The observation error model of the underwater acous-tic positioning system is an important factor to influence the positioning accuracy of the underwater target.For the position inconsistency error caused by consideri...The observation error model of the underwater acous-tic positioning system is an important factor to influence the positioning accuracy of the underwater target.For the position inconsistency error caused by considering the underwater tar-get as a mass point,as well as the observation system error,the traditional error model best estimation trajectory(EMBET)with little observed data and too many parameters can lead to the ill-condition of the parameter model.In this paper,a multi-station fusion system error model based on the optimal polynomial con-straint is constructed,and the corresponding observation sys-tem error identification based on improved spectral clustering is designed.Firstly,the reduced parameter unified modeling for the underwater target position parameters and the system error is achieved through the polynomial optimization.Then a multi-sta-tion non-oriented graph network is established,which can address the problem of the inaccurate identification for the sys-tem errors.Moreover,the similarity matrix of the spectral cluster-ing is improved,and the iterative identification for the system errors based on the improved spectral clustering is proposed.Finally,the comprehensive measured data of long baseline lake test and sea test show that the proposed method can accu-rately identify the system errors,and moreover can improve the positioning accuracy for the underwater target positioning.展开更多
The similarity measure is crucial to the performance of spectral clustering. The Gaussian kernel function based on the Euclidean distance is usual y adopted as the similarity measure. However, the Euclidean distance m...The similarity measure is crucial to the performance of spectral clustering. The Gaussian kernel function based on the Euclidean distance is usual y adopted as the similarity measure. However, the Euclidean distance measure cannot ful y reveal the complex distribution data, and the result of spectral clustering is very sensitive to the scaling parameter. To solve these problems, a new manifold distance measure and a novel simulated anneal-ing spectral clustering （SASC） algorithm based on the manifold distance measure are proposed. The simulated annealing based on genetic algorithm （SAGA）, characterized by its rapid convergence to the global optimum, is used to cluster the sample points in the spectral mapping space. The proposed algorithm can not only reflect local and global consistency better, but also reduce the sensitivity of spectral clustering to the kernel parameter, which improves the algorithm’s clustering performance. To efficiently apply the algorithm to image segmentation, the Nystrom method is used to reduce the computation complexity. Experimental results show that compared with traditional clustering algorithms and those popular spectral clustering algorithms, the proposed algorithm can achieve better clustering performances on several synthetic datasets, texture images and real images.展开更多
A new fuzzy support vector machine algorithm with dual membership values based on spectral clustering method is pro- posed to overcome the shortcoming of the normal support vector machine algorithm, which divides the ...A new fuzzy support vector machine algorithm with dual membership values based on spectral clustering method is pro- posed to overcome the shortcoming of the normal support vector machine algorithm, which divides the training datasets into two absolutely exclusive classes in the binary classification, ignoring the possibility of ＂overlapping＂ region between the two training classes. The proposed method handles sample ＂overlap＂ effi- ciently with spectral clustering, overcoming the disadvantages of over-fitting well, and improving the data mining efficiency greatly. Simulation provides clear evidences to the new method.展开更多
随着国家大力推进能源供给侧结构性改革,新能源装机容量不断提升,电力市场竞争愈加激烈。另一方面,全球煤炭市场的复杂多变,导致以煤炭为能量来源的发电企业成本上涨。燃煤发热量是衡量煤质的重要评价标准之一,也是采购煤炭最重要的依据...随着国家大力推进能源供给侧结构性改革,新能源装机容量不断提升,电力市场竞争愈加激烈。另一方面,全球煤炭市场的复杂多变,导致以煤炭为能量来源的发电企业成本上涨。燃煤发热量是衡量煤质的重要评价标准之一,也是采购煤炭最重要的依据,对燃煤发热量进行准确预测能够有效地控制电厂运行采购成本。为了实现燃煤发热量的高效预测,采用Pearson系数对相关变量进行特征选取,采用基于密度的噪点空间聚类(Density-Based Spatial Clustering of Applications with Noise,DBSCAN)算法对某电厂自备煤厂近2年1733条化验数据进行去噪,对去噪后数据进行谱聚类(Spectral Clustering,SC)分析。将分类后的子样本集采用极致梯度提升(Extreme Gradient Boosting,XGBoost)算法分别建立预测模型,并与最小二乘法回归(Ordinary Least Squares,OLS)、支持向量机(Support Vector Machines,SVM)模型进行性能比较。结果表明,基于XGBoost的电站燃煤发热量预测模型相较于其他算法准确性有明显提升,泛化能力更强。对经过SC算法分类后的燃煤分别建立预测模型能够进一步提高模型的精细化水平,为燃煤电站发热量预测提供一种可靠高效的方法。展开更多
