A bidirectional hybrid DWDM-PON employing optical injection locking technique and data comparators

A bidirectional hybrid dense-wavelength-division-multiplexing (DWDM)-passive optical network (PON) employing optical injection locking technique on vertical-cavity surface-emitting lasers (VCSELs) and data comparator, as well as optical injection locking technique on distributed feedback laser diodes (DFB LDs) and amplitude/phase compensator is proposed and demonstrated. Improved performances of bit error rate (BER) and carrier-to-noise ratio/composite second order/composite triple beat (CNR/CSO/CTB) were observed in our proposed bidirectional hybrid DWDM-PON. Bidirectional transmission in bidirectional hybrid DWDM-PON architecture is a very attractive option, the capacity of the optical network can be expanded relatively easily by bidirectional transmission technology.     

海洋气象观测技术研发进展

 海洋覆盖地球70%以上的面积,由于没有陆地上布设气象观测站网的便利性,使得海上气象观测资料严重不足,难以支撑对海上和来自海上高影响天气过程的深入研究以及预报准确性的提高。为了加强海上气象观测能力的建设,本文对气象海洋卫星、大气探测飞机、岛屿气象站、海上浮漂和岸基雷达等观测平台及技术进行了综述,并对未来基于无人驾驶平台的智能海洋气象探测技术进行展望。     

我国枸杞产业发展现状及提升路径

枸杞产业作为西北地区传统的农业优势特色产业,在优化地方产业结构、缓解区域"三农问题",精准服务国家脱贫攻坚战略方面具有重要意义。本文在挖掘枸杞种植、生产数据信息的基础上,系统梳理了宁夏、新疆、甘肃、青海和内蒙古等枸杞主产区的对外贸易、产品分类和品牌体系,分析了近年来我国枸杞产业发展现状与问题,探讨了我国枸杞产业在规模、产业链、品牌和产品体系等方面的优化路线和解决途径。构建枸杞产业合理高效的发展路径,形成促进产业发展的建议和意见,能够为我国枸杞产业可持续发展提供决策支撑。     

DPSM technique for ultrasonic field modelling near fluid-solid interface

Distributed point source method (DPSM) is gradually gaining popularity in the field of non-destructive evaluation (NDE). DPSM is a semi-analytical technique that can be used to calculate the ultrasonic fields produced by transducers of finite dimension placed in homogeneous or non-homogeneous media. This technique has been already used to model ultrasonic fields in homogeneous and multi-layered fluid structures. In this paper the method is extended to model the ultrasonic fields generated in both fluid and solid media near a fluid-solid interface when the transducer is placed in the fluid half-space near the interface. Most results in this paper are generated by the newly developed DPSM technique that requires matrix inversion. This technique is identified as the matrix inversion based DPSM technique. Some of these results are compared with the results produced by the Rayleigh-Sommerfield integral based DPSM technique. Theory behind both matrix inversion based and Rayleigh-Sommerfield integral based DPSM techniques is presented in this paper. The matrix inversion based DPSM technique is found to be very efficient for computing the ultrasonic field in non-homogeneous materials. One objective of this study is to model ultrasonic fields in both solids and fluids generated by the leaky Rayleigh wave when finite size transducers are inclined at Rayleigh critical angles. This phenomenon has been correctly modelled by the technique. It should be mentioned here that techniques based on paraxial assumptions fail to model the critical reflection phenomenon. Other advantages of the DPSM technique compared to the currently available techniques for transducer radiation modelling are discussed in the paper under Introduction.     

Influence of thermophoretic diffusion of nanoparticles with Joule heating in flow of Maxwell nanofluid

The aspects of activation energy in magnetized Maxwell nanofluid flow with Brownian movement and thermophoretic diffusion have been elaborated here. Furthermore, Joule heating, variable conductivity and chemical reaction are scrutinized. The Buongiorno nanofluid thought is ratify to incorporate the importance of thermophoretic and Brownian diffusion. The attained ODEs have been solved via homotopic algorithm. The performance of operational variables is inspected. The Maxwell temperature field for Eckert number and variable conductivity factor have similar trend. The fluid concentration exaggerates for activation energy and decelerates for Brownian motion parameter. Furthermore, the brilliant outcomes attained and associated with possible existing prose accurately.     

An Electrocardiogram Signal Classification Algorithm Based on Improved Deep Residual Shrinkage Networks北大核心CSCD

心电信号分类是医疗保健领域的重要研究内容.针对大多数方法不能很好地降低样本数量少的类别漏诊率,以及降低预处理操作的复杂性问题,提出了一种基于改进深度残差收缩网络(IDRSN)的心电信号分类算法(即DRSL算法).首先,使用合成少数类过采样技术(SMOTE)扩充数量少的类别样本,从而解决了类不平衡问题;其次,利用改进深度残差收缩网络提取空间特征,其残差模块可以避免网络层加深造成的过拟合,压缩激励和软阈值化子网络可以提取重要局部特征并自动去除噪声;然后,通过长短期记忆网络(LSTM)提取时间特征;最后,利用全连接网络输出分类结果.在MIT-BIH心律失常数据集上的实验结果表明,该算法的分类性能优于IDRSN、DRSN、GAN+2DCNN、CNN+LSTM_ATTENTION、SE-CNN-LSTM分类算法.     

