Inferential models are widely used in the chemical industry to infer key process variables, which are challenging or expensive to measure, from other more easily measured variables. The aim of this paper is three-fold...Inferential models are widely used in the chemical industry to infer key process variables, which are challenging or expensive to measure, from other more easily measured variables. The aim of this paper is three-fold: to present a theoretical review of some of the well known linear inferential modeling techniques, to enhance the predictive ability of the regularized canonical correlation analysis (RCCA) method, and finally to compare the performances of these techniques and highlight some of the practical issues that can affect their predictive abilities. The inferential modeling techniques considered in this study include full rank modeling techniques, such as ordinary least square (OLS) regression and ridge regression (RR), and latent variable regression (LVR) techniques, such as principal component regression (PCR), partial least squares (PLS) regression, and regularized canonical correlation analysis (RCCA). The theoretical analysis shows that the loading vectors used in LVR modeling can be computed by solving eigenvalue problems. Also, for the RCCA method, we show that by optimizing the regularization parameter, an improvement in prediction accuracy can be achieved over other modeling techniques. To illustrate the performances of all inferential modeling techniques, a comparative analysis was performed through two simulated examples, one using synthetic data and the other using simulated distillation column data. All techniques are optimized and compared by computing the cross validation mean square error using unseen testing data. The results of this comparative analysis show that scaling the data helps improve the performances of all modeling techniques, and that the LVR techniques outperform the full rank ones. One reason for this advantage is that the LVR techniques improve the conditioning of the model by discarding the latent variables (or principal components) with small eigenvalues, which also reduce the effect of the noise on the model prediction. The results also show that PCR and PLS have compara展开更多
在模糊C-均值聚类问题目标函数中使用正则化泛函,将聚类中心解的误差指标引入到模糊聚类的目标函数中,构造出新的模糊C-均值聚类算法RBFCM(Regularization based Fuzzy C-means)算法.算法RBFCM不仅具有较高的聚类精度,且计算结果具有更...在模糊C-均值聚类问题目标函数中使用正则化泛函,将聚类中心解的误差指标引入到模糊聚类的目标函数中,构造出新的模糊C-均值聚类算法RBFCM(Regularization based Fuzzy C-means)算法.算法RBFCM不仅具有较高的聚类精度,且计算结果具有更好的稳定性.进一步,将此RBFCM算法应用于基于T-S模糊模型的系统辨识问题.由于RBFCM算法优化了模糊系统的输入空间划分,提高了隶属度函数的精度,使得后继得到的T-S模糊系统辨识精度也有所提高,且系统辨识过程的收敛速度也有所改善.最后,通过对经典IRIS数据集、带有噪声的IRIS数据集的聚类算例和对Box-Jenkins煤气炉数据集进行辨识算例,验证了RBFCM算法的有效性和优越性.展开更多
Reconstruction of man-made scenes from multi-view images is an important problem in computer vision and computer graphics.Observing that manmade scenes are usually composed of planar surfaces,we encode plane shape pri...Reconstruction of man-made scenes from multi-view images is an important problem in computer vision and computer graphics.Observing that manmade scenes are usually composed of planar surfaces,we encode plane shape prior in reconstructing man-made scenes.Recent approaches for single-view reconstruction employ multi-branch neural networks to simultaneously segment planes and recover 3D plane parameters.However,the scale of available annotated data heavily limits the generalizability and accuracy of these supervised methods.In this paper,we propose multiview regularization to enhance the capability of piecewise planar reconstruction during the training phase,without demanding extra annotated data.Our multi-view regularization enables the consistency among multiple views by making the feature embedding more robust against view change and lighting variations.Thus,the neural network trained by multi-view regularization performs better on a wide range of views and lightings in the test phase.Based on more consistent prediction results,we merge the recovered models from multiple views to reconstruct scenes.Our approach achieves state-of-the-art reconstruction performance compared to previous approaches on the public Scan Net dataset.展开更多