In this paper,the synthesis of a novel polyoxovanadiumborate Cd_(0.75) Na_2 Ni_(0.5) [V_(12) B_(18) O_(50.5)(OH)_(9.5)]·27.5 H_2 O by hydrothermal method.The compound consists of metal M(Cd,Na,Ni)and V_(12) B_(18...In this paper,the synthesis of a novel polyoxovanadiumborate Cd_(0.75) Na_2 Ni_(0.5) [V_(12) B_(18) O_(50.5)(OH)_(9.5)]·27.5 H_2 O by hydrothermal method.The compound consists of metal M(Cd,Na,Ni)and V_(12) B_(18) O_(60) cluster units connected through the M-O bond to form a three-dimensional structure.We performed a series of characterizations of Compound 1 of our team and tested its fluorescence properties at the same time.The luminescence investigations show that the compound 1 displays an interesting luminescence property.The compound 1 exhibits a good potential as a luminescent multi-responsive sensing material for Fe^(3+)ions.展开更多
针对谱聚类在尺度参数计算时需要人为设置近邻参数及聚类结果不稳定等问题,本文将初始类中心值和尺度参数作为决策变量,重点对谱聚类算法进行自适应优化与改进。首先,将样本邻域标准差的倒数作为度量样本局部密度的参数,与密度峰值思想...针对谱聚类在尺度参数计算时需要人为设置近邻参数及聚类结果不稳定等问题,本文将初始类中心值和尺度参数作为决策变量,重点对谱聚类算法进行自适应优化与改进。首先,将样本邻域标准差的倒数作为度量样本局部密度的参数,与密度峰值思想相结合,设计了一种基于密度峰值的初始类中心决策值选择方法(initial class center decision value algorithm based on density peak,DP_KD),解决密度调整谱聚类中聚类结果不稳定的问题。其次,利用样本间的平均距离计算相应的邻域半径,并根据样本标准差自适应地求解每个样本的尺度参数,构造样本间的相似度矩阵,实现了近邻参数的自适应设置,解决尺度参数需要人为设置的问题。然后,基于优化后的初始类中心决策值和近邻参数方法,进一步调整高斯核函数,提出一种基于邻域标准差的密度调整谱聚类算法(density adjusted spectral clustering algorithm based on neighborhood standard deviation,DSSD),通过构建特征向量空间实现了密度谱聚类。最后,将提出的算法与其他聚类算法在多个数据集上进行了对比。结果表明,与其他谱聚类算法相比,本文提出的DSSD算法不仅具有更好的聚类效果,且聚类结果更加稳定,尤其是在类内密集且类间边缘明确的DIM512数据集中,DSSD算法可以正确地进行聚类分簇;在准确率、兰德系数和F-measure上较其他算法至少提升了0.0268、0.0136和0.0247,这表明DSSD算法不仅聚类效果较好且更适合大规模数据集的聚类分析。展开更多
基金This work was supported by the National Natural Science Foundation of China(61903086,61903366,62001115)the Natural Science Foundation of Hunan Province(2019JJ50745,2020JJ4280,2021JJ40133)the Fundamentals and Basic of Applications Research Foundation of Guangdong Province(2019A1515110136).
文摘The observation error model of the underwater acous-tic positioning system is an important factor to influence the positioning accuracy of the underwater target.For the position inconsistency error caused by considering the underwater tar-get as a mass point,as well as the observation system error,the traditional error model best estimation trajectory(EMBET)with little observed data and too many parameters can lead to the ill-condition of the parameter model.In this paper,a multi-station fusion system error model based on the optimal polynomial con-straint is constructed,and the corresponding observation sys-tem error identification based on improved spectral clustering is designed.Firstly,the reduced parameter unified modeling for the underwater target position parameters and the system error is achieved through the polynomial optimization.Then a multi-sta-tion non-oriented graph network is established,which can address the problem of the inaccurate identification for the sys-tem errors.Moreover,the similarity matrix of the spectral cluster-ing is improved,and the iterative identification for the system errors based on the improved spectral clustering is proposed.Finally,the comprehensive measured data of long baseline lake test and sea test show that the proposed method can accu-rately identify the system errors,and moreover can improve the positioning accuracy for the underwater target positioning.
基金supported by the National Natural Science Foundationof China(61272119)
文摘The similarity measure is crucial to the performance of spectral clustering. The Gaussian kernel function based on the Euclidean distance is usual y adopted as the similarity measure. However, the Euclidean distance measure cannot ful y reveal the complex distribution data, and the result of spectral clustering is very sensitive to the scaling parameter. To solve these problems, a new manifold distance measure and a novel simulated anneal-ing spectral clustering （SASC） algorithm based on the manifold distance measure are proposed. The simulated annealing based on genetic algorithm （SAGA）, characterized by its rapid convergence to the global optimum, is used to cluster the sample points in the spectral mapping space. The proposed algorithm can not only reflect local and global consistency better, but also reduce the sensitivity of spectral clustering to the kernel parameter, which improves the algorithm’s clustering performance. To efficiently apply the algorithm to image segmentation, the Nystrom method is used to reduce the computation complexity. Experimental results show that compared with traditional clustering algorithms and those popular spectral clustering algorithms, the proposed algorithm can achieve better clustering performances on several synthetic datasets, texture images and real images.
