The variable selection of high dimensional nonparametric nonlinear systems aims to select the contributing variables or to eliminate the redundant variables.For a high dimensional nonparametric nonlinear system,howeve...The variable selection of high dimensional nonparametric nonlinear systems aims to select the contributing variables or to eliminate the redundant variables.For a high dimensional nonparametric nonlinear system,however,identifying whether a variable contributes or not is not easy.Therefore,based on the Fourier spectrum of densityweighted derivative,one novel variable selection approach is developed,which does not suffer from the dimensionality curse and improves the identification accuracy.Furthermore,a necessary and sufficient condition for testing a variable whether it contributes or not is provided.The proposed approach does not require strong assumptions on the distribution,such as elliptical distribution.The simulation study verifies the effectiveness of the novel variable selection algorithm.展开更多
Deep learning algorithms based on neural networks make remarkable achievements in machine fault diagnosis,while the noise mixed in measured signals harms the prediction accuracy of networks.Existing denoising methods ...Deep learning algorithms based on neural networks make remarkable achievements in machine fault diagnosis,while the noise mixed in measured signals harms the prediction accuracy of networks.Existing denoising methods in neural networks,such as using complex network architectures and introducing sparse techniques,always suffer from the difficulty of estimating hyperparameters and the lack of physical interpretability.To address this issue,this paper proposes a novel interpretable denoising layer based on reproducing kernel Hilbert space(RKHS)as the first layer for standard neural networks,with the aim to combine the advantages of both traditional signal processing technology with physical interpretation and network modeling strategy with parameter adaption.By investigating the influencing mechanism of parameters on the regularization procedure in RKHS,the key parameter that dynamically controls the signal smoothness with low computational cost is selected as the only trainable parameter of the proposed layer.Besides,the forward and backward propagation algorithms of the designed layer are formulated to ensure that the selected parameter can be automatically updated together with other parameters in the neural network.Moreover,exponential and piecewise functions are introduced in the weight updating process to keep the trainable weight within a reasonable range and avoid the ill-conditioned problem.Experiment studies verify the effectiveness and compatibility of the proposed layer design method in intelligent fault diagnosis of machinery in noisy environments.展开更多
The exploration of low-cost and metal-free nanozymes with oxidase-mimicking activity is highly desired due to their attractive properties and potential applications.However,it is still challenging and remains unexploi...The exploration of low-cost and metal-free nanozymes with oxidase-mimicking activity is highly desired due to their attractive properties and potential applications.However,it is still challenging and remains unexploited to fully realize oxidase-like nanozyme in the emerging covalent organic frameworks(COFs)due to their polymeric nature and weak photoelectric activity.We herein report the first example of the preparation and oxidase-mimicking activity of novel ultrathin two-dimensional(2D)COF(termed as TTPA-COF)nanosheets.The ultrathin TTPA-COF nanosheets with hexagonal layered structure are constructed from two flexible photoactive(diarylamino)benzene-based linkers,and exhibit remarkable catalytic activity toward the oxidation of 3,3',5,5'-tetramethylbenzidine(TMB)in the presence of O_(2) due to their large specific surface areas and abundant active sites.Moreover,it is worth noting that the nanozyme activity could be regulated by external light irradiation.Based on the oxidasemimicking activity of TTPA-COF nanosheets,a green colorimetric sensor is proposed for the sensitive and selective determination of glutathione(GSH)in a wide linear range of 0.5–40μM with a detection limit of 0.5μM.This work reported here would open new avenues for the exploration of low-cost and high-efficiency nanozymes,as well as extend the application of 2D COF nanosheets in the fields of catalysis and sensing.展开更多
基金Project supported by the National Key Research and Development Program of China(No.2021YFB3400700)the National Natural Science Foundation of China(Nos.12422201,12072188,12121002,and 12372017)。
文摘The variable selection of high dimensional nonparametric nonlinear systems aims to select the contributing variables or to eliminate the redundant variables.For a high dimensional nonparametric nonlinear system,however,identifying whether a variable contributes or not is not easy.Therefore,based on the Fourier spectrum of densityweighted derivative,one novel variable selection approach is developed,which does not suffer from the dimensionality curse and improves the identification accuracy.Furthermore,a necessary and sufficient condition for testing a variable whether it contributes or not is provided.The proposed approach does not require strong assumptions on the distribution,such as elliptical distribution.The simulation study verifies the effectiveness of the novel variable selection algorithm.
