[Objectives]To optimize the water extraction process of Chinese Herbal Compound Man Gan Ning and establish a method for its extraction and content determination.[Methods]The L9(34)orthogonal test was carried out,the c...[Objectives]To optimize the water extraction process of Chinese Herbal Compound Man Gan Ning and establish a method for its extraction and content determination.[Methods]The L9(34)orthogonal test was carried out,the comprehensive scores of ginsenoside Rb1 content,notoginsenoside R1 content,tanshinone I content and saikosaponin A content in the compound extract were used as evaluation indicators to explore the effects of water addition,the water addition,decoction time and decoction times on the water extraction process,and verification tests were carried out.[Results]The optimized water extraction process was as follows:adding 9 times of water in the first extraction,and extracting for 90 min;adding 7 times of water in the second extraction,and extracting for 60 min;adding 7 times of water in the third extraction,and extracting for 60 min.[Conclusions]The optimized water extraction process is stable and feasible,and can be used for the extraction of various herbs in Compound Man Gan Ning.It can also provide experimental basis for the formulation development of Compound Man Gan Ning sustained release tablet.展开更多
近年来随着氮化镓器件制造工艺的迅速发展,氮化镓高电子迁移率晶体管(GaN HEMT)已经开始应用在电力电子领域。GaN HEMT以其低寄生参数、无反向恢复损耗、高开通速度等特点,可降低开关管的开关损耗。本文以600V GaN HEMT为研究对象,研究...近年来随着氮化镓器件制造工艺的迅速发展,氮化镓高电子迁移率晶体管(GaN HEMT)已经开始应用在电力电子领域。GaN HEMT以其低寄生参数、无反向恢复损耗、高开通速度等特点,可降低开关管的开关损耗。本文以600V GaN HEMT为研究对象,研究其共源共栅(Cascode)结构引起的开关动态过程及其寄生参数的影响。建立了600V GaN HEMT等效模型并详细推导了其在单相逆变器中开关管正向导通、正向关断、反向续流导通和反向续流关断四种情况的动态过程。GaN HEMT的等效电路考虑了对开关过程及开关损耗有重要影响的寄生电感和寄生电容。理论、仿真及实验证明了Cascode GaN HEMT器件中寄生电感L_(int1)、L_(int3)和L_S直接影响开关管的动态过程进而影响开关管的开关损耗。展开更多
为了解决点云处理过程中空间信息损失的问题,同时在融合过程中最大程度地提取可见光图像的纹理信息,本文提出了一种基于特征切片的激光点云与可见光图像融合车辆检测方法(FVOIRGAN-Detection)。在CrossGAN-Detection方法中加入了FVOI(Fr...为了解决点云处理过程中空间信息损失的问题,同时在融合过程中最大程度地提取可见光图像的纹理信息,本文提出了一种基于特征切片的激光点云与可见光图像融合车辆检测方法(FVOIRGAN-Detection)。在CrossGAN-Detection方法中加入了FVOI(Front View Based on Original Information)的点云处理思路,将点云投影到前视角度并把原始点云信息的各个维度切片为特征通道,在不降低网络性能的情况下显著提高点云信息利用效率。并且引入了相对概率的思想,采用鉴别器鉴别图像的相对真实概率替代绝对真实概率,使得融合图像提取的纹理信息更加接近真实的纹理信息。在KITTI数据集上进行检测性能实验验证结果表明,本文方法在容易、中等和困难三个类别中的AP指标分别达到97.67%、87.86%和79.03%。在光线受限的场景下,AP指标达到了88.49%,与CrossGAN-Detection方法相比提高了2.37%,提高了目标检测的性能。展开更多
基金Supported by Project of National Natural Science Foundation of China(81360524)Project for Improving Basic Research Ability of Middle Aged and Young Teachers in Colleges and Universities of Guangxi in 2019(2019KY0341)+2 种基金National Traditional Chinese Medicine Characteristic Technology Heritage Talent Training Program(20184828005)Youth Foundation Project of Guangxi University of Chinese Medicine(2019QN036)Program of Scientific Research Third-level Laboratory"Chinese(Zhuang)Medicine Chemical and Quality Analysis Laboratory"of State Administration of Traditional Chinese Medicine(Guo Zhong Yi Yao Fa 2009[21]).
