Diagnosing the cardiovascular disease is one of the biggest medical difficulties in recent years.Coronary cardiovascular(CHD)is a kind of heart and blood vascular disease.Predicting this sort of cardiac illness leads ...Diagnosing the cardiovascular disease is one of the biggest medical difficulties in recent years.Coronary cardiovascular(CHD)is a kind of heart and blood vascular disease.Predicting this sort of cardiac illness leads to more precise decisions for cardiac disorders.Implementing Grid Search Optimization(GSO)machine training models is therefore a useful way to forecast the sickness as soon as possible.The state-of-the-art work is the tuning of the hyperparameter together with the selection of the feature by utilizing the model search to minimize the false-negative rate.Three models with a cross-validation approach do the required task.Feature Selection based on the use of statistical and correlation matrices for multivariate analysis.For Random Search and Grid Search models,extensive comparison findings are produced utilizing retrieval,F1 score,and precision measurements.The models are evaluated using the metrics and kappa statistics that illustrate the three models’comparability.The study effort focuses on optimizing function selection,tweaking hyperparameters to improve model accuracy and the prediction of heart disease by examining Framingham datasets using random forestry classification.Tuning the hyperparameter in the model of grid search thus decreases the erroneous rate achieves global optimization.展开更多
Deep learning algorithm is an effective data mining method and has been used in many fields to solve practical problems.However,the deep learning algorithms often contain some hyper-parameters which may be continuous,...Deep learning algorithm is an effective data mining method and has been used in many fields to solve practical problems.However,the deep learning algorithms often contain some hyper-parameters which may be continuous,integer,or mixed,and are often given based on experience but largely affect the effectiveness of activity recognition.In order to adapt to different hyper-parameter optimization problems,our improved Cuckoo Search(CS)algorithm is proposed to optimize the mixed hyper-parameters in deep learning algorithm.The algorithm optimizes the hyper-parameters in the deep learning model robustly,and intelligently selects the combination of integer type and continuous hyper-parameters that make the model optimal.Then,the mixed hyper-parameter in Convolutional Neural Network(CNN),Long-Short-Term Memory(LSTM)and CNN-LSTM are optimized based on the methodology on the smart home activity recognition datasets.Results show that the methodology can improve the performance of the deep learning model and whether we are experienced or not,we can get a better deep learning model using our method.展开更多
A theoretical methodology is suggested for finding the malaria parasites’presence with the help of an intelligent hyper-parameter tuned Deep Learning(DL)based malaria parasite detection and classification(HPTDL-MPDC)...A theoretical methodology is suggested for finding the malaria parasites’presence with the help of an intelligent hyper-parameter tuned Deep Learning(DL)based malaria parasite detection and classification(HPTDL-MPDC)in the smear images of human peripheral blood.Some existing approaches fail to predict the malaria parasitic features and reduce the prediction accuracy.The trained model initiated in the proposed system for classifying peripheral blood smear images into the non-parasite or parasite classes using the available online dataset.The Adagrad optimizer is stacked with the suggested pre-trained Deep Neural Network(DNN)with the help of the contrastive divergence method to pre-train.The features are extracted from the images in the proposed system to train the DNN for initializing the visible variables.The smear images show the concatenated feature to be utilized as the feature vector in the proposed system.Lastly,hyper-parameters are used to fine-tune DNN to calculate the class labels’probability.The suggested system outperforms more modern methodologies with an accuracy of 91%,precision of 89%,recall of 93%and F1-score of 91%.The HPTDL-MPDC has the primary application in detecting the parasite of malaria in the smear images of human peripheral blood.展开更多
