Magnetic induction (MI) communication has the characteristics of being less affected by the medium and having a stable channel, which can effectively solve the problem of excessive transmission loss of electromagnetic wave (EW) communication in extreme environments. However, as a near−field communication technology, the limitation of its near−field effective range leads to the restricted coverage of magnetic induction communication. Therefore, in−depth research on the magnetic induction boundary is of great significance for guiding the design and optimization of magnetic induction communication systems. This paper studies the spatial distribution of the electromagnetic field of a thin ring antenna in free space, comprehensively considers parameters such as antenna size, and proposes a decision parameter to give a stricter magnetic induction boundary. On this basis, the range of ring parameters under different frequencies and spatial angles is determined through error analysis, providing theoretical support for the parameter design of magnetic induction communication. By analyzing the variation laws of three verification parameters, namely the amplitude of wave impedance, the phase difference between the spatial electric field and magnetic field, and the spatial variation rate of the magnetic induction receiving intensity factor, with distance, the physical characteristics at the magnetic induction boundary are deeply revealed. The research finds that at the magnetic induction boundary, the magnetic field energy is dominant, the electromagnetic field distribution has not yet reached a stable state, and the physical property of rapid attenuation of the magnetic induction receiving energy is presented. Simulation results show that the proposed magnetic induction boundary formula can accurately describe the magnetic induction limit distance and ensure the effective performance of magnetic induction communication within the boundary range.

Major depressive disorder is a prevalent psychological disorder, and screening is currently based on depression diagnostic scales and physician interviews. Based on artificial intelligence technology, computer−aided depression detection is an emerging approach to depression screening. Aiming at the current status and shortcomings of traditional measurement tools, this paper reviews current computer−aided depression detection methods, discusses the current research status of depression detection datasets and depression detection methods based on multimodal data such as facial pictures, speech, and text, and summarizes and outlooks the advantages and challenges of computer−aided depression detection. Computer−aided depression detection can provide a relatively simple and standardized screening method with the potential to synergize with the widely used scale screening and physician diagnosis, but still faces the challenges of the insufficient interpretation of model parameters and features, the Chinese dataset to be expanded, and the small sample size of the existing dataset. In the future, researchers need to further improve the sample size and model accuracy of depression detection datasets, conduct theoretical and experimental analyses of feature extraction and model construction, and promote the clinical application of computer−aided depression detection.

The accurate and efficient extraction of building from remote sensing images is fundamental for applications such as fine urban management, high−precision mapping, and land resource investigation. It is essential to investigate how to leverage image features for intelligent interpretation. This study introduces a global self−attention network with edge−enhancement (E−GSANet) for remote sensing building extraction. The network integrate the edge enhancement module into the encoder backbone, providing the network with a priori knowledge about boundaries, and then establish long−distance dependency relationships between features using the global self−attention feature expression module, enabling the fusion of salient features with edge−enhanced features. A stepwise up−sampling decoding module is designed to fusing the shallow features with rich spatial detail information and the deep features with high−order semantic information to obtain accurate extraction results of buildings. The comparison experiments between E−GSANet and the current mainstream methods is conducted based on two open−source remote sensing building datasets. The quantitative analysis and qualitative demonstrations prove that E−GSANet achieves optimal results across all evaluation metrics, yielding more complete building extractions, precise edges, and higher accuracy. Additionally, network structure ablation experiments and analysis demonstrate the effectiveness of each module.

With the intensification of global population aging, smart walkers, as rehabilitative assistive devices that enhance the quality of life and independence for individuals with gait disorders, have garnered widespread attention regarding the development of their human−machine interaction (HMI) strategies. This paper surveys the research progress on HMI technologies for smart walkers: Firstly, the application background and market demand for smart walkers are introduced. Subsequently, HMI technologies based on motion signals, visual signals, and other interaction modalities (such as physiological electrical signals and virtual reality) are explored. Despite significant advances in related research on smart walker HMI, challenges persist in areas such as motion intention recognition, navigation and obstacle avoidance, and interaction safety. Future research should focus on optimizing motion intention recognition algorithms, developing navigation strategies for complex environments, strengthening research on interaction safety, and promoting the development of multi−modal HMI technologies. These endeavors aim to achieve more accurate intention recognition and a higher level of synergy within the human−machine−environment systems, ultimately assisting individuals with gait disorders in achieving their independent living aspirations.

Breast cancer is one of the most prevalent malignant tumors in women worldwide. Nearly half of patients exhibit a human epidermal growth factor receptor 2 (HER2)-low expression profile, rendering them ineligible for traditional HER2-directed therapy as they fall under the category of HER2-negative breast cancer. The distinct clinicopathologic features of HER2-low breast cancer as compared with HER2-negative breast cancer have intrigued a growing number of researchers. This review provides a comprehensive overview of advancements in the definition of HER2-low breast cancer, its characteristics in epidemiology, clinicopathology and prognosis, its molecular genetic features, heterogeneity, and relevant agents. Emerging evidence suggests that HER2-low breast cancer may represent a distinct clinicopathological subtype. However, current data remain insufficient to establish it as an independent prognostic factor. Furthermore, the lack of standardized criteria for interpreting HER2 immunohistochemistry (IHC) 0 and 1+scores has led to inconsistencies in classification. The absence of refined subclassification within the HER2 IHC 0 population may further obscure potential benefits from targeted treatments for this group. This review emphasizes the need for future studies to refine and standardize the diagnostic procedures and treatment regimens for HER2-low breast cancer. Establishing consistent diagnostic protocols and treatment strategies may enhance clinical management for these patients.

