Field experiments were conducted to examine the influence factors of cultivar, nitrogen application and irrigation on grain protein content, gluten content and grain hardness in three winter wheat cultivars under fo...Field experiments were conducted to examine the influence factors of cultivar, nitrogen application and irrigation on grain protein content, gluten content and grain hardness in three winter wheat cultivars under four levels of nitrogen and irrigation treatments. Firstly, the influence of cultivars and environment factors on grain quality were studied, the effective factors were cultivars, irrigation, fertilization, etc. Secondly, total nitrogen content around winter wheat anthesis stage was proved to be significantly correlative with grain protein content, and spectral vegetation index significantly correlated to total nitrogen content around anthesis stage were the potential indicators for grain protein content. Accumulation of total nitrogen content and its transfer to grain is the physical link to produce the final grain protein, and total nitrogen content at anthesis stage was proved to be an indicator of final grain protein content. The selected normalized photochemical reflectance index (NPRI) was proved to be able to predict grain protein content on the close correlation between the ratio of total carotenoid to chlorophyll a and total nitrogen content. The method contributes towards developing optimal procedures for predicting wheat grain quality through analysis of their canopy reflected spectrum at anthesis stage. Regression equations were established to forecast grain protein and dry gluten content by total nitrogen content at anthesis stage, so it is feasible for forecasting grain quality by establishing correlation equations between biochemical constitutes and canopy reflected spectrum.展开更多
针对中长期电力负荷序列噪声含量高、难以直接提取序列周期规律从而影响预测精度的问题,提出了一种基于完全自适应噪声集合经验模态分解(complete ensemble empirical mode decomposition with adaptive noise,CEEMDAN)和奇异谱分析(sin...针对中长期电力负荷序列噪声含量高、难以直接提取序列周期规律从而影响预测精度的问题,提出了一种基于完全自适应噪声集合经验模态分解(complete ensemble empirical mode decomposition with adaptive noise,CEEMDAN)和奇异谱分析(singular spectrum analysis,SSA)双重分解的双向长短时记忆网络(bidirectional long and short time memory,BiLSTM)预测模型。首先,采用CEEMDAN对历史负荷进行分解,以得到若干个周期规律更为清晰的子序列;再利用多尺度熵(multiscale entropy,MSE)计算所有子序列的复杂程度,根据不同时间尺度上的样本熵值将相似的子序列重构聚合;然后,利用SSA去噪的功能,对高度复杂的新序列进行二次分解,去除序列中的噪声并提取更为主要的规律,从而进一步提高中长序列预测精度;再将得到的最终一组子序列输入BiLSTM进行预测;最后,考虑到天气、节假日等外部因素对电力负荷的影响,提出了一种误差修正技术。选取了巴拿马某地区的用电负荷进行实验,实验结果表明,经过双重分解可以将均方根误差降低87.4%;预测未来一年的负荷序列时,采用的BiLSTM模型将拟合系数最高提高2.5%;所提出的误差修正技术可将均方根误差降低9.7%。展开更多
目的应用Spectrum/EPP模型估计新疆兵团成人艾滋病(acquired immune deficiency syndrome,AIDS)疫情及发展趋势预测,为AIDS防治策略制定提供有效参考数据和科学基础。方法收集整理新疆兵团1996—2022年的AIDS相关历史数据,利用Spectrum...目的应用Spectrum/EPP模型估计新疆兵团成人艾滋病(acquired immune deficiency syndrome,AIDS)疫情及发展趋势预测,为AIDS防治策略制定提供有效参考数据和科学基础。方法收集整理新疆兵团1996—2022年的AIDS相关历史数据,利用Spectrum模型将全人群分成8类亚人群,结合新疆兵团人口学数据、AIDS流行参数和监测数据、抗病毒治疗及预防母婴传播数据对新疆兵团AIDS疫情进行估计和预测。结果截至2022年底,新疆兵团累计存活≥15岁人类免疫缺陷病毒(human immunodeficiency virus,HIV)感染者/AIDS病例数共1547例,各类人群感染HIV情况,其中吸毒人群37例、暗娼人群26例、嫖客人群114例、男同性恋人群190例、献血人群25例、感染者配偶/性伴人群86例、剩余男730例、剩余女323例和母婴传播16例。同EPP-Spectrum模型估计结果比较发现,累计存活病例比例高达91.43%,经Spectrum/EPP模型估计,新疆兵团AIDS疫情仍呈逐年上升趋势,估计2022年新疆兵团≥15岁人群HIV感染率为7/10万,传播途径主要以异性传播为主。经过R-Spline模型分别拟合各类亚人群HIV阳性率,暗娼人群HIV阳性率逐渐上升与中位数数值趋势偏离,男男性行为者(men who have sex with men,MSM)人群HIV阳性率呈缓慢上升趋势;注射吸毒者人群HIV阳性率整体处于先升后降的趋势;嫖客人群HIV阳性率一直处于平稳状态;有偿献血人群HIV阳性率为下降趋势;配偶/性伴人群HIV阳性率高于其他亚人群,阳性率在快速上升之后转至平滑曲线。结论经过Spectrum/EPP模型预测发现,新疆兵团仍属于AIDS低流行地区,嫖客、MSM和配偶/性伴人群将是新疆兵团今后AIDS防控工作的重点关注人群。针对不同人群及时做好干预工作以及宣传教育,及时发现更多的HIV感染者,从而降低新疆兵团AIDS发病率。展开更多
基金financially supported by the Special Funds for Major State Basic Research Project,China(G20000779)the China National High Tech R&D Program(2002AA243011,2003AA209010,H020821020130).