High-rise buildings are usually considered as flexible structures with low inherent damping. Therefore, these kinds of buildings are susceptible to wind-induced vibration. Tuned Mass Damper（TMD） can be used as an ef...High-rise buildings are usually considered as flexible structures with low inherent damping. Therefore, these kinds of buildings are susceptible to wind-induced vibration. Tuned Mass Damper（TMD） can be used as an effective device to mitigate excessive vibrations. In this study, Artificial Neural Networks is used to find optimal mechanical properties of TMD for high-rise buildings subjected to wind load. The patterns obtained from structural analysis of different multi degree of freedom（MDF） systems are used for training neural networks. In order to obtain these patterns, structural models of some systems with 10 to 80 degrees-of-freedoms are built in MATLAB/SIMULINK program. Finally, the optimal properties of TMD are determined based on the objective of maximum displacement response reduction. The Auto-Regressive model is used to simulate the wind load. In this way, the uncertainties related to wind loading can be taken into account in neural network’s outputs. After training the neural network, it becomes possible to set the frequency and TMD mass ratio as inputs and get the optimal TMD frequency and damping ratio as outputs. As a case study, a benchmark 76-story office building is considered and the presented procedure is used to obtain optimal characteristics of the TMD for the building.展开更多
Due to outstanding performance in cheminformatics,machine learning algorithms have been increasingly used to mine molecular properties and biomedical big data.The performance of machine learning models is known to cri...Due to outstanding performance in cheminformatics,machine learning algorithms have been increasingly used to mine molecular properties and biomedical big data.The performance of machine learning models is known to critically depend on the selection of the hyper-parameter configuration.However,many studies either explored the optimal hyper-parameters per the grid searching method or employed arbitrarily selected hyper-parameters,which can easily lead to achieving a suboptimal hyper-parameter configuration.In this study,Hyperopt library embedding with the Bayesian optimization is employed to find optimal hyper-parameters for different machine learning algorithms.Six drug discovery datasets,including solubility,probe-likeness,h ERG,Chagas disease,tuberculosis,and malaria,are used to compare different machine learning algorithms with ECFP6 fingerprints.This contribution aims to evaluate whether the Bernoulli Na?ve Bayes,logistic linear regression,Ada Boost decision tree,random forest,support vector machine,and deep neural networks algorithms with optimized hyper-parameters can offer any improvement in testing as compared with the referenced models assessed by an array of metrics including AUC,F1-score,Cohen’s kappa,Matthews correlation coefficient,recall,precision,and accuracy.Based on the rank normalized score approach,the Hyperopt models achieve better or comparable performance on 33 out 36 models for different drug discovery datasets,showing significant improvement achieved by employing the Hyperopt library.The open-source code of all the 6 machine learning frameworks employed in the Hyperopt python package is provided to make this approach accessible to more scientists,who are not familiar with writing code.展开更多
Regularized system identification has become the research frontier of system identification in the past decade.One related core subject is to study the convergence properties of various hyper-parameter estimators as t...Regularized system identification has become the research frontier of system identification in the past decade.One related core subject is to study the convergence properties of various hyper-parameter estimators as the sample size goes to infinity.In this paper,we consider one commonly used hyper-parameter estimator,the empirical Bayes(EB).Its convergence in distribution has been studied,and the explicit expression of the covariance matrix of its limiting distribution has been given.However,what we are truly interested in are factors contained in the covariance matrix of the EB hyper-parameter estimator,and then,the convergence of its covariance matrix to that of its limiting distribution is required.In general,the convergence in distribution of a sequence of random variables does not necessarily guarantee the convergence of its covariance matrix.Thus,the derivation of such convergence is a necessary complement to our theoretical analysis about factors that influence the convergence properties of the EB hyper-parameter estimator.In this paper,we consider the regularized finite impulse response(FIR)model estimation with deterministic inputs,and show that the covariance matrix of the EB hyper-parameter estimator converges to that of its limiting distribution.Moreover,we run numerical simulations to demonstrate the efficacy of ourtheoretical results.展开更多
