A novel control method for the nonlinear and complex plants with environmental uncertainties and variable parameters has been proposed by use of the nearest neighborhood clustering algorithm, the fuzzy control and the...A novel control method for the nonlinear and complex plants with environmental uncertainties and variable parameters has been proposed by use of the nearest neighborhood clustering algorithm, the fuzzy control and the variable regressive estimation (VRE) technology. It overcomes the defects of the other adaptive methods such as the strong dependence to the system and the difficulty of the acquirement of the professional knowledge during the modifying period of the rules. The application of new algorithm to the electrical heating furnace with multiple zones demonstrates the advantages of the proposed method.展开更多
The percolation fields constructed around the elements of a cluster system in the phase spaces of properties are studied.It is shown that such neighborhoods significantly increase the number of structure parameters of...The percolation fields constructed around the elements of a cluster system in the phase spaces of properties are studied.It is shown that such neighborhoods significantly increase the number of structure parameters of the system under study,expanding the possibilities of analytical description.To study the structure and properties of such systems in the proposed model,a three-dimensional continuum percolation problem with interacting elements is solved.The dependences of the structure and properties of clusters on the parameters of the generation processes of the cluster system are studied,and analytical dependences are obtained.展开更多
针对现有离群点检测算法在运用于大规模数据集时时间效率较低的问题,提出一种基于K近邻的并行离群点检测算法PODKNN(Parallel Outlier Detection Based on K-nearest Neighborhood)。该算法利用划分策略对数据集进行预处理,在规模较小...针对现有离群点检测算法在运用于大规模数据集时时间效率较低的问题,提出一种基于K近邻的并行离群点检测算法PODKNN(Parallel Outlier Detection Based on K-nearest Neighborhood)。该算法利用划分策略对数据集进行预处理,在规模较小的子集中寻找K近邻并计算离群度,最后合并结果并遴选出离群点,设计算法过程使其符合MapReduce的编程模型,实现并行化,从而提高了离群点检测算法处理大规模数据的计算效率。实验结果表明,PODKNN具有较高的加速比及较好的扩展性。展开更多
针对水声目标信号复杂、样本获取难度大且富含不确定信息的问题,研究了一种新的证据K类近邻识别算法(Evidence K Nearest Neighbor,EK-NN)。首先在水声目标的各类训练样本中,根据特征距离大小选取待识别目标的K近邻,并构造其基本置信指...针对水声目标信号复杂、样本获取难度大且富含不确定信息的问题,研究了一种新的证据K类近邻识别算法(Evidence K Nearest Neighbor,EK-NN)。首先在水声目标的各类训练样本中,根据特征距离大小选取待识别目标的K近邻,并构造其基本置信指派函数。然后使用证据理论中的Dempster-Shafer(D-S)规则对各类别下的近邻证据进行组合,最后再应用冲突置信的比例分配规则5(Redistribute Conflicting mass proportionally rule5,PCR5)将所有类别的组合证据进行融合,并根据融合结果和所设立的分类规则来判断目标的类别属性。根据水声目标实测数据,将新算法与其他几种常见的水声目标识别算法进行了对比分析,结果表明新算法能有效提高识别的准确率。展开更多
针对复杂工业过程数据的动态性、非线性和多阶段性等特征,提出基于时空近邻标准化和KNN规则(Time-Space Nearest Neighborhood Standardization and K Nearest Neighbor Rule,TSNS-KNN)的复杂多阶段过程故障检测方法。首先使用训练样本...针对复杂工业过程数据的动态性、非线性和多阶段性等特征,提出基于时空近邻标准化和KNN规则(Time-Space Nearest Neighborhood Standardization and K Nearest Neighbor Rule,TSNS-KNN)的复杂多阶段过程故障检测方法。首先使用训练样本在时间和空间域上的两层嵌套近邻集的统计信息对样本预处理,然后将标准样本的累积近邻距离作为检测统计量进行故障检测。TSNS-KNN在排除非线性干扰的同时,消除了前后时刻样本间的动态相关性,将多阶段数据转换为单阶段数据,从而实现对复杂多阶段过程的检测。将该方法运用于数值实验和青霉素发酵过程,并与其他方法进行比较,对比结果进一步验证了TSNS-KNN方法的优越性。展开更多
文摘A novel control method for the nonlinear and complex plants with environmental uncertainties and variable parameters has been proposed by use of the nearest neighborhood clustering algorithm, the fuzzy control and the variable regressive estimation (VRE) technology. It overcomes the defects of the other adaptive methods such as the strong dependence to the system and the difficulty of the acquirement of the professional knowledge during the modifying period of the rules. The application of new algorithm to the electrical heating furnace with multiple zones demonstrates the advantages of the proposed method.