文摘[Objectives]To optimize the water extraction process of Chinese Herbal Compound Man Gan Ning and establish a method for its extraction and content determination.[Methods]The L9(34)orthogonal test was carried out,the comprehensive scores of ginsenoside Rb1 content,notoginsenoside R1 content,tanshinone I content and saikosaponin A content in the compound extract were used as evaluation indicators to explore the effects of water addition,the water addition,decoction time and decoction times on the water extraction process,and verification tests were carried out.[Results]The optimized water extraction process was as follows:adding 9 times of water in the first extraction,and extracting for 90 min;adding 7 times of water in the second extraction,and extracting for 60 min;adding 7 times of water in the third extraction,and extracting for 60 min.[Conclusions]The optimized water extraction process is stable and feasible,and can be used for the extraction of various herbs in Compound Man Gan Ning.It can also provide experimental basis for the formulation development of Compound Man Gan Ning sustained release tablet.
文摘基于生成对抗网络(generative adversarial networks,GAN),提出了面向肝脏肿瘤CT图像仿真数据集生成深度学习算法.首先,将CT图像数据文件进行格式解析,单独保存为PNG格式的图像文件;然后,将肝脏病变区域统一标注为白色,并结合肝脏CT原图组成配对图片;最后,用生成对抗网络的pix2pix架构仿真生成病变肝脏图像.为将生成图像与目标图像进行定量分析、比较,本文采用了峰值信噪比和结构相似性作为模型的评价指标.实验结果表明,本文算法所生成的肝脏肿瘤CT仿真数据集的平均峰值信噪比为64.72 d B,平均结构相似性为0.9973,证明了所生成的仿真图像数据有着非常高的真实度.
文摘近年来随着氮化镓器件制造工艺的迅速发展,氮化镓高电子迁移率晶体管(GaN HEMT)已经开始应用在电力电子领域。GaN HEMT以其低寄生参数、无反向恢复损耗、高开通速度等特点,可降低开关管的开关损耗。本文以600V GaN HEMT为研究对象,研究其共源共栅(Cascode)结构引起的开关动态过程及其寄生参数的影响。建立了600V GaN HEMT等效模型并详细推导了其在单相逆变器中开关管正向导通、正向关断、反向续流导通和反向续流关断四种情况的动态过程。GaN HEMT的等效电路考虑了对开关过程及开关损耗有重要影响的寄生电感和寄生电容。理论、仿真及实验证明了Cascode GaN HEMT器件中寄生电感L_(int1)、L_(int3)和L_S直接影响开关管的动态过程进而影响开关管的开关损耗。
文摘为了解决点云处理过程中空间信息损失的问题,同时在融合过程中最大程度地提取可见光图像的纹理信息,本文提出了一种基于特征切片的激光点云与可见光图像融合车辆检测方法(FVOIRGAN-Detection)。在CrossGAN-Detection方法中加入了FVOI(Front View Based on Original Information)的点云处理思路,将点云投影到前视角度并把原始点云信息的各个维度切片为特征通道,在不降低网络性能的情况下显著提高点云信息利用效率。并且引入了相对概率的思想,采用鉴别器鉴别图像的相对真实概率替代绝对真实概率,使得融合图像提取的纹理信息更加接近真实的纹理信息。在KITTI数据集上进行检测性能实验验证结果表明,本文方法在容易、中等和困难三个类别中的AP指标分别达到97.67%、87.86%和79.03%。在光线受限的场景下,AP指标达到了88.49%,与CrossGAN-Detection方法相比提高了2.37%,提高了目标检测的性能。
