高精度且鲁棒的预测模型建立高度依赖于样本数据的大小、多样性和分布;日益积累的文献数据为获得大量的多样性样本数据提供了可能。以SLM-ed IN 718合金的相对密度(RD)为研究对象,针对从文献中挖掘的激光功率P、扫描速度V、扫描间距HS...高精度且鲁棒的预测模型建立高度依赖于样本数据的大小、多样性和分布;日益积累的文献数据为获得大量的多样性样本数据提供了可能。以SLM-ed IN 718合金的相对密度(RD)为研究对象,针对从文献中挖掘的激光功率P、扫描速度V、扫描间距HS和铺粉厚度LT与RD样本数据存在缺失参数和分布不均问题,采用最大期望化(EM)算法对缺失参数进行补齐;采用带有梯度惩罚的WGAN算法(WGAN-GP)对数据稀疏的低RD区间生成虚拟样本数据。然后,分别基于补齐文献数据和补充虚拟数据,采用常青藤算法优化的随机森林(IVYA-RF)构建了RD预测模型,并对模型预测精度进行了定量评估和实验验证。结果表明:基于补充虚拟数据集构建的IVYA-RF模型II比基于补齐文献数据集构建的IVYA-RF模型I具有更好的预测精度,其原因主要来源于在低RD区间生成虚拟数据后,使建模样本数据的分布均匀性得到改善,这也是数据增强与机器学习相结合的意义所在。对新实验数据的验证取得了满意的预测精度,其中,IVYA-RF模型I验证结果的统计学参数R2(决定系数)、RMSE(均方根误差)、MAE(平均绝对误差)和MRE(平均相对误差)分别达到了0.891、1.352%、0.915%和0.98%;IVYA-RF模型II验证结果的R2增大至0.956,RMSE、MAE和MRE分别减小至0.833%、0.687%和0.71%,同样显示出后者比前者具有更好的预测精度。实验验证结果表明,所构建的RD预测模型具有较好的鲁棒性,从而具备了较好的工程应用价值。展开更多
DNA methylation plays important roles in regulating gene expression during development.However,little is known about the influence of DNA methylation on secondary metabolism during leaf development in the tea plant(Ca...DNA methylation plays important roles in regulating gene expression during development.However,little is known about the influence of DNA methylation on secondary metabolism during leaf development in the tea plant(Camellia sinensis).In this study,we combined the methylome,transcriptome,and metabolome to investigate the dynamic changes in DNA methylation and its potential regulatory roles in secondary metabolite biosynthesis.In this study,the level of genomic DNA methylation increased as leaf development progressed from tender to old leaf.It additionally exhibited a similar distribution across the genomic background at the two distinct developmental stages studied.Notably,integrated analysis of transcriptomic and methylomic data showed that DNA hypermethylation primarily occurred in genes of the phenylpropanoid,flavonoid,and terpenoid biosynthesis pathways.The effect of methylation on transcription of these secondary metabolite biosynthesis genes was dependent on the location of methylation(i.e.,in the promoter,gene or intergenic regions)and the sequence context(i.e.,CpG,CHG,or CHH).Changes in the content of catechins and terpenoids were consistent with the changes in gene transcription and the methylation state of structural genes,such as serine carboxypeptidase-like acyltransferases 1A(SCPL1A),leucoanthocyanidin reductase(LAR),and nerolidol synthase(NES).Our study provides valuable information for dissecting the effects of DNA methylation on regulation of genes involved in secondary metabolism during tea leaf development.展开更多
Hard carbon (HC) has been considered as promising anode material for sodium-ion batteries (SIBs).The optimization of hard carbon’s microstructure and solid electrolyte interface (SEI) property are demonstrated effect...Hard carbon (HC) has been considered as promising anode material for sodium-ion batteries (SIBs).The optimization of hard carbon’s microstructure and solid electrolyte interface (SEI) property are demonstrated effective in enhancing the Na+storage capability,however,a one-step regulation strategy to achieve simultaneous multi-scale structures optimization is highly desirable.Herein,we have systematically investigated the effects of boron doping on hard carbon’s microstructure and interface chemistry.A variety of structure characterizations show that appropriate amount of boron doping can increase the size of closed pores via rearrangement of carbon layers with improved graphitization degree,which provides more Na+storage sites.In-situ Fourier transform infrared spectroscopy/electrochemical impedance spectroscopy (FTIR/EIS) and X-ray photoelectron spectroscopy (XPS) analysis demonstrate the presence of more BC3and less B–C–O structures that result in enhanced ion diffusion kinetics and the formation of inorganic rich and robust SEI,which leads to facilitated charge transfer and excellent rate performance.As a result,the hard carbon anode with optimized boron doping content exhibits enhanced rate and cycling performance.In general,this work unravels the critical role of boron doping in optimizing the pore structure,interface chemistry and diffusion kinetics of hard carbon,which enables rational design of sodium-ion battery anode with enhanced Na+storage performance.展开更多