文摘The percolation fields constructed around the elements of a cluster system in the phase spaces of properties are studied.It is shown that such neighborhoods significantly increase the number of structure parameters of the system under study,expanding the possibilities of analytical description.To study the structure and properties of such systems in the proposed model,a three-dimensional continuum percolation problem with interacting elements is solved.The dependences of the structure and properties of clusters on the parameters of the generation processes of the cluster system are studied,and analytical dependences are obtained.
文摘针对现有离群点检测算法在运用于大规模数据集时时间效率较低的问题,提出一种基于K近邻的并行离群点检测算法PODKNN(Parallel Outlier Detection Based on K-nearest Neighborhood)。该算法利用划分策略对数据集进行预处理,在规模较小的子集中寻找K近邻并计算离群度,最后合并结果并遴选出离群点,设计算法过程使其符合MapReduce的编程模型,实现并行化,从而提高了离群点检测算法处理大规模数据的计算效率。实验结果表明,PODKNN具有较高的加速比及较好的扩展性。
文摘为了改善利用SCATS交通数据估计路段行程时间的效果,通过分析SCATS实际交通数据获取时间间隔不一致的特征,构建了SCATS交通数据虚拟时间序列,将利用因子分析法提取的累计贡献率在85%以上的主因子作为交通模式特征向量的构成要素,用欧氏距离作为当前交通模式特征向量和历史交通模式特征向量相似性的测度指标,以路段行程时间估计误差最小为目标选取当前交通模式的近邻数,对交通模式之间距离的倒数进行归一化处理,确定了相似交通模式的行程时间权重,设计了基于SCATS交通数据的路段行程时间估计方法.实例结果表明:与多元线性回归方法相比,本文方法估计的路段行程时间平均绝对误差、平均绝对百分比误差和均方根误差分别平均减少了9.68 s、8.07%和4.5 s.
文摘针对水声目标信号复杂、样本获取难度大且富含不确定信息的问题,研究了一种新的证据K类近邻识别算法(Evidence K Nearest Neighbor,EK-NN)。首先在水声目标的各类训练样本中,根据特征距离大小选取待识别目标的K近邻,并构造其基本置信指派函数。然后使用证据理论中的Dempster-Shafer(D-S)规则对各类别下的近邻证据进行组合,最后再应用冲突置信的比例分配规则5(Redistribute Conflicting mass proportionally rule5,PCR5)将所有类别的组合证据进行融合,并根据融合结果和所设立的分类规则来判断目标的类别属性。根据水声目标实测数据,将新算法与其他几种常见的水声目标识别算法进行了对比分析,结果表明新算法能有效提高识别的准确率。
文摘针对复杂工业过程数据的动态性、非线性和多阶段性等特征,提出基于时空近邻标准化和KNN规则(Time-Space Nearest Neighborhood Standardization and K Nearest Neighbor Rule,TSNS-KNN)的复杂多阶段过程故障检测方法。首先使用训练样本在时间和空间域上的两层嵌套近邻集的统计信息对样本预处理,然后将标准样本的累积近邻距离作为检测统计量进行故障检测。TSNS-KNN在排除非线性干扰的同时,消除了前后时刻样本间的动态相关性,将多阶段数据转换为单阶段数据,从而实现对复杂多阶段过程的检测。将该方法运用于数值实验和青霉素发酵过程,并与其他方法进行比较,对比结果进一步验证了TSNS-KNN方法的优越性。
