This study focuses on age,growth rate and diameter distribution of pine forests in the Malam Jabba area,Swat District,Pakistan.Wood core samples were taken from twenty stands.Picea smithiana was the oldest at 234 year...This study focuses on age,growth rate and diameter distribution of pine forests in the Malam Jabba area,Swat District,Pakistan.Wood core samples were taken from twenty stands.Picea smithiana was the oldest at 234 years with a 112-cm diameter.Abies pindrow was 125 years with an 80-cm diameter while the oldest Pinus wallichiana was 122 years with 75-cm diameter.The fastest overall growth rate of 1.5±0.1 year/cm was for P.wallichiana on a west-facing aspect,while the slowest 5.8±2.6 year/cm growth was P.smithiana on an eastfacing exposure.P.wallichiana and A.pindrow exhibited marked differences in growth rates over a 5-year period.The highest growth was by P.wallichiana from 1966 to 2006.A.pindrow showed less growth over the same years,such pattern simultaneously reverse from 1911 to 1965.The relationship between diameter and age,diameter and growth rate and age and growth rate were correlated.P.wallichiana and A.pindrow ages were correlated with diameter and growth rates.P.smithiana age was positive correlated with diameter.Generally,topographic and edaphic factors did not show significant correlations with growth rates,although some appreciable correlations were recorded.The growth of P.wallichiana was correlated with elevation while A.pindrow was correlated with maximum water retaining capacity.Diameter and age produced uneven size classes and many size gaps,which could be the result of anthropogenic disturbances.展开更多
针对机械系统状态监测与故障诊断中存在的故障特征维数较高及模式识别导致的耗时较高问题,提出了一种基于自适应局部保持投影(Locality Preserving Projection,LPP)特征降维和改进多变量预测模型(Variable Predictive Model based Class...针对机械系统状态监测与故障诊断中存在的故障特征维数较高及模式识别导致的耗时较高问题,提出了一种基于自适应局部保持投影(Locality Preserving Projection,LPP)特征降维和改进多变量预测模型(Variable Predictive Model based Class Discriminate,VPMCD)的故障诊断方法。首先,从滚动轴承振动信号中提取时频域特征、能量特征,以及复杂度特征组成高维故障特征数据集;其次,利用自适应LPP方法对高维故障特征数据集进行降维处理,得到低维敏感故障特征;最后,采用改进VPMCD方法对低维敏感故障特征进行分类识别,进而判断故障类型。通过滚动轴承故障诊断试验分析表明,自适应LPP方法克服了传统LPP方法需要人工选取参数的缺陷,在获得低维敏感故障特征的基础上具有较少计算时间,相比主成分分析(Principal Component Analysis,PCA)、局部切空间排列(Local Tangent Space Alignment,LTSA)、线性局部切空间排列(Linear Local Tangent Space Alignment,LLTSA)、等距特征映射(Isometric Mapping,Isomap),以及局部线性嵌入(Locally Linear Embedding,LLE)等算法具有明显的优势;改进VPMCD方法可克服人工选择模型的偶然性和片面性,在滚动轴承10种故障状态的识别中获得了99.4%的诊断精度,相比优化参数支持向量机方法提高了故障诊断效率,大大降低了识别时间,具有一定的优越性。展开更多
The author brought forward two suppositions in this thesis:The students' achievements have the direct proportion with the degree of the appreciation to their teachers.Aiming at the supposition,the author developed...The author brought forward two suppositions in this thesis:The students' achievements have the direct proportion with the degree of the appreciation to their teachers.Aiming at the supposition,the author developed a set of statistics modeling and simulation study,which focused on the teachers' characters,the students' appreciations and students' exam scores in order to reveal the influence of teachers' affective variables on students' achievements.Therefore,according to the series of the suppositions,the author suggested that teachers should pay full attention to the students' emotional needs and decrease the negative effects on students.Meanwhile,the students should encourage the teachers to be absorbed in their instruction by cooperating with their teachers actively in order to acquire abundant knowledge in the limited time.展开更多
提出了基于VPMCD(Variable Predictive Model Based Class Discriminate,简称VPMCD)和EMD(Empirical mode decomposition,简称EMD)的齿轮故障诊断方法,并将它应用于齿轮稳态信号的分析。VPMCD方法是一种新的模式识别方法,特别适合于非...提出了基于VPMCD(Variable Predictive Model Based Class Discriminate,简称VPMCD)和EMD(Empirical mode decomposition,简称EMD)的齿轮故障诊断方法,并将它应用于齿轮稳态信号的分析。VPMCD方法是一种新的模式识别方法,特别适合于非线性分类问题,它充分利用从原始数据中所提取的特征值之间的相互内在关系建立数学模型,从而进行模式识别。在基于VPMCD和EMD的齿轮故障诊断方法中,首先采用EMD方法将齿轮振动信号自适应地分解为若干个单分量信号,然后提取各个分量的样本熵并将其作为特征值,最后采用VPMCD分类器进行故障识别和分类。结果表明该方法能够有效地突出齿轮故障振动信号的故障特征,提高了齿轮故障诊断的准确性。展开更多