Previous mining excavation in upper sublevels left several mined-out areas in Haigou gold mine. To ensure safety of the main and auxiliary shafts and mining production in deeper sublevels, systematical studies on regu...Previous mining excavation in upper sublevels left several mined-out areas in Haigou gold mine. To ensure safety of the main and auxiliary shafts and mining production in deeper sublevels, systematical studies on regularity, prediction, and control of ground pressure in the mine were carried out. Through 3D-numerical modeling and in-situ monitoring of acoustic emission, pressure and displacement, the ground pressure activity and the stability status of surrounding rock masses and the two shafts were assessed. Based on in-situ monitoring practice in Haigou mine,4 modes to judge rock stability according to the monitoring information of acoustic emission,pressure,and displacement were presented.展开更多
在化工过程的建模中,由于过程数据的高维度和高非线性,导致计算量大幅提升和建模难度加大。为了解决这一问题,提出了一种基于正则化方法的函数连接神经网络模型(regularization based functional link neural network, RFLNN)。所提出的...在化工过程的建模中,由于过程数据的高维度和高非线性,导致计算量大幅提升和建模难度加大。为了解决这一问题,提出了一种基于正则化方法的函数连接神经网络模型(regularization based functional link neural network, RFLNN)。所提出的RFLNN方法里,通过使用正则化的方法对函数连接神经网络的权值进行优化,一方面大幅降低网络计算复杂度和计算量,另一方面极大程度上克服网络局部极值和过拟合的问题,以提高函数连接神经网络的学习速度和精度。为了验证所提出方法的有效性,首先采用UCI数据中Real estate valuation数据对其性能进行测试;随后将所提的方法应用于高密度聚乙烯(high density polyethylene,HDPE)复杂生产过程进行建模。UCI标准数据与工业数据的仿真结果表明,与传统FLNN对比,RFLNN在处理高维复杂化工过程数据时具有收敛速度快、建模精度高等特点。展开更多
Regular vine copula provides rich models for dependence structure modeling.It combines vine structures and families of bivariate copulas to construct a number of multivariate distributions that can model a wide range ...Regular vine copula provides rich models for dependence structure modeling.It combines vine structures and families of bivariate copulas to construct a number of multivariate distributions that can model a wide range dependence patterns with different tail dependence for different pairs.Two special cases of regular vine copulas,C-vine and D-vine copulas,have been extensively investigated in the literature.We propose the Python package,pyvine,for modeling,sampling and testing a more generalized regular vine copula(R-vine for short).R-vine modeling algorithm searches for the R-vine structure which maximizes the vine tree dependence in a sequential way.The maximum likelihood estimation algorithm takes the sequential estimations as initial values and uses L-BFGS-B algorithm for the likelihood value optimization.R-vine sampling algorithm traverses all edges of the vine structure from the last tree in a recursive way and generates the marginal samples on each edge according to some nested conditions.Goodness-of-fit testing algorithm first generates Rosenblatt’s transformed data E and then tests the hypothesis H^(∗)_(0):E∼C_(⊥)by using Anderson–Darling statistic,where C_(⊥)is the independence copula.Bootstrap method is used to compute an adjusted p-value of the empirical distribution of replications of Anderson–Darling statistic.The computing of related functions of copulas such as cumulative distribution functions,Hfunctions and inverse H-functions often meets with the problem of overflow.We solve this problem by reinvestigating the following six families of bivariate copulas:Normal,Student t,Clayton,Gumbel,Frank and Joe’s copulas.Approximations of the above related functions of copulas are given when the overflow occurs in the computation.All these are implemented in a subpackage bvcopula,in which subroutines are written in Fortran and wrapped into Python and,hence,good performance is guaranteed.展开更多
Because it is hard to search similar structure for low similarity unknown structure proteins directly from the Protein Data Bank （PDB） database, 3D-structure is modeled in this paper for secondary structure regular ...Because it is hard to search similar structure for low similarity unknown structure proteins directly from the Protein Data Bank （PDB） database, 3D-structure is modeled in this paper for secondary structure regular fragments （α-Helices, β-Strands） of such proteins by the protein secondary structure prediction software, the Basic Local Alignment Search Tool （BLAST） and the side chain construction software SCWRL3. First, the protein secondary structure prediction software is adopted to extract secondary structure fragments from the unknown structure proteins. Then, regular fragments are regulated by BLAST based on comparative modeling, providing main chain configurations. Finally, SCWRL3 is applied to assemble side chains for regular fragments, so that 3D-structure of regular fragments of low similarity unknown structure protein is obtained. Regular fragments of several neurotoxins are used for test. Simulation results show that the prediction errors are less than 0.06nm for regular fragments less than 10 amino acids, implying the simpleness and effectiveness of the proposed method.展开更多