In this paper,a novel four-prong quartz tuning fork(QTF)was designed with enlarged deformation area,large prong gap,and low resonant frequency to improve its performance in laser spectroscopy sensing.A theoretical sim...In this paper,a novel four-prong quartz tuning fork(QTF)was designed with enlarged deformation area,large prong gap,and low resonant frequency to improve its performance in laser spectroscopy sensing.A theoretical simulation model was established to optimize the design of the QTF structure.In the simulation of quartz-enhanced photoacoustic spectroscopy(QEPAS)technology,the maximum stress and the surface charge density of the four-prong QTF demonstrated increases of 11.1-fold and 15.9-fold,respectively,compared to that of the standard two-prong QTF.In the simulation of light-induced thermoelastic spectroscopy(LITES)technology,the surface temperature difference of the four-prong QTF was found to be 11.4 times greater than that of the standard QTF.Experimental results indicated that the C_(2)H_(2)-QEPAS system based on this innovative design improved the signal-to-noise-ratio(SNR)by 4.67 times compared with the standard QTF-based system,and the SNR could increase up to 147.72 times when the four-prong QTF was equipped with its optimal acoustic micro-resonator(AmR).When the average time of the system reached 370 s,the system achieved a MDL as low as 21 ppb.The four-prong QTF-based C_(2)H_(2)-LITES system exhibited a SNR improvement by a factor of 4.52,and a MDL of 96 ppb was obtained when the average time of the system reached 100 s.The theoretical and experimental results effectively demonstrated the superiority of the four-prong QTF in the field of laser spectroscopy sensing.展开更多
Hydroxyapatite nanoparticles(HAP NPs)were synthesized by a one‐step hydrothermal method.The surface of HAP NPs was grafted-SH and-COOH chelating groups via in situ surface‐modification with iminodiacetic acid(IDA)an...Hydroxyapatite nanoparticles(HAP NPs)were synthesized by a one‐step hydrothermal method.The surface of HAP NPs was grafted-SH and-COOH chelating groups via in situ surface‐modification with iminodiacetic acid(IDA)and 3‐mercaptopropyl trimethoxysilane(MPS)to afford dual surface‐capped nano‐amendment HAPIDA/MPS.The structure of HAP‐IDA/MPS was characterized,and its adsorption performance for Hg^(2+),Cu^(2+),Zn^(2+),Ni^(2+),Co^(2+),and Cd^(2+)was evaluated.The total adsorption capacity of 0.10 g HAP‐IDA/MPS nano‐amendment for Hg^(2+),Cu^(2+),Zn^(2+),Ni^(2+),Co^(2+),and Cd^(2+)with an initial mass concentration of 20 mg·L^(-1) reached 13.7 mg·g^(-1),about 4.3 times as much as that of HAP.Notably,HAP‐IDA/MPS nano‐amendment displayed the highest immobilization rate for Hg^(2+),possibly because of its chemical reaction with-SH to form sulfide,possessing the lowest solubility product constant among a variety of metal sulfides.展开更多
The next generation of synchrotron radiation light sources features extremely low emittance,enabling the generation of synchrotron radiation with significantly higher brilliance,which facilitates the exploration of ma...The next generation of synchrotron radiation light sources features extremely low emittance,enabling the generation of synchrotron radiation with significantly higher brilliance,which facilitates the exploration of matter at smaller scales.However,the extremely low emittance results in stronger sextupole magnet strengths,leading to high natural chromaticity.This necessitates the use of sextupole magnets to correct the natural chromaticity.For the Shanghai Synchrotron Radiation Facility Upgrade(SSRF-U),a lattice was designed for the storage ring that can achieve an ultra-low natural emittance of 72.2 pm·rad at the beam energy of 3.5 GeV.However,the significant detuning effects,driven by high second-order resonant driving terms due to strong sextupoles,will degrade the performance of the facility.To resolve this issue,installation of octupoles in the SSRF-U storage ring has been planned.This paper presents the study results on configuration selection and optimization method for the octupoles.An optimal solution for the SSRF-U storage ring was obtained to effectively mitigate the amplitude-dependent tune shift and the second-order chromaticity,consequently leading to an increased dynamic aperture(DA),momentum acceptance(MA),and reduced sensitivity to magnetic field errors.展开更多