In 5G/6G radio-over-fiber (RoF) networks, remote radio units (RUs) require multi-beamforming functionality to ensure reliable access for ubiquitous mobile terminals. To meet this requirement, multi-core fibers (MCFs) have emerged as a promising solution for RoF links due to their advantages of supporting multiple channels and maintaining excellent inter-channel delay consistency. Here, we proposes a remote optical true-time-delay multi-beamforming architecture based on MCFs, which is suitable for 5G RoF networks. The architecture utilizes MCFs as the link, while deploying chirped fiber Bragg gratings with equal dispersion spacing to provide equal-space time delays at the centralized unit. By independently tuning the wavelengths of each optical carriers, the corresponding beam direction can be continuously adjusted, enabling centralized multi-beam control. To validate the feasibility of this architecture, we use a 2-km 7-core fiber as the RoF link for experiment and build a 2×2 remote beamforming system. Experimental results demonstrate that by tuning the wavelength of each optical carrier, independent control of each beam direction can be achieved. Compared to single-mode fibers, MCF reduces inter-channel delay jitter by more than an order of magnitude, with a maximum delay jitter of 1.7 ps, ensuring long-term stability of the beam direction. Furthermore, the inter-core crosstalk of MCF has a negligible impact on both the preset delays and the signal-to-noise ratio of broadband wireless signals. This architecture provides a feasible and stable solution for realizing remote beamforming, offering significant application value for 5G/6G mobile access networks.

Searchable encryption is a key technology for enabling data encrypted search, and it has significant application value for cloud storage. However, existing schemes generally adopt a single-user model and are vulnerable to insider keyword guessing attacks, which exposes cloud data to the risk of privacy leakage. Therefore, there is an urgent need to design a searchable encryption scheme that support multi-user models and provide higher security to meet the privacy-preserving of cloud data. In response, this paper proposes an authenticated ciphertext retrieval scheme for cloud data access control. In terms of access control, the scheme embeds attributes into users' secret key to generate search trapdoor and incorporates access policies into the keyword ciphertext. The matching of attributes and access policies is achieved through threshold secret sharing techniques, thus establishing a fine-grained retrieval permission control mechanism. To enhance security, the secret key of the data owner is embedded into the keyword ciphertext to provide ciphertext authentication, effectively preventing insider keyword guessing attacks. Performance analysis shows that the trapdoor generation algorithm in our proposed scheme are computationally efficient, while the user secret key has relatively low storage overhead, making our scheme suitable for cloud storage applications.

Soil antibiotic pollution has become a significant issue in China's agricultural environment, impacting ecosystem health and food safety. This study, based on the Geographically Optimal Similarity (GOS) model, systematically predicts the spatial distribution of soil antibiotics in Chinese farmlands and explores the impact of agricultural management measures on soil antibiotic residues through scenario analysis. The results indicate that the average antibiotic concentration in China's farmland soil is approximately 16.9 ng/g, with distinct regional variations. The eastern and central regions generally show higher soil antibiotic concentrations, which are primarily associated with irrigation and fertilization management intensity. When the amount of organic fertilizer is reduced by more than 40%, the soil antibiotic concentration significantly decreases; conversely, when the irrigation area increases by more than 60%, the concentration of soil antibiotics increases rapidly. The study suggests that optimizing agricultural management measures, particularly reasonable fertilization and irrigation in high-risk areas, can effectively control soil antibiotic pollution and promote agricultural environmental protection and sustainable agricultural development.

近年来，通过材料创新、结构优化和管理系统升级，动力电池在能量密度、充电速度和使用寿命等方面取得了显著进展。然而，随着新技术的应用，动力电池的安全设计问题依然不容忽视，尤其在快充、底部磕碰和热事件等使用场景中仍存在较高的安全风险。系统梳理了动力电池技术革新的主要方向，包括结构设计创新及化学体系升级，并分析了由此引入的安全性能提升策略。从测试评价的角度，深入分析了电池快充、底部冲击及热扩散行为等前沿安全问题，并提出了针对电池全生命周期的安全评价技术。通过综述实验与仿真测试技术，构建了多层级的动力电池安全评价框架，为动力电池技术的进一步发展和测试评价的优化提供了理论支持与实践指导。

选取人胃癌BGC-823细胞系为模型，通过多组学研究方法，探讨丁酸盐和B族维生素联用对癌细胞生长的影响及可能的作用机制。通过cck-8方法观察维生素B2与丁酸盐联合作用对胃癌细胞BGC-823的抑制作用，Chou-Talay中效分析法评价两组分的联合效应，并以光学显微镜观察细胞形态变化；通过流式细胞术观察维生素B2和丁酸盐联用对BGC-823细胞凋亡及细胞周期的影响；以Western blot方法测定抗癌基因蛋白PTEN及其下游PI3K蛋白表达情况；通过对荷瘤裸鼠模型的作用，观察二者联合体内抑瘤效果；通过胃癌组和健康人群对照组血清中多代谢物的代谢组学分析建立多元判别模型，初步筛选差异代谢物，并在这些代谢物基础上进行通路聚集分析和代谢途径分析。结果表明，cck-8细胞活力和Chou-Talay中效分析结果证明维生素B2可协同增强丁酸盐对BGC-823细胞的杀伤作用，且作用过程中影响了细胞分化，导致细胞形态变大，分裂受阻；流式细胞术显示维生素B2与丁酸盐联用可显著增加胃癌细胞的凋亡比例，将细胞周期阻滞在G2/M期；Western blot结果显示维生素B2和丁酸盐联用可使PTEN蛋白水平显著上升，其下游PI3K表达水平则显著下降；体内荷瘤裸鼠模型显示维生素B2和丁酸盐联用可显著抑制肿瘤的生长；人群代谢组学分析建立多元OPLS-DA模型，并提示丁酸和维生素B2均为胃癌患者体内显著差异代谢物；通路富集分析和途径分析显示脂肪酸和B族维生素的代谢通路在胃癌患者体内具有重要的作用。维生素B2和丁酸盐联用可协同增强对胃癌的抑制作用。

