摘要：在药物发现的早期阶段,深度生成模型正逐渐成为分子设计的关键工具。简化分子线输入系统(simplified moleaclar input line entry system,SMILES)作为标准的化学表示法,被广泛用于模型训练与生成。然而,由于分子环系统线性表示的非唯一性和非方向性,现有的单向编码器在捕捉样本全局语义结构和生成有效分子环时存在局限性。因此,提出一种名为Chemical RWKV BERT(ChemRB)的方法,旨在从海量的未标记数据中深度挖掘双向信息。为实现这一目标,创新性地设计了2个预训练任务:环级特征预测与全局跨度闭合预测。这些预训练任务不仅为模型提供了丰富多元的上下文信息,更进一步提升了其对复杂分子结构特性的深入理解。实验结果显示:ChemRB模型不仅在性能上取得了显著提升,而且在新分子/样本的评估指标上也达到了最优的基准性能。结果充分验证了ChemRB在精准捕获分子固有结构信息方面的有效性,为其在相关领域的应用提供了坚实的实证基础。最后,通过在EGFR抑制剂上的测试和应用,进一步验证了ChemRB模型的实用性和广泛应用前景。

摘要：汞污染来源广泛、排放量大,已成为世界上最严重的十大污染源之一。在众多汞离子去除技术中,吸附法由于具有高效且经济的特点,成为了较为成熟且应用广泛的处理技术。性能优异的吸附剂是吸附法的关键,而传统的吸附材料吸附性能有限。基于此,本文介绍了近年来报道的金属有机骨架材料(MOFs)、共价有机框架材料(COFs)、共轭微孔聚合物(CMPs)、导电聚合物、二维过渡金属碳(氮)化物(MXenes)和复合材料等新型汞离子吸附材料,详细论述了这些吸附材料的结构特征及其与汞离子的吸附特性之间的关系。总结了水中汞离子吸附材料的发展前景,指出调节材料结构以及添加特定官能团是进一步研发新材料的方向,希望对新型汞离子吸附材料的设计和开发提供一定参考和启发。

摘要：Periodic composites with band gaps that prevent the propagation of elastic waves in certain frequency ranges can be used to control waves for a variety of engineering applications. Although studies on the characteristics of these materials, which are called phononic crystals (PCs), have yielded a large number of positive results in recent years, there is still a lack of effective design methods. In this work, a multi-objective optimization approach based on the band gap mechanism and an intelligent algorithm is used to design a one-dimensional (1D) slab construction of PCs. The design aims to fit pre-determined bands by arranging the available materials properly. Obtained by analyzing the wave transmission in periodic layers, the objective functions are linked to the optimization program to obtain a proper solution set. The results of the numerical simulations demonstrate that without constructing complicated structures, the design method is able to produce PCs that overcome the limitations of two-component PCs and hence can feasibly and effectively achieve the design targets. The design approach presented in this paper can be extended to two-or three-dimensional systems and has great potential for the development of sound/ultrasound isolation structures, multiple band frequency filters, and other applications.

摘要：Highly versatile machines, such as wheel loaders, forklifts, and mining haulers, are subject to many kinds of working conditions, as well as indefinite factors that lead to the complexity of the load. The load probability distribution function （PDF） of transmission gears has many distributions centers; thus, its PDF cannot be well represented by just a single-peak function. For the purpose of representing the distribution characteristics of the complicated phenomenon accurately, this paper proposes a novel method to establish a mixture model. Based on linear regression models and correlation coefficients, the proposed method can be used to automatically select the best-fitting function in the mixture model. Coefficient of determination, the mean square error, and the maximum deviation are chosen and then used as judging criteria to describe the fitting precision between the theoretical distribution and the corresponding histogram of the available load data. The applicability of this modeling method is illustrated by the field testing data of a wheel loader. Meanwhile, the load spectra based on the mixture model are compiled. The comparison results show that the mixture model is more suitable for the description of the load-distribution characteristics. The proposed research improves the flexibility and intelligence of modeling, reduces the statistical error and enhances the fitting accuracy, and the load spectra complied by this method can better reflect the actual load characteristic of the gear component.

摘要：To overcome imperfection of exhaustive based beam searching scheme in IEEE 802.15.3c and IEEE 802.11 ad and accelerate the beam training process, combined with the fast beam searching algorithm previously proposed, this paper proposed a beam codebook design scheme for phased array to not only satisfy the fast beam searching algorithm's demand, but also make good use of the advantage of the searching algorithm. The simulation results prove that the proposed scheme not only performs well on flexibility and searching time complexity, but also has high success ratio.