Multiple tuned mass dampers(MTMDs)reduce dynamic response with multiple specified frequencies of building structures.Many optimization algorithms for placement design exist,though they rarely conform to code-based ver...Multiple tuned mass dampers(MTMDs)reduce dynamic response with multiple specified frequencies of building structures.Many optimization algorithms for placement design exist,though they rarely conform to code-based verification nor produce high quality solutions without high computational effort and high complexity.This study proposes an inverse element exchange method(IEEM)with multi-level programming and compares it to a single tuned mass damper(STMD)and uniform distribution of multiple tuned mass dampers in the frequency and time domains.A ten-story shear building is used for the numerical case study.The results show that the proposed method can offer improvement over the STMD,uniform distribution of multiple tuned mass dampers,and distribution optimized by genetic algorithms(GA)with regard to minimizing the interstory drift ratio(IDR)in both the frequency and time domains and the time consumption for optimization.展开更多
Under the perspective of translation aesthetics,the article studies three Chinese-to-English translations of Slow,Slow Tune as research objects,comparing and contrasting the differences between the different translati...Under the perspective of translation aesthetics,the article studies three Chinese-to-English translations of Slow,Slow Tune as research objects,comparing and contrasting the differences between the different translations under the perspective of translation aesthetics.The study finds that:Kenneth Rexroth’s translation is poor in textual understanding,with a lot of errors,and is too shallow in emotional expression,remaining only on the surface of the text;Lin Yutang makes occasional mistakes in understanding individual words,and is able to explore the implicit feelings of the lyricist,and his translation focuses on textual and phonological beauty;Xu Yuanchong’s translation is not only capable of conveying the meaning of the original work,but also of conveying the lyricist’s feelings more vividly and focusing on the phonological beauty.Xu Yuanchong’s translation not only conveys the meaning of the original work,but also vividly conveys the emotion of the lyricist,and pays attention to the beauty of sound.展开更多
We present a gain adaptive tuning method for fiber Raman amplifier(FRA) using two-stage neural networks(NNs) and double weights updates. After training the connection weights of two-stage NNs separately in training ph...We present a gain adaptive tuning method for fiber Raman amplifier(FRA) using two-stage neural networks(NNs) and double weights updates. After training the connection weights of two-stage NNs separately in training phase, the connection weights of the unified NN are updated again in verification phase according to error between the predicted and target gains to eliminate the inherent error of the NNs. The simulation results show that the mean of root mean square error(RMSE) and maximum error of gains are 0.131 d B and 0.281 d B, respectively. It shows that the method can realize adaptive adjustment function of FRA gain with high accuracy.展开更多
Lanthanide-sensitized upconverting nanoparticles(UCNPs)are widely studied because of their unusual optical characteristics,such as large antenna-generated anti-Stokes shifts,high photostability,and narrow emission ban...Lanthanide-sensitized upconverting nanoparticles(UCNPs)are widely studied because of their unusual optical characteristics,such as large antenna-generated anti-Stokes shifts,high photostability,and narrow emission bandwidths,which can be harnessed for a variety of applications including bioimaging,sensing,information security and high-level anticounterfeiting.The diverse requirements of these applications typically require precise control over upconversion luminescence(UCL).Recently,the concept of energy migration upconversion has emerged as an effective approach to modulate UCL for various lanthanide ions.Moreover,it provides valuable insights into the fundamental comprehension of energy transfer mechanisms on the nanoscale,thereby contributing to the design of efficient lanthanide-sensitized UCNPs and their practical applications.Here we present a comprehensive overview of the latest developments in energy migration upconversion in lanthanide-sensitized nanoparticles for photon upconversion tuning,encompassing design strategies,mechanistic investigations and applications.Additionally,some future prospects in the field of energy migration upconversion are also discussed.展开更多