入侵检测作为软件定义网络(software defined networks，SDN)架构的关键安全防护手段，能有效保障SDN安全稳定运行。通过汇总基于机器学习、基于深度学习、基于强化学习和基于信息熵的入侵检测方法，总结并分析SDN环境中仍存在的问题总结并分析了SDN环境中仍存在的问题：单控制器易受网络威胁、缺乏可扩展性、缺乏缓解和预防的方法、缺乏低速率DDoS的攻击检测、缺乏用于训练的SDN特定数据集、应用层的防御方法较少，并指出了未来的研究方向。

基于静止轨道海洋彩色成像仪（GOCI）卫星遥感数据对黄河口及其毗邻海域表层水体叶绿素a（Chl-a）浓度进行反演，并根据2019年的实测黄河口表层水体Chl-a浓度对GOCI数据进行了验证，结果表明，反演得到的Chl-a浓度准确。同时，通过对比分析调水调沙年份和未调水调沙年份的黄河口及其毗邻海域表层水体Chl-a浓度的时空变化特征后发现：（1）在调水调沙年份，从调水调沙前到调水期结束Chl-a浓度呈持续降低的状态；从调沙期开始，由于短时间内携带大量营养物质的泥沙大量汇入研究海域，导致该海域内Chl-a浓度大幅增加；（2）Chl-a在调水调沙年份的平均浓度整体低于未调水调沙年份的平均浓度，从侧面表明调水调沙使得大量泥沙入海导致水体浊度增加，降低了光合作用效率，这在一定程度上抑制了浮游生物的爆发，进而防止了赤潮等生态问题的发生。研究结果表明，实施调水调沙工程在缓解黄河口上游水库及河床泥沙淤积的同时，并未对黄河口湿地及其毗邻海域的生态环境产生破坏；构建的监测模型对未来评估及动态监测调水调沙工程对黄河口湿地及其毗邻海域生态环境影响具有重要意义。

交叉学科是新科学的生长点，是科学发展的必然趋势。基于位置服务的重要应用——兴趣点推荐，作为计算机学科与地理信息学科相交叉的研究课题，对于推动2个学科在时空数据分析等相关领域的交叉融合研究具有重要作用。分析了兴趣点推荐的影响因素即地理位置、时间因素、社交关系和流行度，重点阐述了基于图神经网络的兴趣点推荐方法，包括基于图注意力网络、图卷积网络、图自编码器的推荐，并对其特点进行对比；讨论了在兴趣点推荐中存在的一些关键挑战，如数据稀疏性、冷启动问题和用户动态偏好问题，并针对各项挑战提出相应的解决思路，提出了结合多种影响因素的推荐，跨领域推荐以及动态偏好推荐的发展趋势。

智能建造数字化平台作为建筑行业数字化转型的关键支撑，在建筑信息模型（BIM）、云计算和大数据等核心技术的驱动下实现了突破性发展。从相关研究总体特征、相关研究现状和平台案例分析3个维度对2010—2023年的相关研究进行分析。研究发现，BIM技术通过三维可视化建模和全生命周期数据管理构建了平台的基础框架；云计算技术凭借其弹性计算资源和分布式处理能力为平台提供了强大的技术支撑；大数据技术则通过实时数据分析和智能决策优化了工程管理流程。以北京城市副中心项目为例，该平台创新性地采用“环状”数字集成交付模式，整合区块链数据确权技术，实现了政府投资项目设计-施工-运维全过程的数字化协同管理，有效解决了传统工程管理中的“三超”问题。然而当前研究仍存在明显局限，比如应用场景主要集中于单体项目层面，缺乏行业级系统整合；技术标准体系尚未统一，制约了数据互联互通；人工智能等新兴技术的融合应用研究相对不足。未来发展方向应着重于构建产业级集成平台，推动规模化应用；建立统一的技术标准和数据接口规范；深化人工智能、数字孪生等创新技术与传统建造技术的融合应用；探索数据资产交易机制，促进数据要素市场化。

域名系统作为互联网核心基础资源，在日常网络访问中发挥着不可或缺的关键作用。根区的稳定性直接决定全球域名体系的运行安全。基于互联网数字分配机构发布的根区顶级域名数据，通过分析顶级域权威服务器的解析服务地址，揭示了当前顶级域名注册与托管的集中化特征。研究发现，1458个活跃顶级域名的解析服务高度依赖少数顶级域名托管机构，其中Afilias和Identity Digital Limited分别托管了32.58%和32.17%的顶级域名，前3大机构合计占比超90%。通过基尼系数量化评估，顶级域名托管资源的中心化程度达到0.83，显著高于注册管理机构（0.44），表明托管环节存在严重的寡头垄断风险。研究指出，过度集中的托管架构可能引发单点故障隐患（如区域性网络攻击波及多顶级域名解析服务），并削弱域名生态多样性。为此，建议推动顶级域名注册管理机构自建权威服务器、优化托管机构准入机制，并探索基于区块链的分散化域名解析技术，以提升根区抗风险能力与长期稳定性。