基金supported by the National Natural Science Foundation of China (7083100170821061)
文摘A new fuzzy support vector machine algorithm with dual membership values based on spectral clustering method is pro- posed to overcome the shortcoming of the normal support vector machine algorithm, which divides the training datasets into two absolutely exclusive classes in the binary classification, ignoring the possibility of ＂overlapping＂ region between the two training classes. The proposed method handles sample ＂overlap＂ effi- ciently with spectral clustering, overcoming the disadvantages of over-fitting well, and improving the data mining efficiency greatly. Simulation provides clear evidences to the new method.
文摘随着国家大力推进能源供给侧结构性改革,新能源装机容量不断提升,电力市场竞争愈加激烈。另一方面,全球煤炭市场的复杂多变,导致以煤炭为能量来源的发电企业成本上涨。燃煤发热量是衡量煤质的重要评价标准之一,也是采购煤炭最重要的依据,对燃煤发热量进行准确预测能够有效地控制电厂运行采购成本。为了实现燃煤发热量的高效预测,采用Pearson系数对相关变量进行特征选取,采用基于密度的噪点空间聚类(Density-Based Spatial Clustering of Applications with Noise,DBSCAN)算法对某电厂自备煤厂近2年1733条化验数据进行去噪,对去噪后数据进行谱聚类(Spectral Clustering,SC)分析。将分类后的子样本集采用极致梯度提升(Extreme Gradient Boosting,XGBoost)算法分别建立预测模型,并与最小二乘法回归(Ordinary Least Squares,OLS)、支持向量机(Support Vector Machines,SVM)模型进行性能比较。结果表明,基于XGBoost的电站燃煤发热量预测模型相较于其他算法准确性有明显提升,泛化能力更强。对经过SC算法分类后的燃煤分别建立预测模型能够进一步提高模型的精细化水平,为燃煤电站发热量预测提供一种可靠高效的方法。
基金supported by the NNSFC(21473030,1371033)the Natural Science Foundation of Fujian Province(2013J01042)
文摘In this paper,the synthesis of a novel polyoxovanadiumborate Cd_(0.75) Na_2 Ni_(0.5) [V_(12) B_(18) O_(50.5)(OH)_(9.5)]·27.5 H_2 O by hydrothermal method.The compound consists of metal M(Cd,Na,Ni)and V_(12) B_(18) O_(60) cluster units connected through the M-O bond to form a three-dimensional structure.We performed a series of characterizations of Compound 1 of our team and tested its fluorescence properties at the same time.The luminescence investigations show that the compound 1 displays an interesting luminescence property.The compound 1 exhibits a good potential as a luminescent multi-responsive sensing material for Fe^(3+)ions.
文摘针对谱聚类在尺度参数计算时需要人为设置近邻参数及聚类结果不稳定等问题,本文将初始类中心值和尺度参数作为决策变量,重点对谱聚类算法进行自适应优化与改进。首先,将样本邻域标准差的倒数作为度量样本局部密度的参数,与密度峰值思想相结合,设计了一种基于密度峰值的初始类中心决策值选择方法(initial class center decision value algorithm based on density peak,DP_KD),解决密度调整谱聚类中聚类结果不稳定的问题。其次,利用样本间的平均距离计算相应的邻域半径,并根据样本标准差自适应地求解每个样本的尺度参数,构造样本间的相似度矩阵,实现了近邻参数的自适应设置,解决尺度参数需要人为设置的问题。然后,基于优化后的初始类中心决策值和近邻参数方法,进一步调整高斯核函数,提出一种基于邻域标准差的密度调整谱聚类算法(density adjusted spectral clustering algorithm based on neighborhood standard deviation,DSSD),通过构建特征向量空间实现了密度谱聚类。最后,将提出的算法与其他聚类算法在多个数据集上进行了对比。结果表明,与其他谱聚类算法相比,本文提出的DSSD算法不仅具有更好的聚类效果,且聚类结果更加稳定,尤其是在类内密集且类间边缘明确的DIM512数据集中,DSSD算法可以正确地进行聚类分簇;在准确率、兰德系数和F-measure上较其他算法至少提升了0.0268、0.0136和0.0247,这表明DSSD算法不仅聚类效果较好且更适合大规模数据集的聚类分析。
文摘为了提高辨识稳定图中真实模态的准确性与自动化程度,首先,从稳定点定义方式的角度论述了聚类算法效果欠佳的原因,并采用异阶系统非等权重的定义方式输出稳定点;其次,基于数据挖掘思想,采用改进的辨识聚类结构的有序点(ordering points to identify the clustering structure,简称OPTICS)算法自动清洗稳定点集,通过遍历性搜索的方式确定输入参数;然后,提出结合度矩阵去噪的自适应局部密度谱聚类(local density adaptive spectral clustering,简称SC-DA)算法分析稳定点集,并以簇中值作为模态参数的代表值,实现模态参数的自动化识别;最后,将含有密集模态的外滩大桥作为识别对象进行试验验证。试验结果表明:所提出方法具有较高的精度,与频域分解(frequency domain decomposition,简称FDD)法的频率结果最大相差仅为0.012 3 Hz,且在线识别的准确率达到82.86%,显著高于基于层次聚类的自动识别方法,实现了无人工干预下模态参数的自动、准确识别,具有一定的工程应用前景。
