基于改进SOA-VMD的磁声发射信号去噪算法

磁声发射（MAE）是铁磁性材料磁化过程中产生的声发射信号，在构件应力检测和微观损伤检测中有着广泛的应用。针对MAE信号非稳态、复杂性、衰减性等特点，提出海鸥算法结合变分模态分解（SOA-VMD）的去噪方法，为克服海鸥算法求解过程中易陷入局部最优解问题，利用柯西变异算子产生随机迭代过程，使改进算法即柯西变异海欧算法（CVSOA）跳出早熟收敛。采用以幅值谱熵为适应度函数，优化VMD算法中分解模态个数K和二次惩戒因子α两个参数，将含噪声的MAE信号进行VMD分解重构。经仿真信号和实际检测信号分析表明，改进后的CVSOA-VMD算法全局寻优能力和去噪性能优于传统的SOA-VMD算法，降噪后的MAE信号特征值对于不同应力下均方根、偏斜度特征值的重复性更好，可靠性更高。     

One-Dimensional Deep Convolutional Neural Network for Mineral Classification from Raman Spectroscopy

Neural Processing Letters - Raman spectroscopy is often used for the composition determination and rapid classification of materials because it can reflect the molecular information of materials....     

Fractional order battery modelling methodologies for electric vehicle applications: Recent advances and perspectives

Science China Technological Sciences - Accurate modelling of lithium ion batteries is crucial for battery management in electric vehicles. Recent studies have revealed the fractional order nature...     

共享交通的时空轨迹检索与群体发现

为解决共享交通下的共乘用户群体发现效率低、准确率不高问题，依据R-树原理建立GeoOD-Tree索引，并在此基础上提出以最大化共乘率为目标的群体发现策略。首先，对原始时空轨迹数据进行特征提取与标定处理，挖掘有效出行起讫点（OD）轨迹；其次，针对用户起讫点轨迹的特征，建立GeoOD-Tree索引进行有效的存储管理；最后，给出以最大化共乘行程为目标的群体发现模型，并运用K最近邻（KNN）查询对搜索空间剪枝压缩，提高群体发现效率。采用西安市近12000辆出租车营运轨迹数据，选取动态时间规整（DTW）等典型算法与所提算法在查询效率与准确率上进行性能对比分析。与DTW算法相比，所提算法的准确率提高了10.12%，查询效率提高了约15倍。实验结果表明提出的群体发现策略能有效提高共乘用户群体发现的准确率和效率，可有效提升共乘出行方式的出行率。     

快速切换(SMED)在空调制造厂的研究与应用

在目前市场需求呈现规模扩张和品类个性化的冲击下,制造业面对的竞争压力愈发明显.且随着新时代自动化、信息化、大数据智能化等先进制造特征的兴起,更加促使制造业向转型升级不断进军.以空调制造业的控制器版块为例,从制约此单位效率的SMT车间切换浪费为切入点,在SMED快速换模的理念基础上,通过使用IE分析工具对切换流程进行研究,提炼出一套多模组固定料站切换模型,同时结合自动化、信息化等智能制造技术引进,完成SMT车间人力投入减少35人,日均产出点数提升18.29％,设备运行率提升15.46％,综合成本降低110.6万元/年的效益.成功推广应用至电子物料拣选版块,并开发信息化看板,夯实应用效益,再次证实"多模组固定料站"快速切换模型是典型小、多、高生产模式,具有可行性的解决方案.     

Assembly of a DNA Origami Chinese Knot by Only 15% of the Staple Strands

As a giant leap in DNA self-assembly, DNA origami has exhibited an unprecedented ability to construct nanostructures with arbitrary shapes and sizes. In typical DNA origami, hundreds of short DNA staple strands fold a long, single-stranded (ss) DNA scaffold cooperatively into designed nanostructures. However, large numbers of DNA strands are expensive and would hinder applications such as pharmaceutical investigations because of the complicated components. Therefore, one challenge is how to reduce the number of staple strands needed to construct DNA origami. For a DNA origami structure, the scale-free folding pattern of the scaffold strand is determined by staple strands at the branching vertexes. Simple duplex regions help to define the size-related features of the origami geometry. In this study, we hypothesized that a scaffold strand can be correctly folded into a designed topology by using only staple strands involved in branching vertexes. After assembly, any remaining, flexible, single-stranded regions of the scaffold could be converted into rigid duplexes by DNA polymerase to achieve the designed geometric structures. To demonstrate the concept, we used only 18 staple strands (covering 15 % of the scaffold strand) to assemble a porous DNA nanostructure, which was visualized by atomic force microscopy (AFM). This study helps understanding of the role of cooperativity in origami folding, and provides a cost-effective approach for small-scale prototyping DNA origami.     

An effectiveness analysis of altmetrics indices for different levels of artificial intelligence publications

Scientometrics - Altmetrics indices are increasingly applied to measure scholarly influence in recent years because they can reflect the influence of research outputs more timely comparing with...