结合2D-COS和光谱融合技术的小麦淀粉回生特性定量表征

回生是淀粉加工、运输和储藏过程中的重要理化性质,快速检测淀粉回生程度对淀粉制品的品质和保质期有重要意义。为了探究二维相关光谱法(2D-COS)优选回生淀粉特征变量的可行性,研究结合2D-COS和光谱融合技术对小麦淀粉的回生特性进行定量表征。首先,将不同回生时间的小麦淀粉测定结晶度和回生度,从淀粉体系中晶体含量和对淀粉酶水解抗性的角度表征淀粉回生特性。然后,分别采集样品的近红外和中红外光谱数据,对采集的原始光谱进行Savitzky-Golay平滑和标准正态变量变换预处理后,结合偏最小二乘法分别基于近红外光谱、中红外光谱和融合光谱构建全光谱的预测模型。在此基础上,以回生天数为外部扰动,分别选取回生0, 1, 2, 3, 5, 7, 10, 14, 21和35 d的10条淀粉光谱进行2D-COS分析。通过分析同步谱和自相关谱,辨识了近红外13个和中红外11个与回生特性有关的特征波长。最后,基于这些特征波长进一步建立回生度和结晶度的预测模型。结果表明,全光谱模型结果中,光谱融合后的模型预测效果较好,结晶度模型的相对分析误差(RPD)值由1.203 4和2.069 0提高至3.980 9,回生度...     

基于连续光谱法联合测定NH3-N、COD方法研究

针对地表水中Ⅰ—Ⅴ类水域、地下水、工业废水对氨氮及化学需氧量准确快速联合测定的需求,融合光谱分析法中的连续光谱法与顺序注射技术（SIA）,基于地表水国家检测标准,以氨态氮（NH₃-N）、化学需氧量（COD）原位水质参数为检测对象,设计了一种小型微量原位水质NH₃-N、COD高效快速检测仪。该系统主要依据自主设计的基于紫外灯照射消解与加热密闭消解方法同时进行的消解池结构以及基于光谱扫描设计的的检测池结构,达到快速消解稳定检测的目的。同时系统基于分光光度法优化了检测流程,测定开始COD消解的同时,待检测池中的NH₃-N指数显色后的配位化合物进行光谱扫描测定,消解后,进行COD的测定,整个检测过程相比国标检测法缩短至少60 min,可以实现25min内自动完成NH₃-N及COD的测定,大大节约时间成本。绘制经光谱扫描显色反应后配位化合物的吸光度与连续波长曲线可得：NH₃-N、COD分别于690和445 nm处具有明显吸收峰,经读取峰的吸光度值,采用最小二乘法分别建立NH₃-N、COD回归建模,拟合回归方程并计算相关系数,绘制相应参数的吸光度-浓度工作曲线。实验结果表明：在0~2 mg·L-1浓度范围内,NH₃-N标准工作曲线相关系数r≥0.998 7,且浓度与吸光度成正相关。重复性相对标准偏差为1.36%~1.68%,加标回收率为97%~102%;在0~50 mg·L-1浓度范围内,COD标准工作曲线相关系数r≥0.997 8,浓度与吸光度呈负相关。重复性相对标准偏差为2.14%~2.48%,加标回收率为97.6%~102.95%。系统的测定结果准确、直线性与稳定性良好,具有较高的可行性与可靠性。基于SIA与连续光谱法联合测定NH₃-N、COD的方法研究,为拓宽光谱法在水质快速检测领域的应用、提高检测效率等方面的研究具有重大价值。     

有源头靠中一种分布式虚拟传声器优化方法

有源降噪头靠系统中,远程虚拟传声器技术能够解决控制点处与误差传声器处降噪量不匹配的问题。在实际应用中,多通道虚拟传声器技术存在收敛速度慢和运算复杂度高等问题。针对这个问题,本文通过重新设计远程虚拟传声技术的离线优化过程,提出一种分布式远程虚拟传声器技术优化方法。该方法将虚拟次级通路矩阵作对角化限制,同时对观测传递函数矩阵进行联合寻优,以实现一种分布式的更新算法。有源降噪头靠实验结果表明,所提算法能够有效降低远程虚拟传声器技术算法的运算复杂度,并且提升了算法的收敛速度。     

基于网络技术的综合导航系统研究

通过对局域网的分析,提出了基于1553B数据总线技术综合导航系统设计方案,并对其可行性进行了充分论证,这种方案在很大程度上保证了系统的可靠性和实时性。