Grooved tuning forks with hierarchical structures have become some of the most widely used piezoelectric quartz microelectromechanical system devices;however,fabricating these devices requires multi-step processes due...Grooved tuning forks with hierarchical structures have become some of the most widely used piezoelectric quartz microelectromechanical system devices;however,fabricating these devices requires multi-step processes due to the complexity of etching of quartz,particularly in specific orientations of the crystal lattice.This paper proposes a one-step fabrication strategy that can form a complete hierarchical structure with only a single etching process using novel lithography patterns.The core principle of this strategy is based on the effect of the size of the groove patterns on quartz etching,whereby trenches of varying depths can be created in a fixed etching time by adjusting the width of the hard mask.Specifically,the device outline and grooved structure can be completed using a seamlessly designed etching pattern and optimized time.Furthermore,the etching structure itself influences the etching results.It was found that dividing a wide trench by including a wall to separate it into two narrow trenches significantly reduces the etching rate,allowing for predictable tuning of the etching rate for wider grooves.This effectively increases the usability and flexibility of the one-step strategy.This was applied to the manufacture of an ultra-small quartz grooved tuning fork resonator with a frequency of 32.768 kHz in a single step,increasing production efficiency by almost 45%and reducing costs by almost 30%compared to current methods.This has great potential for improving the productivity of grooved tuning fork devices.It can also be extended to the fabrication of other quartz crystal devices requiring hierarchical structures.展开更多
P2-type layered oxide Na_(2/3)Ni_(1/3)Mn_(2/3)O_(2)(NM)is a promising cathode material for sodium-ion batteries(SIBs).However,the severe irreversible phase transition,sluggish Na+diffusion kinetics,and interfacial sid...P2-type layered oxide Na_(2/3)Ni_(1/3)Mn_(2/3)O_(2)(NM)is a promising cathode material for sodium-ion batteries(SIBs).However,the severe irreversible phase transition,sluggish Na+diffusion kinetics,and interfacial side reactions at high-voltage result in grievous capacity degradation and inferior electrochemical performance.Herein,a dual-function strategy of entropy tuning and artificial cathode electrolyte interface(CEI)layer construction is reported to generate a novel P2-type medium-entropy Na_(0.75)Li_(0.1)Mg_(0.05)Ni_(0.18)Mn_(0.66)Ta_(0.01)O_(2)with NaTaO_(3)surface modification(LMNMT)to address the aforementioned issues.In situ X-ray diffraction reveals that LMNMT exhibits a near zero-strain phase transition with a volume change of only 1.4%,which is significantly lower than that of NM(20.9%),indicating that entropy tuning effectively suppresses irreversible phase transitions and enhances ion diffusion.Kinetic analysis and post-cycling interfacial characterization further confirm that the artificial CEI layer promotes the formation of a stable,thin NaF-rich CEI and reduces interfacial side reactions,thereby further enhancing ion transport kinetics and surface/interface stability.Consequently,the LMNMT electrode exhibits outstanding rate capability(46 mA h g^(−1)at 20 C)and cycling stability(89.5%capacity retention after 200 cycles at 2 C)within the voltage range of 2–4.35 V.The LMNMT also exhibits superior all-climate performance and air stability.This study provides a novel path for the design of high-voltage cathode materials for SIBs.展开更多
Hydrogenation catalysts frequently impose a compromise between activity and selectivity,where maximizing one property inevitably diminishes the other.Researchers from the Dalian Institute of Chemical Physics(DICP)of t...Hydrogenation catalysts frequently impose a compromise between activity and selectivity,where maximizing one property inevitably diminishes the other.Researchers from the Dalian Institute of Chemical Physics(DICP)of the Chinese Academy of Sciences,in collaboration with scholars from University of Science and Technology of China and the Karlsruhe Institute of Technology in Germany,cracked this dilemma by engineering bimetallic catalysts with atomic precision-a breakthrough that boosts hydrogenation efficiency by 35-fold while maintaining pinpoint accuracy,resolving the stubborn activity-selectivity paradox.展开更多
Fire can cause significant damage to the environment,economy,and human lives.If fire can be detected early,the damage can be minimized.Advances in technology,particularly in computer vision powered by deep learning,ha...Fire can cause significant damage to the environment,economy,and human lives.If fire can be detected early,the damage can be minimized.Advances in technology,particularly in computer vision powered by deep learning,have enabled automated fire detection in images and videos.Several deep learning models have been developed for object detection,including applications in fire and smoke detection.This study focuses on optimizing the training hyperparameters of YOLOv8 andYOLOv10models usingBayesianTuning(BT).Experimental results on the large-scale D-Fire