变量预测模型的模式识别方法(Variable predictive model based class discriminate,VPMCD)是一种利用特征值相互内在关系进行模式识别的新方法。论文提出了基于局部均值分解LMD(Local mean decomposition,LMD)能量矩概念,并针对轴承故...变量预测模型的模式识别方法(Variable predictive model based class discriminate,VPMCD)是一种利用特征值相互内在关系进行模式识别的新方法。论文提出了基于局部均值分解LMD(Local mean decomposition,LMD)能量矩概念,并针对轴承故障振动信号特征值的相互内在联系,将LMD能量矩与变量预测模型模式识别相结合,提出了一种轴承故障智能诊断新方法。首先利用LMD方法将复杂非平稳的原始信号分解为若干PF(Product function,PF)分量;然后利用相关分析剔除LMD方法中的虚假PF分量,并提取真实PF分量能量矩组成特征向量来有效地表达故障信息;最后采用VPMCD方法进行轴承故障诊断。通过仿真信号验证了PF能量矩比PF能量更能反映非平稳信号本质特征。轴承故障诊断实验结果表明,论文提出的方法能有效地应用于小样本多分类轴承故障智能诊断。展开更多
文摘This study focuses on age,growth rate and diameter distribution of pine forests in the Malam Jabba area,Swat District,Pakistan.Wood core samples were taken from twenty stands.Picea smithiana was the oldest at 234 years with a 112-cm diameter.Abies pindrow was 125 years with an 80-cm diameter while the oldest Pinus wallichiana was 122 years with 75-cm diameter.The fastest overall growth rate of 1.5±0.1 year/cm was for P.wallichiana on a west-facing aspect,while the slowest 5.8±2.6 year/cm growth was P.smithiana on an eastfacing exposure.P.wallichiana and A.pindrow exhibited marked differences in growth rates over a 5-year period.The highest growth was by P.wallichiana from 1966 to 2006.A.pindrow showed less growth over the same years,such pattern simultaneously reverse from 1911 to 1965.The relationship between diameter and age,diameter and growth rate and age and growth rate were correlated.P.wallichiana and A.pindrow ages were correlated with diameter and growth rates.P.smithiana age was positive correlated with diameter.Generally,topographic and edaphic factors did not show significant correlations with growth rates,although some appreciable correlations were recorded.The growth of P.wallichiana was correlated with elevation while A.pindrow was correlated with maximum water retaining capacity.Diameter and age produced uneven size classes and many size gaps,which could be the result of anthropogenic disturbances.
文摘针对机械系统状态监测与故障诊断中存在的故障特征维数较高及模式识别导致的耗时较高问题,提出了一种基于自适应局部保持投影(Locality Preserving Projection,LPP)特征降维和改进多变量预测模型(Variable Predictive Model based Class Discriminate,VPMCD)的故障诊断方法。首先,从滚动轴承振动信号中提取时频域特征、能量特征,以及复杂度特征组成高维故障特征数据集;其次,利用自适应LPP方法对高维故障特征数据集进行降维处理,得到低维敏感故障特征;最后,采用改进VPMCD方法对低维敏感故障特征进行分类识别,进而判断故障类型。通过滚动轴承故障诊断试验分析表明,自适应LPP方法克服了传统LPP方法需要人工选取参数的缺陷,在获得低维敏感故障特征的基础上具有较少计算时间,相比主成分分析(Principal Component Analysis,PCA)、局部切空间排列(Local Tangent Space Alignment,LTSA)、线性局部切空间排列(Linear Local Tangent Space Alignment,LLTSA)、等距特征映射(Isometric Mapping,Isomap),以及局部线性嵌入(Locally Linear Embedding,LLE)等算法具有明显的优势;改进VPMCD方法可克服人工选择模型的偶然性和片面性,在滚动轴承10种故障状态的识别中获得了99.4%的诊断精度,相比优化参数支持向量机方法提高了故障诊断效率,大大降低了识别时间,具有一定的优越性。
基金the staged achievement of the funded programe of 2014 Fine and Resource Sharing Course of the Middlebury College of Northern Polytechnical University
文摘The author brought forward two suppositions in this thesis:The students' achievements have the direct proportion with the degree of the appreciation to their teachers.Aiming at the supposition,the author developed a set of statistics modeling and simulation study,which focused on the teachers' characters,the students' appreciations and students' exam scores in order to reveal the influence of teachers' affective variables on students' achievements.Therefore,according to the series of the suppositions,the author suggested that teachers should pay full attention to the students' emotional needs and decrease the negative effects on students.Meanwhile,the students should encourage the teachers to be absorbed in their instruction by cooperating with their teachers actively in order to acquire abundant knowledge in the limited time.
文摘提出了基于VPMCD(Variable Predictive Model Based Class Discriminate,简称VPMCD)和EMD(Empirical mode decomposition,简称EMD)的齿轮故障诊断方法,并将它应用于齿轮稳态信号的分析。VPMCD方法是一种新的模式识别方法,特别适合于非线性分类问题,它充分利用从原始数据中所提取的特征值之间的相互内在关系建立数学模型,从而进行模式识别。在基于VPMCD和EMD的齿轮故障诊断方法中,首先采用EMD方法将齿轮振动信号自适应地分解为若干个单分量信号,然后提取各个分量的样本熵并将其作为特征值,最后采用VPMCD分类器进行故障识别和分类。结果表明该方法能够有效地突出齿轮故障振动信号的故障特征,提高了齿轮故障诊断的准确性。
文摘变量预测模型的模式识别方法(Variable predictive model based class discriminate,VPMCD)是一种利用特征值相互内在关系进行模式识别的新方法。论文提出了基于局部均值分解LMD(Local mean decomposition,LMD)能量矩概念,并针对轴承故障振动信号特征值的相互内在联系,将LMD能量矩与变量预测模型模式识别相结合,提出了一种轴承故障智能诊断新方法。首先利用LMD方法将复杂非平稳的原始信号分解为若干PF(Product function,PF)分量;然后利用相关分析剔除LMD方法中的虚假PF分量,并提取真实PF分量能量矩组成特征向量来有效地表达故障信息;最后采用VPMCD方法进行轴承故障诊断。通过仿真信号验证了PF能量矩比PF能量更能反映非平稳信号本质特征。轴承故障诊断实验结果表明,论文提出的方法能有效地应用于小样本多分类轴承故障智能诊断。