基金Supported by National Natural Science Foundation of China(Grant Nos.12072188,11632011,11702171,11572189,51121063)Shanghai Municipal Natural Science Foundation of China(Grant No.20ZR1425200).
文摘Deep learning algorithms based on neural networks make remarkable achievements in machine fault diagnosis,while the noise mixed in measured signals harms the prediction accuracy of networks.Existing denoising methods in neural networks,such as using complex network architectures and introducing sparse techniques,always suffer from the difficulty of estimating hyperparameters and the lack of physical interpretability.To address this issue,this paper proposes a novel interpretable denoising layer based on reproducing kernel Hilbert space(RKHS)as the first layer for standard neural networks,with the aim to combine the advantages of both traditional signal processing technology with physical interpretation and network modeling strategy with parameter adaption.By investigating the influencing mechanism of parameters on the regularization procedure in RKHS,the key parameter that dynamically controls the signal smoothness with low computational cost is selected as the only trainable parameter of the proposed layer.Besides,the forward and backward propagation algorithms of the designed layer are formulated to ensure that the selected parameter can be automatically updated together with other parameters in the neural network.Moreover,exponential and piecewise functions are introduced in the weight updating process to keep the trainable weight within a reasonable range and avoid the ill-conditioned problem.Experiment studies verify the effectiveness and compatibility of the proposed layer design method in intelligent fault diagnosis of machinery in noisy environments.
基金supported by the financial support from the National Natural Science Foundation of China(NSFC)(Nos.21976166,22006122,and 21405144)the research funding of“Leading Innovative and Entrepreneur Team Introduction Program of Zhejiang”(No.2020R01002)+4 种基金the Class D of Qianjiang Talent Program(No.ZD20011250001)the Science Challenge Project(No.TZ2016004)he Project of State Key Laboratory of Environment-friendly Energy Materials(No.18zd320303)the Scientific Research Starting Foundation for Returned Overseas Chinese Scholars of Sichuan Province(No.19zd3200)support from the Opening Project of Jiangsu Province Engineering Research Center of Agricultural Breeding Pollution Control and Resource(No.2021ABPCR004).
文摘The exploration of low-cost and metal-free nanozymes with oxidase-mimicking activity is highly desired due to their attractive properties and potential applications.However,it is still challenging and remains unexploited to fully realize oxidase-like nanozyme in the emerging covalent organic frameworks(COFs)due to their polymeric nature and weak photoelectric activity.We herein report the first example of the preparation and oxidase-mimicking activity of novel ultrathin two-dimensional(2D)COF(termed as TTPA-COF)nanosheets.The ultrathin TTPA-COF nanosheets with hexagonal layered structure are constructed from two flexible photoactive(diarylamino)benzene-based linkers,and exhibit remarkable catalytic activity toward the oxidation of 3,3',5,5'-tetramethylbenzidine(TMB)in the presence of O_(2) due to their large specific surface areas and abundant active sites.Moreover,it is worth noting that the nanozyme activity could be regulated by external light irradiation.Based on the oxidasemimicking activity of TTPA-COF nanosheets,a green colorimetric sensor is proposed for the sensitive and selective determination of glutathione(GSH)in a wide linear range of 0.5–40μM with a detection limit of 0.5μM.This work reported here would open new avenues for the exploration of low-cost and high-efficiency nanozymes,as well as extend the application of 2D COF nanosheets in the fields of catalysis and sensing.
