress resistance were performed using a customized setup with controlled environment. It could be shown that the selected modules have on average 0.55‰ ± 0.01‰ defective pixels and that all performance parameters stay within their specifications.