为了解决传统接触式位移传感器在实际桥梁位移检测中存在的安装复杂、测量范围受限等问题,将SURF(Speeded Up Robust Features)特征点检测与KLT(Kanade-Lucas-Tomasi)特征追踪算法相结合,提出了一种基于计算机视觉的特征光流桥梁位移测...为了解决传统接触式位移传感器在实际桥梁位移检测中存在的安装复杂、测量范围受限等问题,将SURF(Speeded Up Robust Features)特征点检测与KLT(Kanade-Lucas-Tomasi)特征追踪算法相结合,提出了一种基于计算机视觉的特征光流桥梁位移测量方法。利用消费级相机对实验室斜拉桥模型进行测试,并与传统激光测量方法进行了对比,验证了本文所提方法的有效性。同时,从有无靶标、相机距离、光线强弱3个方面探讨了该算法的鲁棒性,最后将本文所提方法与主流视觉方法进行了对比分析。结果表明,特征光流法能够精确测量桥梁的振动位移,最大误差仅为0.73 mm,为桥梁在复杂环境下的非接触式实时位移监测提供了新的思路。展开更多
Due to the high capacity and moderate volume expansion of silicon protoxide SiO_(x)(160%)compared with that of Si(300%),reducing silicon dioxide SiO_(2)into SiO_(x)while maintaining its special nano-morphology makes i...Due to the high capacity and moderate volume expansion of silicon protoxide SiO_(x)(160%)compared with that of Si(300%),reducing silicon dioxide SiO_(2)into SiO_(x)while maintaining its special nano-morphology makes it attractive as an anode of Li-ion batteries.Herein,through a one-pot facile high-temperature annealing route,using SBA15 as the silicon source,and embedding tin dioxide SnO_(2)particles into carbon coated SiO_(x),the mesoporous SiO_(x)-SnO_(2)@C rod composite was prepared and tested as the anode material.The results revealed that the SnO_(2)particles were distributed uniformly in the wall,which could further improve their volume energy densities.The coated carbon plays a role in maintaining structural integrality during lithiation,and the rich mesopores structure can release the expanded volume and enhance Li-ion transfer.At 0.1 A·g^(-1),the gravimetric and volumetric capacities of the composite were as high as 1271 mAh·g^(-1)and 1573 mAh·cm^(-3),respectively.After 200 cycles,the 95%capacity could be retained compared with that upon the 2nd cycle at 0.5 A·g^(-1).And the rod morphology was well kept,except that the diameter of the rod was 3 times larger than its original size after the cell was discharged into 0.01 V.展开更多
文摘高精度且鲁棒的预测模型建立高度依赖于样本数据的大小、多样性和分布;日益积累的文献数据为获得大量的多样性样本数据提供了可能。以SLM-ed IN 718合金的相对密度(RD)为研究对象,针对从文献中挖掘的激光功率P、扫描速度V、扫描间距HS和铺粉厚度LT与RD样本数据存在缺失参数和分布不均问题,采用最大期望化(EM)算法对缺失参数进行补齐;采用带有梯度惩罚的WGAN算法(WGAN-GP)对数据稀疏的低RD区间生成虚拟样本数据。然后,分别基于补齐文献数据和补充虚拟数据,采用常青藤算法优化的随机森林(IVYA-RF)构建了RD预测模型,并对模型预测精度进行了定量评估和实验验证。结果表明:基于补充虚拟数据集构建的IVYA-RF模型II比基于补齐文献数据集构建的IVYA-RF模型I具有更好的预测精度,其原因主要来源于在低RD区间生成虚拟数据后,使建模样本数据的分布均匀性得到改善,这也是数据增强与机器学习相结合的意义所在。对新实验数据的验证取得了满意的预测精度,其中,IVYA-RF模型I验证结果的统计学参数R2(决定系数)、RMSE(均方根误差)、MAE(平均绝对误差)和MRE(平均相对误差)分别达到了0.891、1.352%、0.915%和0.98%;IVYA-RF模型II验证结果的R2增大至0.956,RMSE、MAE和MRE分别减小至0.833%、0.687%和0.71%,同样显示出后者比前者具有更好的预测精度。实验验证结果表明,所构建的RD预测模型具有较好的鲁棒性,从而具备了较好的工程应用价值。
基金supported by the Natural Science Foundation of Guangdong Province(Grant Nos.2022A1515111141 and 2023A1515010786)。
文摘DNA methylation plays important roles in regulating gene expression during development.However,little is known about the influence of DNA methylation on secondary metabolism during leaf development in the tea plant(Camellia sinensis).In this study,we combined the methylome,transcriptome,and metabolome to investigate the dynamic changes in DNA methylation and its potential regulatory roles in secondary metabolite biosynthesis.In this study,the level of genomic DNA methylation increased as leaf development progressed from tender to old leaf.It additionally exhibited a similar distribution across the genomic background at the two distinct developmental stages studied.Notably,integrated analysis of transcriptomic and methylomic data showed that DNA hypermethylation primarily occurred in genes of the phenylpropanoid,flavonoid,and terpenoid biosynthesis pathways.The effect of methylation on transcription of these secondary metabolite biosynthesis genes was dependent on the location of methylation(i.e.,in the promoter,gene or intergenic regions)and the sequence context(i.e.,CpG,CHG,or CHH).Changes in the content of catechins and terpenoids were consistent with the changes in gene transcription and the methylation state of structural genes,such as serine carboxypeptidase-like acyltransferases 1A(SCPL1A),leucoanthocyanidin reductase(LAR),and nerolidol synthase(NES).Our study provides valuable information for dissecting the effects of DNA methylation on regulation of genes involved in secondary metabolism during tea leaf development.
