基于改进MobileFaceNet的人脸识别方法北大核心

针对目前大多数人脸识别算法参数多、计算量大,难以部署到移动端和嵌入式设备中的问题,提出了一种基于改进MobileFaceNet的人脸识别方法。通过对MobileFaceNet模型结构的调整,将bottleneck模块优化为sandglass模块,改良深度卷积和逐点卷积的相对位置,适当增大sandglass模块的输出通道数,从而减少特征压缩时的信息丢失,增强人脸空间特征的提取。实验结果表明:改进后的方法在LFW测试数据集上准确率达99.15%,模型大小和计算量分别仅为原算法的61%和45%,验证了所提方法的有效性。     

Mutual information maximizing GAN inversion for real face with identity preservation

Recent generative adversarial networks (GANs) have yielded remarkable performance in face image synthesis. GAN inversion embeds an image into the latent space of a pretrained generator, enabling it to be used for real face manipulation. However, current inversion approaches for real faces suffer the dilemma of initialization collapse and identity loss. In this paper, we propose a hierarchical GAN inversion for real faces with identity preservation based on mutual information maximization. We first use a facial domain guaranteed initialization to avoid the initialization collapse. Furthermore, we prove that maximizing the mutual information between inverted faces and their identities is equivalent to minimizing the distance between identity features from inverted and original faces. Optimization for real face inversion with identity preservation is implemented on this mutual information-maximizing constraint. Extensive experimental results show that our approach outperforms state-of-the-art solutions for inverting and editing real faces, particularly in terms of face identity preservation.     

Multi-scale and multi-patch transformer for sandstorm image enhancement

Sandstorm is a meteorological phenomenon common in arid and semi-arid regions. A sandstorm can carry large volumes of sand unexpectedly, which leads to severe color deviations and significantly degraded visibility when an image is taken in such a scenario. However, existing image enhancement methods cannot enhance sandstorm images well due to the challenging degradations and the scarcity of sandstorm training data. In this paper, we propose a Transformer with rotary position embedding to perform sandstorm image enhancement via building multi-scale and multi-patch dependencies. Our key insights in this work are 1) a multi-scale Transformer can globally eliminate the color deviations of sandstorm images via aggregating global information, 2) a multi-patch Transformer can recover local details well via learning the spatial variant degradations, and 3) a U-shape Transformer with rotary position embedding as the core unit of multi-scale and multi-patch Transformer can effectively build the long-range dependencies. We also contribute a real-world Sandstorm Image Enhancement (SIE) dataset including 1,400 sandstorm images with different degrees of degradations and various scenes. Experiments performed on synthetic images and real-world sandstorm images demonstrate that our proposed method not only obtains visually pleasing results but also outperforms state-of-the-art methods qualitatively and quantitatively.     

碳中和背景下配电网差异化节能降耗潜力评估分

为了提高配电网差异化节能降耗效果，解决现有潜力评估方法存在的应用性能差的问题，提出碳中和背景下配电网差异化节能降耗潜力优化评估方法。根据配电网的空间结构，构建相应的等值电路模型。在该模型下，从设备损耗和运行附加损耗2个方面计算配电网的损耗量。根据损耗量计算结果，确定配电网差异化碳中和节能降耗方式。从静态和动态2个角度设置潜力评估指标，通过指标数据处理、指标权重求解等步骤，得出配电网差异化节能降耗潜力的综合量化评估结果。将设计潜力评估方法应用到配电网的差异化节能降耗改造工作中，能够有效降低配电网的实际线损量、降低区域损耗费用，并具有较高的应用价值。     

Accelerating the decarbonization of carbonaceous gold ore by suspension oxidation roasting towards the improvement of gold leaching efficiency

Eliminating the gold preg-robbing effect of carbonaceous matter in carbonaceous gold ores is crucial for gold leaching. In this study, suspension oxidation roasting was proposed to accelerate the decarbonization of carbonaceous gold ore. The characteristics of oxidation reaction process and gas release were analyzed by TG-DTA-FTIR. The phase transformation and microstructure evolution of samples during roasting were analyzed by XRD, SEM and BET. The results show that the gold preg-robbing effect was eliminated after the gasification of carbonaceous matter, and the CaO generated by decomposition of carbonates can effectively capture the SO₂. After roasting for 75 min at 650 °C in a 20% O₂ atmosphere, the total carbon removal rate reached 99.42%, the distribution of exposed gold increased from 28.85% to 77.10% and the gold leaching efficiency increased from 4.55% to 84.83%. In addition, about 70% sulfur was mainly fixed in the roasted products in the form of sulfate. Therefore, the suspension oxidation roasting process is an efficient and clean pretreatment method for carbonaceous gold ores.     