dataset demonstrate that this approach enhances detection performance.Specifically,the proposed approach improves the mean average precision at an Intersection over Union(IoU)threshold of 0.5(mAP50)of the YOLOv8s,YOLOv10s,YOLOv8l,and YOLOv10lmodels by 0.26,0.21,0.84,and 0.63,respectively,compared tomodels trainedwith the default hyperparameters.The performance gains are more pronounced in larger models,YOLOv8l and YOLOv10l,than in their smaller counterparts,YOLOv8s and YOLOv10s.Furthermore,YOLOv8 models consistently outperform YOLOv10,with mAP50 improvements of 0.26 for YOLOv8s over YOLOv10s and 0.65 for YOLOv8l over YOLOv10l when trained with BT.These results establish YOLOv8 as the preferred model for fire detection applications where detection performance is prioritized.展开更多
Separation of ternary C_(4) olefins(n-butene,iso-butene and 1,3-butadiene)is very challenging but crucial in the petrol-chemical industry due to their similar molecular sizes and properties.Herein,to optimize the sepa...Separation of ternary C_(4) olefins(n-butene,iso-butene and 1,3-butadiene)is very challenging but crucial in the petrol-chemical industry due to their similar molecular sizes and properties.Herein,to optimize the separation efficiency for separation of C_(4) olefins,a new Hofmann-type MOF,[Ni(piz)Ni(CN)_(4)](piz=piperazine)-isostructural to the typical one[Ni(pyz)Ni(CN)_(4)](pyz=pyrazine),has been synthesized by a facile method from aqueous solution.The pore size reduction of[Ni(piz)Ni(CN)_(4)](3.62A,in contrast to 3.85A in[Ni(pyz)Ni(CN)_(4)])results in negligible iso-butene(i-C_(4)H_(8))uptake(from 2.92 to 0.04 mmol g^(-1))whereas retaining significant uptake for 1,3-butadiene(1,3-C_(4)H_(6),1.96 mmol g^(-1))and n-butene(n-C_(4)H_(8),1.47 mmol g^(-1)),showing much higher uptake ratios of 1,3-C_(4)H_(6)/i-C_(4)H_(8)(47)and n-C_(4)H_(8)/i-C_(4)H_(8)(35)that outperform most of the benchmark porous materials for separating C_(4) olefins.Breakthrough experiments demonstrate successful separation of high-purity(99.9999%)i-C_(4)H_(8) and 1,3-C_(4)H_(6) from equimolar 1,3-C_(4)H_(6)/i-C_(4)H_(8),n-C_(4)H_(8)/i-C_(4)H_(8) and 1,3-C_(4)H_(6)/n-C_(4)H_(8)/i-C_(4)H_(8) mixtures.展开更多
文摘Diagnosing the cardiovascular disease is one of the biggest medical difficulties in recent years.Coronary cardiovascular(CHD)is a kind of heart and blood vascular disease.Predicting this sort of cardiac illness leads to more precise decisions for cardiac disorders.Implementing Grid Search Optimization(GSO)machine training models is therefore a useful way to forecast the sickness as soon as possible.The state-of-the-art work is the tuning of the hyperparameter together with the selection of the feature by utilizing the model search to minimize the false-negative rate.Three models with a cross-validation approach do the required task.Feature Selection based on the use of statistical and correlation matrices for multivariate analysis.For Random Search and Grid Search models,extensive comparison findings are produced utilizing retrieval,F1 score,and precision measurements.The models are evaluated using the metrics and kappa statistics that illustrate the three models’comparability.The study effort focuses on optimizing function selection,tweaking hyperparameters to improve model accuracy and the prediction of heart disease by examining Framingham datasets using random forestry classification.Tuning the hyperparameter in the model of grid search thus decreases the erroneous rate achieves global optimization.
基金Supported by the Anhui Province Sports Health Information Monitoring Technology Engineering Research Center Open Project (KF2023012)。
文摘Deep learning algorithm is an effective data mining method and has been used in many fields to solve practical problems.However,the deep learning algorithms often contain some hyper-parameters which may be continuous,integer,or mixed,and are often given based on experience but largely affect the effectiveness of activity recognition.In order to adapt to different hyper-parameter optimization problems,our improved Cuckoo Search(CS)algorithm is proposed to optimize the mixed hyper-parameters in deep learning algorithm.The algorithm optimizes the hyper-parameters in the deep learning model robustly,and intelligently selects the combination of integer type and continuous hyper-parameters that make the model optimal.Then,the mixed hyper-parameter in Convolutional Neural Network(CNN),Long-Short-Term Memory(LSTM)and CNN-LSTM are optimized based on the methodology on the smart home activity recognition datasets.Results show that the methodology can improve the performance of the deep learning model and whether we are experienced or not,we can get a better deep learning model using our method.