文摘Inferential models are widely used in the chemical industry to infer key process variables, which are challenging or expensive to measure, from other more easily measured variables. The aim of this paper is three-fold: to present a theoretical review of some of the well known linear inferential modeling techniques, to enhance the predictive ability of the regularized canonical correlation analysis (RCCA) method, and finally to compare the performances of these techniques and highlight some of the practical issues that can affect their predictive abilities. The inferential modeling techniques considered in this study include full rank modeling techniques, such as ordinary least square (OLS) regression and ridge regression (RR), and latent variable regression (LVR) techniques, such as principal component regression (PCR), partial least squares (PLS) regression, and regularized canonical correlation analysis (RCCA). The theoretical analysis shows that the loading vectors used in LVR modeling can be computed by solving eigenvalue problems. Also, for the RCCA method, we show that by optimizing the regularization parameter, an improvement in prediction accuracy can be achieved over other modeling techniques. To illustrate the performances of all inferential modeling techniques, a comparative analysis was performed through two simulated examples, one using synthetic data and the other using simulated distillation column data. All techniques are optimized and compared by computing the cross validation mean square error using unseen testing data. The results of this comparative analysis show that scaling the data helps improve the performances of all modeling techniques, and that the LVR techniques outperform the full rank ones. One reason for this advantage is that the LVR techniques improve the conditioning of the model by discarding the latent variables (or principal components) with small eigenvalues, which also reduce the effect of the noise on the model prediction. The results also show that PCR and PLS have compara
文摘在模糊C-均值聚类问题目标函数中使用正则化泛函,将聚类中心解的误差指标引入到模糊聚类的目标函数中,构造出新的模糊C-均值聚类算法RBFCM(Regularization based Fuzzy C-means)算法.算法RBFCM不仅具有较高的聚类精度,且计算结果具有更好的稳定性.进一步,将此RBFCM算法应用于基于T-S模糊模型的系统辨识问题.由于RBFCM算法优化了模糊系统的输入空间划分,提高了隶属度函数的精度,使得后继得到的T-S模糊系统辨识精度也有所提高,且系统辨识过程的收敛速度也有所改善.最后,通过对经典IRIS数据集、带有噪声的IRIS数据集的聚类算例和对Box-Jenkins煤气炉数据集进行辨识算例,验证了RBFCM算法的有效性和优越性.
基金supported by the National Key R&D Program of China under Grant 2017YFB1002202the National Natural Science Foundation of China(NSFC)under Grant 61632006the Fundamental Research Funds for the Central Universities under Grants WK3490000003 and WK2100100030.
文摘Reconstruction of man-made scenes from multi-view images is an important problem in computer vision and computer graphics.Observing that manmade scenes are usually composed of planar surfaces,we encode plane shape prior in reconstructing man-made scenes.Recent approaches for single-view reconstruction employ multi-branch neural networks to simultaneously segment planes and recover 3D plane parameters.However,the scale of available annotated data heavily limits the generalizability and accuracy of these supervised methods.In this paper,we propose multiview regularization to enhance the capability of piecewise planar reconstruction during the training phase,without demanding extra annotated data.Our multi-view regularization enables the consistency among multiple views by making the feature embedding more robust against view change and lighting variations.Thus,the neural network trained by multi-view regularization performs better on a wide range of views and lightings in the test phase.Based on more consistent prediction results,we merge the recovered models from multiple views to reconstruct scenes.Our approach achieves state-of-the-art reconstruction performance compared to previous approaches on the public Scan Net dataset.