利用小动物超声建立稳定、可靠的无创评价肺动脉高压大鼠右心结构和功能评价的方法，将雄性SD大鼠随机分为对照组和模型组，模型组采用颈部皮下注射血管内皮细胞生长因子受体抑制剂SU5416（20 mg/kg体重）后置于10%氧浓度低氧箱中饲养3周后转至常氧环境饲养3周，构建重度肺动脉高压大鼠模型。对照组置于正常环境进行饲养。造模6周后对动物进行麻醉后，进行心脏超声测量及右心导管术进行血流动力学相关指标采集，并对大鼠心脏超声指标进行均一性和重测信度分析，与大鼠右心室收缩压进行相关性分析，评价其对肺动脉高压大鼠的诊断价值。结果发现通过小动物心脏超声建立的肺动脉高压大鼠右心结构与功能评价指标具有较好的均一性及重测信度。舒张末期右室游离壁厚度、收缩末期右室游离壁厚度、三尖瓣环收缩期位移、肺动脉血流加速时间/射血时间和肺动脉血流加速时间与右心导管测得右心室收缩压具有较好的相关性（P < 0.001）及较高的诊断价值。因此，小动物超声是一种无创、稳定、重复性良好的肺动脉高压大鼠右心结构和功能评价方法，可用于该模型疾病严重程度的评估。

基于2002—2022年Web of Science论文数据，界定基础研究创新韧性概念，运用核心变量法测度中国和美国基础研究创新韧性，并借助Dagum基尼系数和核密度分析探讨其特征、地区差异及动态演变。研究发现：美国基础研究创新韧性呈下降趋势，而中国相对平稳且总体更强；在金融危机和新冠肺炎疫情冲击下，美国各区域韧性分布较均衡，中国区域差异较大，且随时间推移，美国抵抗力和恢复力均下降，中国抵抗力上升但恢复力下降，中国抵抗力与恢复力显著正相关，美国则无明显相关性；美国主要区域论文发表较均衡，中国则呈极化趋势，且在不同冲击下，两国主要区域韧性表现各异；Dagum基尼系数显示，中美基础研究创新韧性综合差距扩大，抵抗期主要源于国内差异，恢复期则由中美间差异主导；Kernel密度估计表明，美国韧性降低且州间差异扩大，抵抗期呈现两极分化趋势，恢复期趋于一致但集中在低水平；中国韧性高于美国，省间差异减小，低水平地区比例降低。

具有超高分辨率的扫描探针显微镜（Scanning Probe Microscope, SPM）是纳米科技、量子现象揭示的重要手段，其应用范围广泛，涵盖了材料科学、物理、化学、生物学等多个领域。SPM重要性体现在其超高分辨率和多功能性，能提供原子级分辨率。综述了近年SPM领域与极端条件集成的进展、发展趋势及存在问题。此外，SPM还能进行力学、磁学、电学等多种基本测量，已成为研究微观世界的多功能平台。当前，在材料性能不断提升但似乎难以取得颠覆性进展的情况下，利用极端环境发现新材料和新物理来实现突破已成为前进方向。为此，SPM与极端外部物理场的结合，如超高磁场、超低温和高真空环境的融合已成为重要发展方向。

为系统评估武器装备体系使命任务需求满足度，提出一种基于质量功能展开（quality functiondeployment，QFD）和基于证据推理的信度规则库推理（belief rule base inference methodology usingevidential reasoning，RIMER）的组合方法。首先，通过QFD方法将高层次的使命任务需求逐层分解到具体装备能力需求，构建不同层级需求的相关映射关系；其次，采用决策实验室分析法（decisionmaking trial and evaluation laboratory，DEMATEL）确定各子任务需求重要度，运用犹豫模糊数评分改进传统关系矩阵确定方式；在底层装备能力评估中，基于RIMER方法构建信度规则库，通过证据推理提高复杂和不确定性指标数据的评估准确度；最后，基于QFD映射关系逆向推导，实现从底层能力到使命任务的需求满足度评估。以城市智能有人/无人协同夺控任务为例，验证了方法的有效性。研究结果表明，该方法在处理复杂和不确定性数据方面具有优势，可以提高评估的准确性和可靠性。

在全球经济快速发展背景下，粤港澳大湾区的建设发展是中国推动战略发展的关键创新引擎，也是中国实现社会经济可持续发展的首要选择。总结研究粤港澳大湾区沿海城市群山海田城的空间发展特征背后的影响因素。基于粤港澳大湾区沿海城市40年间7个时期的遥感监测数据，对山、田、城等自然要素进行了解译提取及整理计算。研究发现：粤港澳大湾区沿海城市山海田城的空间形态呈现“变化东强西弱，城市连绵发展”的演变特征；空间结构在多方面呈现不同特征，空间规模呈现“城镇规模分段式扩张，山海田规模逐年变化”的特征；空间地类的转化每10年间呈现不同趋势的变化特征。

以西江23-1油田H2A砂岩储层为背景，分析了不同细粒含量砂岩孔隙结构。开展了不同细粒含量砂岩应力敏感性实验，采用应力敏感性系数、渗透率损害率和渗透率曲率评价了砂岩应力敏感性，结果表明，随着细粒含量的增加，砂岩渗透率应力敏感性增强。综合分析了砂岩孔隙结构及渗透率应力敏感性，表明孔隙体积分布是影响砂岩应力敏感性的主要因素，即小孔频率越高，渗透率应力敏感性越强。提出了砂岩孔隙率-应力预测模型，预测有效应力增加过程中砂岩孔隙演化。