Combining a new baseboard radiator with a fan coil system for improving heating performance and reduce energy consumption

In the present work, the heating performance of a new system combined with a new modified baseboard radiator and fan coil is investigated. Using longitudinal fins with special geometry and also forced airflow at the end of the system causes that at the lower inlet water temperature compared with the conventional models, higher heat output rate be obtained. The heat output rate of the new modified system is obtained by experimental metrology based on the European Standard No. EN-442. Temperature and velocity distribution in the room space is done by simulation of the modified system in the Flovent software. Computational fluid dynamics (CFD) results are validated against experimental results and there is a good agreement between them. Also, the energy consumption of the system during the winter season is calculated in TRANSYS software. Experimental results show that the heat output rate of a new modified heating system with inlet water temperature in the range of 45–55°C is on average 4.17 times higher compared with the conventional model. CFD simulation also showed that the combined system provides good thermal comfort conditions. Energy consumption of the new system reduced about 13% compared with conventional models.     

Influence of radiation and viscous dissipation on MHD heat transfer Casson nanofluid flow along a nonlinear stretching surface with chemical reaction

The present article investigates the influence of Joule heating and chemical reaction on magneto Casson nanofluid phenomena in the occurrence of thermal radiation through a porous inclined stretching sheet. Consideration is extended to heat absorption/generation and viscous dissipation. The governing partial differential equations were transformed into nonlinear ordinary differential equations and numerically solved using the Implicit Finite Difference technique. The article analyses the effect of various physical flow parameters on velocity, heat, and mass transfer distributions. For the various involved parameters, the graphical and numerical outcomes are established. The analysis reveals that the enhancement of the radiation parameter increases the temperature and the chemical reaction parameter decreases the concentration profile. The empirical data presented were compared with previously published findings.     

疫情下电磁场理论课程的网络教学

本文主要总结了新冠疫情期间作者的电磁场理论课程在线教学经验。对比分析了录播和直播的优缺点后,选择录播教学方式。基于超星网络教学平台,展示了录播网络教学的具体措施,包括网上答疑和学习效果检查以及在线批改作业等。给出了网络教学可以为线下教学继续使用的方法和手段,为疫情结束后的正常教学提供了新的网络教学补充措施。     

5G核心网业务模型的智能化预测研究

摘 要：核心网业务模型的建立是5G网络容量规划和网络建设的基础，通过现有方法得到的理论业务模型是静态不可变的且与实际网络存在偏离。为了克服现有5G核心网业务模型与现网模型适配性较差以及规划设备无法满足用户实际业务需求的问题，提出了一种长短期记忆（long short-term memory，LSTM）网络与卷积LSTM （convolution LSTM，ConvLSTM）网络双通道融合的 5G 核心网业务模型预测方法。该方法基于人工智能（artificial intelligence，AI）技术以实现高质量的核心网业务模型的智能预测，形成数据反馈闭环，实现网络自优化调整，助力网络智能化建设。     

一种可传输轨道角动量的螺旋光子晶体光纤北大核心

设计了一种基于阿基米德螺线的新型螺旋光子晶体光纤,该光纤以二氧化硅为基底材料,包层由24个螺旋臂组成,每个螺旋臂包含11个小空气孔,纤芯设有大空气孔,包层与纤芯中间的环形区域用于传输轨道角动量模式。该结构在1300~1800 nm波段上可支持22种轨道角动量模式稳定传输,在1550 nm波长下,有效折射率差最高可达2.89×10^(-3),色散系数最低可达66.4 ps/(nm·km),非线性系数最低可达2.17 W^(-1)·km^(-1),且1500~1600 nm波段上的色散值变化均小于15.15 ps/(nm·km)。此螺旋光子晶体光纤不仅结构简单,且具有低非线性、色散平坦的性能,为螺旋光子晶体光纤的设计提供了思路。