基金National Key Research and Development Program of China (2022YFE0206300)National Natural Science Foundation of China (U21A2081,22075074, 22209047)+3 种基金Guangdong Basic and Applied Basic Research Foundation (2024A1515011620)Hunan Provincial Natural Science Foundation of China (2024JJ5068)Foundation of Yuelushan Center for Industrial Innovation (2023YCII0119)Student Innovation Training Program (S202410532594,S202410532357)。
文摘Hard carbon (HC) has been considered as promising anode material for sodium-ion batteries (SIBs).The optimization of hard carbon’s microstructure and solid electrolyte interface (SEI) property are demonstrated effective in enhancing the Na+storage capability,however,a one-step regulation strategy to achieve simultaneous multi-scale structures optimization is highly desirable.Herein,we have systematically investigated the effects of boron doping on hard carbon’s microstructure and interface chemistry.A variety of structure characterizations show that appropriate amount of boron doping can increase the size of closed pores via rearrangement of carbon layers with improved graphitization degree,which provides more Na+storage sites.In-situ Fourier transform infrared spectroscopy/electrochemical impedance spectroscopy (FTIR/EIS) and X-ray photoelectron spectroscopy (XPS) analysis demonstrate the presence of more BC3and less B–C–O structures that result in enhanced ion diffusion kinetics and the formation of inorganic rich and robust SEI,which leads to facilitated charge transfer and excellent rate performance.As a result,the hard carbon anode with optimized boron doping content exhibits enhanced rate and cycling performance.In general,this work unravels the critical role of boron doping in optimizing the pore structure,interface chemistry and diffusion kinetics of hard carbon,which enables rational design of sodium-ion battery anode with enhanced Na+storage performance.
文摘为了解决传统接触式位移传感器在实际桥梁位移检测中存在的安装复杂、测量范围受限等问题,将SURF(Speeded Up Robust Features)特征点检测与KLT(Kanade-Lucas-Tomasi)特征追踪算法相结合,提出了一种基于计算机视觉的特征光流桥梁位移测量方法。利用消费级相机对实验室斜拉桥模型进行测试,并与传统激光测量方法进行了对比,验证了本文所提方法的有效性。同时,从有无靶标、相机距离、光线强弱3个方面探讨了该算法的鲁棒性,最后将本文所提方法与主流视觉方法进行了对比分析。结果表明,特征光流法能够精确测量桥梁的振动位移,最大误差仅为0.73 mm,为桥梁在复杂环境下的非接触式实时位移监测提供了新的思路。
文摘Due to the high capacity and moderate volume expansion of silicon protoxide SiO_(x)(160%)compared with that of Si(300%),reducing silicon dioxide SiO_(2)into SiO_(x)while maintaining its special nano-morphology makes it attractive as an anode of Li-ion batteries.Herein,through a one-pot facile high-temperature annealing route,using SBA15 as the silicon source,and embedding tin dioxide SnO_(2)particles into carbon coated SiO_(x),the mesoporous SiO_(x)-SnO_(2)@C rod composite was prepared and tested as the anode material.The results revealed that the SnO_(2)particles were distributed uniformly in the wall,which could further improve their volume energy densities.The coated carbon plays a role in maintaining structural integrality during lithiation,and the rich mesopores structure can release the expanded volume and enhance Li-ion transfer.At 0.1 A·g^(-1),the gravimetric and volumetric capacities of the composite were as high as 1271 mAh·g^(-1)and 1573 mAh·cm^(-3),respectively.After 200 cycles,the 95%capacity could be retained compared with that upon the 2nd cycle at 0.5 A·g^(-1).And the rod morphology was well kept,except that the diameter of the rod was 3 times larger than its original size after the cell was discharged into 0.01 V.