文摘Field experiments were conducted to examine the influence factors of cultivar, nitrogen application and irrigation on grain protein content, gluten content and grain hardness in three winter wheat cultivars under four levels of nitrogen and irrigation treatments. Firstly, the influence of cultivars and environment factors on grain quality were studied, the effective factors were cultivars, irrigation, fertilization, etc. Secondly, total nitrogen content around winter wheat anthesis stage was proved to be significantly correlative with grain protein content, and spectral vegetation index significantly correlated to total nitrogen content around anthesis stage were the potential indicators for grain protein content. Accumulation of total nitrogen content and its transfer to grain is the physical link to produce the final grain protein, and total nitrogen content at anthesis stage was proved to be an indicator of final grain protein content. The selected normalized photochemical reflectance index (NPRI) was proved to be able to predict grain protein content on the close correlation between the ratio of total carotenoid to chlorophyll a and total nitrogen content. The method contributes towards developing optimal procedures for predicting wheat grain quality through analysis of their canopy reflected spectrum at anthesis stage. Regression equations were established to forecast grain protein and dry gluten content by total nitrogen content at anthesis stage, so it is feasible for forecasting grain quality by establishing correlation equations between biochemical constitutes and canopy reflected spectrum.
文摘针对中长期电力负荷序列噪声含量高、难以直接提取序列周期规律从而影响预测精度的问题,提出了一种基于完全自适应噪声集合经验模态分解(complete ensemble empirical mode decomposition with adaptive noise,CEEMDAN)和奇异谱分析(singular spectrum analysis,SSA)双重分解的双向长短时记忆网络(bidirectional long and short time memory,BiLSTM)预测模型。首先,采用CEEMDAN对历史负荷进行分解,以得到若干个周期规律更为清晰的子序列;再利用多尺度熵(multiscale entropy,MSE)计算所有子序列的复杂程度,根据不同时间尺度上的样本熵值将相似的子序列重构聚合;然后,利用SSA去噪的功能,对高度复杂的新序列进行二次分解,去除序列中的噪声并提取更为主要的规律,从而进一步提高中长序列预测精度;再将得到的最终一组子序列输入BiLSTM进行预测;最后,考虑到天气、节假日等外部因素对电力负荷的影响,提出了一种误差修正技术。选取了巴拿马某地区的用电负荷进行实验,实验结果表明,经过双重分解可以将均方根误差降低87.4%;预测未来一年的负荷序列时,采用的BiLSTM模型将拟合系数最高提高2.5%;所提出的误差修正技术可将均方根误差降低9.7%。
文摘目的应用Spectrum/EPP模型估计新疆兵团成人艾滋病(acquired immune deficiency syndrome,AIDS)疫情及发展趋势预测,为AIDS防治策略制定提供有效参考数据和科学基础。方法收集整理新疆兵团1996—2022年的AIDS相关历史数据,利用Spectrum模型将全人群分成8类亚人群,结合新疆兵团人口学数据、AIDS流行参数和监测数据、抗病毒治疗及预防母婴传播数据对新疆兵团AIDS疫情进行估计和预测。结果截至2022年底,新疆兵团累计存活≥15岁人类免疫缺陷病毒(human immunodeficiency virus,HIV)感染者/AIDS病例数共1547例,各类人群感染HIV情况,其中吸毒人群37例、暗娼人群26例、嫖客人群114例、男同性恋人群190例、献血人群25例、感染者配偶/性伴人群86例、剩余男730例、剩余女323例和母婴传播16例。同EPP-Spectrum模型估计结果比较发现,累计存活病例比例高达91.43%,经Spectrum/EPP模型估计,新疆兵团AIDS疫情仍呈逐年上升趋势,估计2022年新疆兵团≥15岁人群HIV感染率为7/10万,传播途径主要以异性传播为主。经过R-Spline模型分别拟合各类亚人群HIV阳性率,暗娼人群HIV阳性率逐渐上升与中位数数值趋势偏离,男男性行为者(men who have sex with men,MSM)人群HIV阳性率呈缓慢上升趋势;注射吸毒者人群HIV阳性率整体处于先升后降的趋势;嫖客人群HIV阳性率一直处于平稳状态;有偿献血人群HIV阳性率为下降趋势;配偶/性伴人群HIV阳性率高于其他亚人群,阳性率在快速上升之后转至平滑曲线。结论经过Spectrum/EPP模型预测发现,新疆兵团仍属于AIDS低流行地区,嫖客、MSM和配偶/性伴人群将是新疆兵团今后AIDS防控工作的重点关注人群。针对不同人群及时做好干预工作以及宣传教育,及时发现更多的HIV感染者,从而降低新疆兵团AIDS发病率。