轨道交通运输规划及控制的数学本质为离散约束下的优化问题，具有NP（non-deterministic polynomial）难的高计算复杂度问题。作为未来计算能力跨越式发展的重要探索方向，量子计算有望为解决现有大规模路网中的复杂问题提供潜在解决方案。介绍了量子计算的基础概念及算法，分析了其在轨道交通领域的潜在应用场景，包括行车调度组织优化、列车自动运行控制和列车群组的动态编解。通过实验验证了量子粒子群算法在解决复杂优化问题中的优势，表明量子计算在处理大规模离散约束优化问题时具有显著的计算效率提升。然而，量子计算技术在轨道交通中的应用仍面临量子比特退相干、硬件集成和结果安全性等挑战。总结了量子计算在轨道交通中的应用前景及可能面临的问题，强调了量子计算技术在推动轨道交通智能化发展中的潜力与挑战。

计算全息技术是动态三维显示的理想解决方案，具有广阔的应用前景。在当前的技术条件下，计算全息技术面临的最大挑战是全息算法难以同时兼顾计算的精度和速度。为此，提出了卷积误差消除的模型驱动相位型全息图生成网络，实现了高保真相位型全息图的快速生成。首先，研究了角谱法中卷积误差的产生机制，提出了无卷积误差的角谱法，开发了基于无卷积误差角谱法的迭代框架，证实了无卷积误差角谱法对于提升相位型全息图计算精度的有效性；其次，以无卷积误差角谱法作为编码器构建了卷积误差消除的模型驱动相位型全息图生成网络，将相位型全息图的计算时间减小了3个数量级。通过网络生成的相位型全息图，抑制了全息光学重建中的散斑噪声，提高了重建结果的细节质量，平均峰值信噪比高达20.38 dB。伴随着深度梯度显著性和通道效率一致性的继续提升，该网络有望广泛应用在虚拟现实、元宇宙和三维视频通讯等领域。

为解决目前海上高温高压低渗气田在清喷排液过程中存在的高密度钻完井液固相含量高，易沉降导致清喷困难；钻井船综合日费高，排液周期长，造成清喷成本高；海上空间受限，清喷设备空间布置及安全性要求高等一系列难题，通过科技攻关，自主形成了一套适用于海上高温高压低渗气田清喷排液关键技术体系。该技术研发了超细重晶石加重的钻开液+封隔液+无固相完井液的工作液体系，设计了一套井口小平台清井放喷地面流程，提出了一套海上高温高压井井口平台清喷安全控制方法，合理优化了井口平台的清喷管线流程及装置布局，形成了清喷流程多节点监测控制系统，设置了地面流程应急关断屏障，在地层高压气体泄漏发生时能够及时关断流体来源，保障清喷安全。该技术在南海莺歌海盆地高温高压低渗气田中成功实现了现场应用。

锂产业链关键产品贸易安全是事关国家战略性矿产资源安全的重要问题。基于2013—2022年中国锂产业链关键产品的对外贸易数据，一方面，从贸易规模、贸易伙伴2个维度，全面探讨中国锂产业链关键产品的贸易格局及演化规律；另一方面，构建进出口结构安全性评价指数，评估中国锂产业链关键产品贸易结构的安全水平，以期为中国制定和调整锂资源安全战略提供参考。研究结果表明：2013年以来，中国锂产业链关键产品的贸易规模不断扩大，其中，碳酸锂以进口为主，而氢氧化锂、锂离子电池均以出口为主；中国锂产业链关键产品的出口贸易伙伴均多于进口，其中，中国碳酸锂和氢氧化锂的贸易伙伴相对集中，而锂离子电池的贸易伙伴数量稳中有升，贸易核心地位逐渐凸显；中国碳酸锂的进出口结构模式在强进口和弱进口之间波动，现阶段体现为强进口，氢氧化锂的进出口结构始终维持在强出口模式，锂离子电池进出口结构模式经历了“相对平衡—弱出口—强出口”的转变；中国碳酸锂进口和氢氧化锂出口的结构安全性均处在一种不安全的状态，而中国锂离子电池的贸易结构长期处于安全状态。

基于空冷型燃料电池原位测试实验平台，对优化的阳极氢气侧双向供气的气路结构和阴极空气侧强化中间区域散热的差异化风速风扇配置方案开展实验分析。结果发现，优化脉排间隔时间可以减小运行过程中的电压衰减，并提出了基于不同负载层次的氢气控制策略。阴极中间区域风速提升可以改善单电池内温度和电流密度的分布均匀性，提出了分段式风扇转速控制策略，在不同负载电流区间内对最佳运行温度和阴极入口风速进行了设计。在此基础上，搭建包含33节电池的空冷型燃料电池系统，验证了控制策略的有效性。

