平衡迭代规约层次聚类(balanced iterative reducing and clustering using hierarchies, BIRCH)算法是一个综合的层次聚类算法。但BIRCH算法为叶子节点中的簇设置统一的空间阈值,根据数据对象与簇之间的距离来决定数据对象的插入位置,...平衡迭代规约层次聚类(balanced iterative reducing and clustering using hierarchies, BIRCH)算法是一个综合的层次聚类算法。但BIRCH算法为叶子节点中的簇设置统一的空间阈值,根据数据对象与簇之间的距离来决定数据对象的插入位置,从而忽略了簇与簇之间的关系;此外,算法在分裂节点时,选取距离最远的2个聚类特征作为子簇,其他聚类特征会根据与这2个聚类特征之间的距离关系分裂为另外的子簇,造成处于簇与簇之间的样本数据错误分类,这样会忽略聚类特征之间的关系。针对BIRCH算法的这2个问题,提出了基于阈值的自适应算法,用于解决原算法统一空间阈值的问题;并在针对聚类特征关系的问题上,结合朴素贝叶斯算法对原算法进行改进。对改进后BIRCH算法与传统的算法进行仿真实验。结果表明,改进算法在损失效率的情况下,聚类效果得到了明显的改善,并且与其他算法相比,所提算法具有不错的表现性,而且具有跨数据集的鲁棒性。展开更多
In order to exactly provide scientific basis for pressure dynamic balance control of working chamber of earth pressure balance shield (EPBS),study on optimal arrangement of pressure measurement points in working chamb...In order to exactly provide scientific basis for pressure dynamic balance control of working chamber of earth pressure balance shield (EPBS),study on optimal arrangement of pressure measurement points in working chamber was conducted. Based on mathematical description of optimal arrangement for pressure measurement points,fuzzy clustering analysis and discriminant analysis were used to divide pressure regions of nodes on bulkhead. Finally,the selection method of optimal measurement points was proposed,and by selecting d6.28 m EPBS as study object,the case study was conducted. By contrast,based on optimal arrangement scheme of pressure measurement points,through adopting weighted algorithm,the absolute error mean of equivalent pressure of working chamber is the smallest. In addition,pressure curve of optimal arrangement points presents parabola,and it can show the state of pressure distribution on bulkhead truly. It is concluded that the optimal arrangement method of pressure measurement points in working chamber is effective and feasible,and the method can provide basis for realizing high precision pressure control of EPBS.展开更多
当前分布式光伏系统短期发电功率预测结构多设定为目标式,预测范围在实际发电环境下受限,导致平均绝对预测误差增加。为此设计基于利用层次方法的平衡迭代规约和聚类(Balanced Iterative Reducing and Clustering Using Hierarchies,BIR...当前分布式光伏系统短期发电功率预测结构多设定为目标式,预测范围在实际发电环境下受限,导致平均绝对预测误差增加。为此设计基于利用层次方法的平衡迭代规约和聚类(Balanced Iterative Reducing and Clustering Using Hierarchies,BIRCH)的分布式光伏系统短期发电功率预测方法。首先,明确预测指标,采用多层级的方式设计预测结构;其次,结合BIRCH原理,设计发电功率预测模型;最后,采用梯度回归处理的方式来实现最终预测。测试结果表明,对比传统变分模态分解-麻雀搜索算法-反向传播(Variational Mode Decomposition-Sparrow Search Algorithm-Back Propagation,VMD-SSA-BP)光伏系统短期发电功率预测小组、传统时序动态回归光伏系统短期发电功率预测小组,此次所设计的方法得出的平均绝对预测误差被较好地控制在2.1以下,预测效果更佳,针对性更强,误差可控,具有实际的应用价值。展开更多
提出了一种新的基于分层的平衡迭代归约及聚类(balance iterative reducing and clustering using hierarchies,BIRCH)算法的木材缺陷识别方法,讨论了关于分支因子(B,L)、阈值T的选取及非缺陷类判别问题。该方法通过在一定阈值内构建CF...提出了一种新的基于分层的平衡迭代归约及聚类(balance iterative reducing and clustering using hierarchies,BIRCH)算法的木材缺陷识别方法,讨论了关于分支因子(B,L)、阈值T的选取及非缺陷类判别问题。该方法通过在一定阈值内构建CF树,产生初始聚类,对初始聚类进行非缺陷类判别,自动识别木材缺陷类及位置并标记。实验结果表明,该算法能有效地进行木材缺陷识别,平均识别查准率约为86.3%,平均识别查全率约为90.1%。展开更多
文摘平衡迭代规约层次聚类(balanced iterative reducing and clustering using hierarchies, BIRCH)算法是一个综合的层次聚类算法。但BIRCH算法为叶子节点中的簇设置统一的空间阈值,根据数据对象与簇之间的距离来决定数据对象的插入位置,从而忽略了簇与簇之间的关系;此外,算法在分裂节点时,选取距离最远的2个聚类特征作为子簇,其他聚类特征会根据与这2个聚类特征之间的距离关系分裂为另外的子簇,造成处于簇与簇之间的样本数据错误分类,这样会忽略聚类特征之间的关系。针对BIRCH算法的这2个问题,提出了基于阈值的自适应算法,用于解决原算法统一空间阈值的问题;并在针对聚类特征关系的问题上,结合朴素贝叶斯算法对原算法进行改进。对改进后BIRCH算法与传统的算法进行仿真实验。结果表明,改进算法在损失效率的情况下,聚类效果得到了明显的改善,并且与其他算法相比,所提算法具有不错的表现性,而且具有跨数据集的鲁棒性。
基金Project(2007CB714006) supported by the National Basic Research Program of China
文摘In order to exactly provide scientific basis for pressure dynamic balance control of working chamber of earth pressure balance shield (EPBS),study on optimal arrangement of pressure measurement points in working chamber was conducted. Based on mathematical description of optimal arrangement for pressure measurement points,fuzzy clustering analysis and discriminant analysis were used to divide pressure regions of nodes on bulkhead. Finally,the selection method of optimal measurement points was proposed,and by selecting d6.28 m EPBS as study object,the case study was conducted. By contrast,based on optimal arrangement scheme of pressure measurement points,through adopting weighted algorithm,the absolute error mean of equivalent pressure of working chamber is the smallest. In addition,pressure curve of optimal arrangement points presents parabola,and it can show the state of pressure distribution on bulkhead truly. It is concluded that the optimal arrangement method of pressure measurement points in working chamber is effective and feasible,and the method can provide basis for realizing high precision pressure control of EPBS.
文摘提出了一种新的基于分层的平衡迭代归约及聚类(balance iterative reducing and clustering using hierarchies,BIRCH)算法的木材缺陷识别方法,讨论了关于分支因子(B,L)、阈值T的选取及非缺陷类判别问题。该方法通过在一定阈值内构建CF树,产生初始聚类,对初始聚类进行非缺陷类判别,自动识别木材缺陷类及位置并标记。实验结果表明,该算法能有效地进行木材缺陷识别,平均识别查准率约为86.3%,平均识别查全率约为90.1%。
