The safety margin criterion of nonlinear dynamic question of an elastic rotor system are given. A series of observing spaces were separated from integral space by resolving and polymerizing method. The stable_state tr...The safety margin criterion of nonlinear dynamic question of an elastic rotor system are given. A series of observing spaces were separated from integral space by resolving and polymerizing method. The stable_state trajectory of high dimensional nonlinear dynamic systems was got within integral space.According to international standard of rotor system vibration, energy limits of safety criterion were determined. The safety margin was calculated within a series of observing spaces by comparative positive_area criterion (CPAC) method. A quantitative example calculating safety margin for unbalance elastic rotor system was given by CPAC. The safety margin criterion proposed includes the calculation of current stability margin in engineering. This criterion is an effective method to solve quantitative calculation question of safety margin and stability margin for nonlinear dynamic systems.展开更多
研究提出一种模糊双向最大间距准则(fuzzy bidirectional maximum margin criterion,FBMMC)特征提取方法,并将其用于人脸识别。在FBMMC中,首先通过引入原始训练样本集的模糊隶属度矩阵,定义了面向图像的行方向模糊离散度矩阵和行方向模...研究提出一种模糊双向最大间距准则(fuzzy bidirectional maximum margin criterion,FBMMC)特征提取方法,并将其用于人脸识别。在FBMMC中,首先通过引入原始训练样本集的模糊隶属度矩阵,定义了面向图像的行方向模糊离散度矩阵和行方向模糊最大间距准则,进一步求得行方向最优投影矩阵;然后,对原始训练样本集中的每一个样本,采用行方向最优投影矩阵进行投影变换,从而得到行方向特征训练样本集。同样地,通过引入行方向特征训练样本集的模糊隶属度矩阵,给出了面向图像的列方向模糊离散度矩阵和列方向模糊最大间距准则的定义,进一步求得列方向最优投影矩阵。FBMMC在得到行、列两个方向的最优投影矩阵后,就可以将原始数据空间的样本数据投影到一个相对低维的特征空间,从而完成对原始样本数据的特征提取。在ORL和Yale人脸数据库上的实验结果表明,文中提出的模糊双向最大间距准则特征提取方法用于人脸识别具有较高的识别率。展开更多
依据最大间距判别准则(Maximum margin criterion,MMC)的基本原理,并结合模糊技术和张量理论,提出一种矩阵模式的模糊最大间距判别准则(Matrix model fuzzy maximum margin criterion,MFMMC),并在此基础上形成具有模糊聚类功能的双向二...依据最大间距判别准则(Maximum margin criterion,MMC)的基本原理,并结合模糊技术和张量理论,提出一种矩阵模式的模糊最大间距判别准则(Matrix model fuzzy maximum margin criterion,MFMMC),并在此基础上形成具有模糊聚类功能的双向二维无监督特征提取方法(Two-directional two-dimensional unsupervised feature extraction method with fuzzy clustering ability,(2D)2UFFCA).该方法不但能直接实现矩阵模式数据的模糊聚类,而且还可以对矩阵模式数据进行双向二维特征提取,实现特征降维.同时我们还从几何的直观含义出发,合理地设定矩阵模式的模糊最大间距判别准则中的调节参数γ,并从理论上证明其合理性.为了提高特征提取的效率,还提出一种能有效计算矩阵模式数据的投影变换矩阵的方法.实验结果表明该方法具有上述优势.展开更多
文摘The safety margin criterion of nonlinear dynamic question of an elastic rotor system are given. A series of observing spaces were separated from integral space by resolving and polymerizing method. The stable_state trajectory of high dimensional nonlinear dynamic systems was got within integral space.According to international standard of rotor system vibration, energy limits of safety criterion were determined. The safety margin was calculated within a series of observing spaces by comparative positive_area criterion (CPAC) method. A quantitative example calculating safety margin for unbalance elastic rotor system was given by CPAC. The safety margin criterion proposed includes the calculation of current stability margin in engineering. This criterion is an effective method to solve quantitative calculation question of safety margin and stability margin for nonlinear dynamic systems.
文摘研究提出一种模糊双向最大间距准则(fuzzy bidirectional maximum margin criterion,FBMMC)特征提取方法,并将其用于人脸识别。在FBMMC中,首先通过引入原始训练样本集的模糊隶属度矩阵,定义了面向图像的行方向模糊离散度矩阵和行方向模糊最大间距准则,进一步求得行方向最优投影矩阵;然后,对原始训练样本集中的每一个样本,采用行方向最优投影矩阵进行投影变换,从而得到行方向特征训练样本集。同样地,通过引入行方向特征训练样本集的模糊隶属度矩阵,给出了面向图像的列方向模糊离散度矩阵和列方向模糊最大间距准则的定义,进一步求得列方向最优投影矩阵。FBMMC在得到行、列两个方向的最优投影矩阵后,就可以将原始数据空间的样本数据投影到一个相对低维的特征空间,从而完成对原始样本数据的特征提取。在ORL和Yale人脸数据库上的实验结果表明,文中提出的模糊双向最大间距准则特征提取方法用于人脸识别具有较高的识别率。
文摘局部保持投影(Locality preserving projections,LPP)算法只保持了目标在投影后的邻域局部信息,为了更好地刻画数据的流形结构,引入了类内和类间局部散度矩阵,给出了一种基于有效且稳定的大间距准则(Maximum margin criterion,MMC)的不相关保局投影分析方法,该方法在最大化散度矩阵迹差时,引入尺度因子α,对类内和类间局部散度矩阵进行加权,以便找到更适合分类的子空间并且可避免小样本问题;更重要的是,大间距准则下提取的判别特征集一般情况下是统计相关的,造成了特征信息的冗余,因此,通过增加一个不相关约束条件,利用推导出的公式提取不相关判别特征集,这样做,对正确识别更为有利.在Yale人脸库、PIE人脸库和MNIST手写数字库上的测试结果表明,本文方法有效且稳定,与LPP、LDA(Linear discriminant analysis)和LPMIP(Locality-preserved maximum information projection)方法等相比,具有更高的正确识别率。