新污染物共暴露会对污水处理厂活性污泥微生物群落结构产生重要影响，但卤代咔唑和纳米材料复合污染对活性污泥系统生态效应的研究尚未开展。选择低（0.05 mg/L）和高（5 mg/L）2种浓度的三氯咔唑（3-chlorocarbazole，3-CCZ）与1 mg/L纳米四氧化三铁（Fe₃O₄ nanoparticles，Fe₃O₄NPs）作为胁迫条件，探究复合污染对活性污泥体系细菌群落的短期影响以及胁迫后体系对低温的响应。结果表明，3-CCZ与Fe₃O₄ NPs复合污染会对微生物群落丰富度产生抑制作用，污染物的添加以及低温环境都会改变微生物群落结构。基于全尺度分类发现各类群微生物中均存在抗性与恢复特性不同的特定菌属，稀有属对污染物的添加更敏感，丰富属中的Saccharimonadales和TM7a一直是优势菌属。利用网络关系图探究6个生物类群间的相互作用，条件稀有菌属（conditionally rare taxa，CRT）与条件丰富菌属（conditionally abundant taxa，CAT）分别是胁迫阶段和低温阶段下的核心分类单元，复合污染胁迫下稀有属与丰富属之间存在着复杂的共现关系。功能微生物中反硝化细菌（denitrifying bacteria，DNB）的相对丰度受到复合污染胁迫影响最显著，尤其低温环境下DNB相对丰度出现大幅上升。高浓度复合污染会引起硝化基因的富集，反硝化功能基因面对复合污染呈现出不同的变化。研究可为污水处理系统卤代咔唑与纳米材料复合污染的风险评价提供新的思路与理论依据。

针对浅水航道水域较浅且环境条件复杂恶劣的现象，以浮筒宽高比过大的新型半潜平台为研究对象，基于时域分析理论和浅水波浪理论，通过数值模型对比分析不同有义波高、谱峰周期、浪向角、拖航速度和拖揽长度对浅水海域半潜平台拖航运动响应和拖揽张力的影响。结果表明：波浪高度越高，平台垂荡、缆绳张力越大；谱峰周期越长，纵摇、缆绳张力越小；浪向角对横纵摇影响较大，极值呈凹凸型分布；不同工况下的新型半潜平台横纵摇均≤5°，满足拖航规范要求；缆绳张力受航速和波高的影响较大，当波高达到5 m，拖揽张力超过安全范围，可以通过降低拖航速度、缩短缆绳长度等措施规避风险。

图像压缩感知是一种能够在低采样率下实现高效信号采样与重构的技术，但在实现高质量图像重构时，面临局部与全局特征难以有效融合的问题。为此，提出一种结合Transformer与卷积神经网络（convolutional neural networks，CNN）优点的图像压缩感知重构框架（transformer-CNN mixture transformer，TCMformer）。该框架充分利用CNN的局部建模能力和Transformer的全局特征捕捉能力；设计了一种特征融合模块（TCM Block），有效桥接局部与全局特征，从而提升特征表示效率；同时，为降低模型复杂度并控制计算成本，框架采用基于窗口的Transformer结构，通过分块实现高效的全局建模。此外，引入渐进式重建策略，利用多尺度特征图逐步优化重建质量。实验结果表明，TCMformer在峰值信噪比、结构相似性和视觉效果上相较于主流的压缩感知重构算法表现更优，为实现高质量的图像重建提供了一种有效的解决方案。

Seismic water temperature observation instruments (abbreviated as thermometer) plays a key role in short and imminent earthquake prediction. However, for a long time, there has been a lack of field detection method with quantitative traceability, which is, mobile calibration method. In this paper, two kinds of mobile calibration methods of thermometers, namely in situ comparison method and constant temperature chamber comparison method, are proposed, and the influence of these methods on the continuity of observation data is preliminarily verified by field experiments. The results show that under the conditions of the experimental well, the standard deviations of the observation data of the measured instrument before and after mobile calibration are close, with the maximum difference of -0.0002℃, the maximum difference of the variation amplitude of the indication value of -0.0053℃, and the calculated results of interference degree (ID) of all<50%. The influence of mobile calibration method on instrument data continuity is in the acceptable range. Although the method has some limitations in the scope of application, it is still a potential scheme to realize the traceability of station operating thermometers.

Gray water is a kind of domestic sewage with light extent of pollution, and rural grey water accounts for 60%~70% of rural domestic sewage. If it is properly used, the situation of rural water shortage will be greatly improved. algal-bacterial symbiosis system is a potentially efficient and economical treatment for rural gray water. In this study, the feasibility of practical application of this method was demonstrated by studying the biological activity and pollutant removal effect of single or symbiotic Chlorella-activated sludge system under free and immobilized states, and the variation rule of microbial growth under different conditions was analyzed. The experimental results show that, in the condition of no aeration, the photosynthesis of Chlorella can supply a certain amount of Dissolved Oxygen (DO) to the system, and the immobilized Chlorella-activated sludge system showed the best removal effect on the pollutants in simulated rural gray water (the removal rate of chemical oxygen demand 87.5%±2.89%; Ammonia nitrogen 65.34%±2.0%; Total Phosphorus (TP) 60.4%±4.29%; Linear Alklybezene Sulfonates (LAS) 91.07%±2.04%). The treatment effect was better than that of single fixed system and free system of bacteria or algae, and the biomass increased significantly compared with other experimental groups (p<0.05). The symbiotic effect between activated sludge and Chlorella was obvious. In the semi-continuous operation test, the immobilized algal-bacterial symbiosis system showed better pollutant removal stability. The unit cell chlorophyll content and Mixed Liquor Suspended Solids (MLSS) accumulation in the immobilized state were 1.6 and 1.5 times of that in the suspended state, respectively, and the biological growth was good in the immobilized state. In the immobilized state, the phylum flora retained more of the same category before and after the experiment, while the changing trend of Actinobacteria in the two experimental groups was different. During the experiment, LAS showed obvious enrichment of bacteria and strong inhibition of nitrobacteria activity, indicating that nitrogen removal in water was closely related to the biological activities of Chlorella.