基金the National Key Technologies R&D Program of China (No. 2006BAK04B02)
文摘Previous mining excavation in upper sublevels left several mined-out areas in Haigou gold mine. To ensure safety of the main and auxiliary shafts and mining production in deeper sublevels, systematical studies on regularity, prediction, and control of ground pressure in the mine were carried out. Through 3D-numerical modeling and in-situ monitoring of acoustic emission, pressure and displacement, the ground pressure activity and the stability status of surrounding rock masses and the two shafts were assessed. Based on in-situ monitoring practice in Haigou mine,4 modes to judge rock stability according to the monitoring information of acoustic emission,pressure,and displacement were presented.
文摘在化工过程的建模中,由于过程数据的高维度和高非线性,导致计算量大幅提升和建模难度加大。为了解决这一问题,提出了一种基于正则化方法的函数连接神经网络模型(regularization based functional link neural network, RFLNN)。所提出的RFLNN方法里,通过使用正则化的方法对函数连接神经网络的权值进行优化,一方面大幅降低网络计算复杂度和计算量,另一方面极大程度上克服网络局部极值和过拟合的问题,以提高函数连接神经网络的学习速度和精度。为了验证所提出方法的有效性,首先采用UCI数据中Real estate valuation数据对其性能进行测试;随后将所提的方法应用于高密度聚乙烯(high density polyethylene,HDPE)复杂生产过程进行建模。UCI标准数据与工业数据的仿真结果表明,与传统FLNN对比,RFLNN在处理高维复杂化工过程数据时具有收敛速度快、建模精度高等特点。
基金This work was supported by the NNSF of China(Nos.11371340,71871208).
文摘Regular vine copula provides rich models for dependence structure modeling.It combines vine structures and families of bivariate copulas to construct a number of multivariate distributions that can model a wide range dependence patterns with different tail dependence for different pairs.Two special cases of regular vine copulas,C-vine and D-vine copulas,have been extensively investigated in the literature.We propose the Python package,pyvine,for modeling,sampling and testing a more generalized regular vine copula(R-vine for short).R-vine modeling algorithm searches for the R-vine structure which maximizes the vine tree dependence in a sequential way.The maximum likelihood estimation algorithm takes the sequential estimations as initial values and uses L-BFGS-B algorithm for the likelihood value optimization.R-vine sampling algorithm traverses all edges of the vine structure from the last tree in a recursive way and generates the marginal samples on each edge according to some nested conditions.Goodness-of-fit testing algorithm first generates Rosenblatt’s transformed data E and then tests the hypothesis H^(∗)_(0):E∼C_(⊥)by using Anderson–Darling statistic,where C_(⊥)is the independence copula.Bootstrap method is used to compute an adjusted p-value of the empirical distribution of replications of Anderson–Darling statistic.The computing of related functions of copulas such as cumulative distribution functions,Hfunctions and inverse H-functions often meets with the problem of overflow.We solve this problem by reinvestigating the following six families of bivariate copulas:Normal,Student t,Clayton,Gumbel,Frank and Joe’s copulas.Approximations of the above related functions of copulas are given when the overflow occurs in the computation.All these are implemented in a subpackage bvcopula,in which subroutines are written in Fortran and wrapped into Python and,hence,good performance is guaranteed.
基金Sponsored by the National Natural Science Foundation of China （60374069） and the Excellent Young Scholars Research Fund of Beijing Institute of Technology （000Y01-3）.
文摘Because it is hard to search similar structure for low similarity unknown structure proteins directly from the Protein Data Bank （PDB） database, 3D-structure is modeled in this paper for secondary structure regular fragments （α-Helices, β-Strands） of such proteins by the protein secondary structure prediction software, the Basic Local Alignment Search Tool （BLAST） and the side chain construction software SCWRL3. First, the protein secondary structure prediction software is adopted to extract secondary structure fragments from the unknown structure proteins. Then, regular fragments are regulated by BLAST based on comparative modeling, providing main chain configurations. Finally, SCWRL3 is applied to assemble side chains for regular fragments, so that 3D-structure of regular fragments of low similarity unknown structure protein is obtained. Regular fragments of several neurotoxins are used for test. Simulation results show that the prediction errors are less than 0.06nm for regular fragments less than 10 amino acids, implying the simpleness and effectiveness of the proposed method.