摘要：针对帝王蝶优化算法存在的收敛速度慢、寻优精度低和易陷入局部极值等缺陷,提出了一种多策略改进的帝王蝶优化算法。首先,采用正向正态云发生器对父代帝王蝶个体执行非线性云化操作,增加候选解数量,提高算法的局部开发能力;之后,引入基于凸透镜成像的反向学习策略,应用到当前最优个体上产生新的个体,提高算法收敛速度和寻优精度;最后,在调整算子中融入自适应策略,增加种群的多样性,避免算法陷入局部最优。通过对8个基准测试函数的寻优对比,以及Wilcoxon秩和检验结果的对比,发现改进算法具有更好的收敛性能、寻优性能和鲁棒性。与此同时,通过工程应用中压力容器设计和焊接梁设计的优化对比,进一步验证了改进算法处理实际工程问题时的优越性。

摘要：为了解决飞机公差设计计算机化的问题,探讨了飞机公差模型的建模技术;在UG(UniGraphics)平台上采用UG的二次开发技术和VC++,开发了计算机辅助飞机公差优化设计模块,实现了飞机公差的计算机辅助分析和分配功能;建立了实用、可靠的飞机公差数据库,实现了飞机公差的计算机自动查询功能.通过飞机公差分析和公差分配的具体设计实例,验证了研究成果的正确性,初步实现了飞机公差设计的计算机化,为飞机公差设计的数字化提供了技术参考.

摘要：为了准确地目视检查和解释脑电图(EEG),提出了一种用于识别EEG信号中癫痫发作信号的异常检测方法。首先,使用小波变换将EEG信号分解为近似和细节系数,并根据阈值准则剔除不显著系数,以限制小波系数的数量;其次,采用自编码器对离散小波系数进行编码;然后,对EEG信号进行分析以检测异常值,通过压缩特征集进行数据重构,利用分类器从无癫痫信号中检测癫痫发作信号;最后,使用波恩大学数据库,将所提方法与既有方法进行比较。所提方法中采用了线性和非线性机器学习分类器从EEG信号中检测癫痫发作信号。实验结果表明,该方法的准确率和特异性分别达到了99.93%和100%。因此,所提方法具有良好的检测能力和鲁棒性,可以用简单的线性分类器识别EEG信号中的癫痫发作信号,适用于时间序列信号分析,同时能够检测和判断异常,也可为癫痫的诊断、治疗和评估提供客观参考,从而减轻医生的工作量,提高治疗效率。

摘要：采用深水护岸摊铺船进行施工的模袋混凝土,其尾段在沉放过程中若不采取牵拉制动措施而在失控状态下脱离摊铺板滑入水中,易发生尾段堆叠现象,而非均匀展铺于水下斜坡上,不能起到有效的护坡作用。为解决上述问题,创新提出一种模袋混凝土尾段沉放技术,即通过研制出一种牵拉装置,使模袋混凝土尾段在受控状态下缓慢沉放入水。经水下摄像检查,模袋混凝土均匀展铺于水下斜坡之上,满足规范和设计要求,护坡效果良好。本技术原理简单,操作便捷,施工效果良好,可供类似工程参考和借鉴。

摘要：为更好地指导矿山开采大型机械设备的创新设计,归纳铲运机外观特点,提出点阵拓扑优化法和熵权法(EWM)设计、评估铲运机外观。首先,通过点阵拓扑优化法设计铲运机外观,使用层次分析法-熵权法(AHP-EWM)对评价指标赋权,搭建层次准则模型;其次,通过调研归纳出创造性指标、美学指标、功能性指标、经济性指标、技术性指标5个指标并建立评价体系;然后,运用层次分析法(AHP)对一级指标主观赋权,通过EWM分析二级指标的客观权重,依据python计算灰色关联度、多属性决策分析权重(CRITIC)、独立性权重、模糊综合评价法和逼近理想解排序法、理想点法(TOPSIS法)5种方法的权重,通过对比分析并验证AHP-EWM的可行性;最后,基于指标权重与方案得分优化备选方案设计,并使用最小支持向量机(LSSVM)来验证模型。结果表明:使用LSSVM模型进行回归分析,其决定系数达到0.93以上,均方误差在0.01以内,明显优于传统模型且精度较高,验证点阵拓扑优化法和EWM相结合的优化决策模型。