In order to improve the service performance of oil pipeline for oilfield development, this paper takes Ф73×5.51 mm J55 oil pipeline as the substrate, and uses self-spreading high-temperature synthesis technology (SHS) to add 4wt%Nb₂O₅, 8wt% ZrO 2, 13wt% CrO₃, and 5wt% SiO₂ to the basic aluminum thermal system of Al+Fe₂O₃ (B component) as a new type of SHS material system, and prepares a new type of ceramic lined wear-resistant coated oil pipe (A component). Using SEM analysis, EDS analysis, XRD analysis, FTIR analysis and other analytical means to test the acid corrosion resistance, bonding strength, bending strength and crush strength and other properties, comparative analysis results show that: the corrosion resistance, bending strength, bonding strength and crushing strength of the ceramic lining layer prepared by component A are better than that of component B. The flow corrosion weight loss rate of component A coating and component B coating at 1200 h is 0.42% and 0.54%, respectively, which is better than that of J55 tubing substrate, and the bending strength of ceramic lined tubing prepared by component A is 15.9% higher than that of component B. Based on the observation of the slip position, the transition temperature between the strength of partial metallurgical bonding at the Fe-Fe interface and the mechanical bonding at the FeAl₂O₃ interface was determined to be about 200℃, with the mechanical bonding strength being greater than the metallurgical bonding strength at low temperatures, and the mechanical bonding strength being less than the metallurgical bonding strength at high temperatures. Both the bond strength and the crush strength properties of the ceramic lining layer showed a significant decreasing trend with increasing temperature.

In response to clear understanding of reservoir rock-forming and pore evolution characteristics，this study takes the tight sandstone reservoir of the Lower Cretaceous Yingcheng Formation and Shahezi Formation in the Longfengshan Sub-sag of the Changling Fault depression in the Songliao Basin as the research object, and the pore evolution process of the main rockforming phases is quantitatively recovered on the basis of reservoir characteristics and rock-forming phase delineation. The conclusions are as follows. The main rock type of the target formation is rock chip sandstone with low structural maturity and poor sorting and rounding; the typical type of reservoir diagenesis is compaction and cementation, the diagenetic stage is intermediate diagenetic stage B, and the diagenetic coeval sequence is: early compaction - chlorite cementation - turbidite cementation - early calcareous cementation - phase I dissolution - siliceous cementation - phase II dissolution - late calcareous cementation; the typical diagenetic phases include chlorite cementation diagenetic phase, turbidite cementation diagenetic phase and acidic dissolution diagenetic phase, and each diagenetic phase has undergone the process of compaction and cementation to reduce pore size, and two stages of dissolution to increase pore size; the pore evolution characteristics and the understanding of tight gas reservoir formation period are integrated, and two reservoir evolution models are established, namely the first tight and then reservoir formation model and the tight unresolved reservoir model.

Information processing of ancient Chinese seldom uses unearthed documents as corpus to carry out relevant research. The number of Liye Qin bamboo manuscripts reached ten times that of all the Qin slips unearthed before, which can fill many gaps in the historical records of the Qin Dynasty. In this paper, we used them as experimental corpus and explored the automatic sentence segmentation and word segmentation of unearthed documents based on the CRF model. We combined the actual characteristics of the corpus and set up different feature templates to verify the generalization ability of model sequence labeling on different tasks. We set up a joint approach to sentence segmentation and word segmentation as comparative experiment to select a better performance processing plan. At the same time, a comparative experiment was designed between deep learning methods and pretrained models. The results proved that the overall performance of the joint approach in each task was improved and that the F1-score of automatic sentence segmentation and word segmentation reached 75.79% and 94.44%, respectively. Since it's faster and takes less time, this approach is more suitable for the Liye Qin bamboo slips. The research results can serve the proofreading work of the last three volumes of Liye Qin bamboo slips and the in-depth processing and construction of the corpus.

The effect of post-heat treatment on the microstructure and microhardness of laser cladded high Co-Ni secondary hardening steel coating was investigated. The microstructure of the coating was analyzed using a scanning electron microscopy (SEM) equipped with an energy dispersive spectrometer (EDS), and the microhardness was measured with a Vickers indenter. Decomposition of the retained austenite in the coating occurred during the post-heat treatment. As the temperature increased from 200 ℃ to 600 ℃, the quantity of the retained austenite at the boundaries decreased significantly, while that of the needleshaped M₃C cementite and M₂C carbides increased. The M₂C carbides evidently coarsened when the temperature was 550 ℃ and 600 ℃. The microhardness of high Co-Ni steel coating increased as the temperature of heat treatment increased from 200 ℃ to 400 ℃ because the fine-scale M₂C carbides were coherent with the matrix and increased distinctly in this temperature range. While the microhardness decreased sharply when the temperature was improved from 500 ℃ to 600 ℃ due to both the incoherency of the coarsened M₂C carbides and the recovery of dislocations in the carbon-supersaturated matrix.