文摘A theoretical methodology is suggested for finding the malaria parasites’presence with the help of an intelligent hyper-parameter tuned Deep Learning(DL)based malaria parasite detection and classification(HPTDL-MPDC)in the smear images of human peripheral blood.Some existing approaches fail to predict the malaria parasitic features and reduce the prediction accuracy.The trained model initiated in the proposed system for classifying peripheral blood smear images into the non-parasite or parasite classes using the available online dataset.The Adagrad optimizer is stacked with the suggested pre-trained Deep Neural Network(DNN)with the help of the contrastive divergence method to pre-train.The features are extracted from the images in the proposed system to train the DNN for initializing the visible variables.The smear images show the concatenated feature to be utilized as the feature vector in the proposed system.Lastly,hyper-parameters are used to fine-tune DNN to calculate the class labels’probability.The suggested system outperforms more modern methodologies with an accuracy of 91%,precision of 89%,recall of 93%and F1-score of 91%.The HPTDL-MPDC has the primary application in detecting the parasite of malaria in the smear images of human peripheral blood.
文摘High-rise buildings are usually considered as flexible structures with low inherent damping. Therefore, these kinds of buildings are susceptible to wind-induced vibration. Tuned Mass Damper（TMD） can be used as an effective device to mitigate excessive vibrations. In this study, Artificial Neural Networks is used to find optimal mechanical properties of TMD for high-rise buildings subjected to wind load. The patterns obtained from structural analysis of different multi degree of freedom（MDF） systems are used for training neural networks. In order to obtain these patterns, structural models of some systems with 10 to 80 degrees-of-freedoms are built in MATLAB/SIMULINK program. Finally, the optimal properties of TMD are determined based on the objective of maximum displacement response reduction. The Auto-Regressive model is used to simulate the wind load. In this way, the uncertainties related to wind loading can be taken into account in neural network’s outputs. After training the neural network, it becomes possible to set the frequency and TMD mass ratio as inputs and get the optimal TMD frequency and damping ratio as outputs. As a case study, a benchmark 76-story office building is considered and the presented procedure is used to obtain optimal characteristics of the TMD for the building.
基金financial support provided by the National Key Research and Development Project(2019YFC0214403)Chongqing Joint Chinese Medicine Scientific Research Project(2021ZY023984)。
文摘Due to outstanding performance in cheminformatics,machine learning algorithms have been increasingly used to mine molecular properties and biomedical big data.The performance of machine learning models is known to critically depend on the selection of the hyper-parameter configuration.However,many studies either explored the optimal hyper-parameters per the grid searching method or employed arbitrarily selected hyper-parameters,which can easily lead to achieving a suboptimal hyper-parameter configuration.In this study,Hyperopt library embedding with the Bayesian optimization is employed to find optimal hyper-parameters for different machine learning algorithms.Six drug discovery datasets,including solubility,probe-likeness,h ERG,Chagas disease,tuberculosis,and malaria,are used to compare different machine learning algorithms with ECFP6 fingerprints.This contribution aims to evaluate whether the Bernoulli Na?ve Bayes,logistic linear regression,Ada Boost decision tree,random forest,support vector machine,and deep neural networks algorithms with optimized hyper-parameters can offer any improvement in testing as compared with the referenced models assessed by an array of metrics including AUC,F1-score,Cohen’s kappa,Matthews correlation coefficient,recall,precision,and accuracy.Based on the rank normalized score approach,the Hyperopt models achieve better or comparable performance on 33 out 36 models for different drug discovery datasets,showing significant improvement achieved by employing the Hyperopt library.The open-source code of all the 6 machine learning frameworks employed in the Hyperopt python package is provided to make this approach accessible to more scientists,who are not familiar with writing code.