The high temperature of self-propagation is about 1000℃, which exceeds the tolerable temperature of 502~638℃ of the bearing temperature of aluminum alloy tube and causes thermal damage to the aluminum alloy substrate, while Si, Ti and Mo, as the additive phases, can reduce the temperature of self-propagation aluminum thermal reaction by activating the crystal lattice and wetting the interface. In this paper, low-temperature self-propagation technology is used to prepare wear-resistant coatings on the inner wall of aluminum alloy tubes with small inner diameters, and investigated the effect of added phases (Si, Ti, and Mo) on reducing the temperature of the self-propagating aluminothermal reaction system. Orthogonal experiments were used to determine process parameters such as the optimal ratio of aluminothermic agents, additive phases, and cooling pressure, and the cooling mechanism of additive phases was studied. The results show that the traditional aluminothermic reaction process releases a large amount of heat. Adding Si and Ti phases can lower the reaction temperature, slow down the severity of the reaction, and even inhibit the progress of the reaction. Adding an appropriate amount of Mo will cause the aluminothermic reaction to occur again. The Exothermic reaction temperature of the self-propagating reaction system with all added phases is about 100℃ lower than that of the aluminothermic reaction, and the weight loss at high temperature is at least 1.33%. The optimal ratio is: the cooling gas pressure is 0.2 MPa, the mass fraction of the added phase is 6%, and the mass fraction of thermite is 94%.

This study observed the effects of intravenous patient-controlled analgesia with hydromorphone combined with butorphanol on postoperative pain neurotransmitters, inflammatory factors, and oxidative stress levels, aiming to evaluate the efficacy, safety, and application value of comprehensive analgesic methods. Postoperative orthopedic patients were randomly divided into a study group (with a combination of hydromorphone and butorphanol) and a control group (with butorphanol alone for pain relief). Serum substance P (SP), prostaglandin E2 (PGE2) and beta endorphin (β-EP) were used as pain neurotransmitter markers, interleukin-6 (IL-6) and tumor necrosis factor alpha (TNF-α) were used as inflammatory cytokine markers, cortisol (Cor) and norepinephrine (NE) were used as oxidative stress markers, and Visual Analog Scale (VAS) was used to measure clinical analgesic efficacy. The results showed that the combined group had a lower post-operative VAS, which further reduced the serum SP, PGE2, NE, Cor levels, increased the β-endorphin levels, and reduced the incidence of adverse reactions. The combination of butorphanol and low-dose hydromorphone can further reduce the levels of pain transmitters, alleviate oxidative stress reactions, increase pain thresholds, reduce patient pain levels, and improve anesthesia safety.

Entity alignment, as an important research direction in knowledge graph research, aims to connect different entities pointing to the same real-world object in different knowledge graphs, and thus to achieve the expansion of knowledge graphs. At present, there are two mainstream research approaches in this field. One is to analyze the structural characteristics of knowledge graphs, and the other is to analyze the element information (such as entity name, relation name, attribute name) of knowledge graphs. In this article, a novel entity alignment model EAFF (Entity Alignment based on Feature Fusion) is proposed to analyze the features of knowledge graphs from the perspectives of graph structure and element information. First, a graph neural networkbased entity alignment algorithm was designed to obtain aligned entity pairs based on graph structures. Then, an entity alignment algorithm based on element information was designed to obtain aligned entity pairs based on element information. Finally, using feature transformation and sorting algorithms, two sets of aligned entity pairs are sorted to obtain aligned entities in the knowledge graph. In the experiment, EAFF achieved relatively good results, surpassing current mainstream algorithms.

Marine high-tech enterprises are the core carrier and main organization of the marine high-tech industry. The identification of their location characteristics and influencing factors is of great significance for optimizing the spatial layout of regional industries and accelerating the high-quality development of the marine economy. Based on the data analysis of China's marine high-tech enterprises, in this paper the nearest neighbor index, kernel density estimation, hot spot analysis and other methods are adopted to explore the spatial distribution, sub-industry development characteristics, agglomeration characteristics and hotspot division of marine high-tech enterprises. Meanwhile, geographic detectors are used to explore influencing factors and interactions of their spatial distribution. The results show that: (1) The temporal and spatial evolution of China's marine hightech enterprises in three different development stages has obvious regional differences and industry orientation, that is, the overall pattern is "dense in the north and sparse in the south", "high in the east and low in the west", and the distribution is concentrated in the Yangtze River Delta, Beijing-Tianjin-Hebei, Shandong Peninsula, Pearl River Delta and other regions; industries such as seawater utilization, marine biomedicine, and marine equipment manufacturing are the main industries. (2) Both the overall and sub-industries of China's marine high-tech enterprises have the characteristics of spatial agglomeration. All the three development stages have passed the significance test, and the degree of spatial agglomeration is becoming more and more obvious; the spatial distribution density has formed 4 high-density core areas and 3 sub-density core areas and several small core areas; hot spots are mainly concentrated in the Bohai Rim, Yangtze River Delta, and Pearl River Delta. (3) The main core influencing factors are industrial base, economic base, regional characteristics and regional policies. The influence of the interaction of different factors on the spatial distribution of enterprises is greater than the influence of each factor alone.

After China has made remarkable achievements in tackling poverty, overcoming relative poverty has become a difficult problem in the new period. Overcoming energy poverty will be an important task for China to promote energy transformation and achieve common prosperity. Based on the panel data of 30 provinces (autonomous regions and municipalities directly under the Central Government) in China from 2005 to 2021, this paper constructs a comprehensive evaluation index system to measure the energy poverty index, and studies the regional differences and distribution dynamics of energy poverty in China by using the two-dimensional decomposition method of Gini coefficient, kernel density estimation and Markov chain. Come to a conclusion:China's energy poverty has eased as a whole, with obvious regional differences; The overall difference is convergent, with the most obvious difference between regions and the largest difference in energy consumption structure. Energy poverty is persistent and characterized by "club convergence". The above conclusions will provide inspiration and decisionmaking reference for understanding the current situation of energy poverty in China and solving it.