基金supported in part by the National Natural Science Foundation of China(No.62273287)by the Shenzhen Science and Technology Innovation Council(Nos.JCYJ20220530143418040,JCY20170411102101881)the Thousand Youth Talents Plan funded by the central government of China.
文摘Regularized system identification has become the research frontier of system identification in the past decade.One related core subject is to study the convergence properties of various hyper-parameter estimators as the sample size goes to infinity.In this paper,we consider one commonly used hyper-parameter estimator,the empirical Bayes(EB).Its convergence in distribution has been studied,and the explicit expression of the covariance matrix of its limiting distribution has been given.However,what we are truly interested in are factors contained in the covariance matrix of the EB hyper-parameter estimator,and then,the convergence of its covariance matrix to that of its limiting distribution is required.In general,the convergence in distribution of a sequence of random variables does not necessarily guarantee the convergence of its covariance matrix.Thus,the derivation of such convergence is a necessary complement to our theoretical analysis about factors that influence the convergence properties of the EB hyper-parameter estimator.In this paper,we consider the regularized finite impulse response(FIR)model estimation with deterministic inputs,and show that the covariance matrix of the EB hyper-parameter estimator converges to that of its limiting distribution.Moreover,we run numerical simulations to demonstrate the efficacy of ourtheoretical results.
基金supports from the National Natural Science Foundation of China(Grant Nos.62335006,62022032,62275065,and 62405078)Key Laboratory of Opto-Electronic Information Acquisition and Manipulation(Anhui University),Ministry of Education(Grant No.OEIAM202202)+2 种基金Fundamental Research Funds for the Central Universities(Grant No.HIT.OCEF.2023011)China Postdoctoral Science Foundation(Grant No.2024M764172)Heilongjiang Postdoctoral Fund(Grant No.LBH-Z23144).
文摘In this paper,a novel four-prong quartz tuning fork(QTF)was designed with enlarged deformation area,large prong gap,and low resonant frequency to improve its performance in laser spectroscopy sensing.A theoretical simulation model was established to optimize the design of the QTF structure.In the simulation of quartz-enhanced photoacoustic spectroscopy(QEPAS)technology,the maximum stress and the surface charge density of the four-prong QTF demonstrated increases of 11.1-fold and 15.9-fold,respectively,compared to that of the standard two-prong QTF.In the simulation of light-induced thermoelastic spectroscopy(LITES)technology,the surface temperature difference of the four-prong QTF was found to be 11.4 times greater than that of the standard QTF.Experimental results indicated that the C_(2)H_(2)-QEPAS system based on this innovative design improved the signal-to-noise-ratio(SNR)by 4.67 times compared with the standard QTF-based system,and the SNR could increase up to 147.72 times when the four-prong QTF was equipped with its optimal acoustic micro-resonator(AmR).When the average time of the system reached 370 s,the system achieved a MDL as low as 21 ppb.The four-prong QTF-based C_(2)H_(2)-LITES system exhibited a SNR improvement by a factor of 4.52,and a MDL of 96 ppb was obtained when the average time of the system reached 100 s.The theoretical and experimental results effectively demonstrated the superiority of the four-prong QTF in the field of laser spectroscopy sensing.
文摘Hydroxyapatite nanoparticles(HAP NPs)were synthesized by a one‐step hydrothermal method.The surface of HAP NPs was grafted-SH and-COOH chelating groups via in situ surface‐modification with iminodiacetic acid(IDA)and 3‐mercaptopropyl trimethoxysilane(MPS)to afford dual surface‐capped nano‐amendment HAPIDA/MPS.The structure of HAP‐IDA/MPS was characterized,and its adsorption performance for Hg^(2+),Cu^(2+),Zn^(2+),Ni^(2+),Co^(2+),and Cd^(2+)was evaluated.The total adsorption capacity of 0.10 g HAP‐IDA/MPS nano‐amendment for Hg^(2+),Cu^(2+),Zn^(2+),Ni^(2+),Co^(2+),and Cd^(2+)with an initial mass concentration of 20 mg·L^(-1) reached 13.7 mg·g^(-1),about 4.3 times as much as that of HAP.Notably,HAP‐IDA/MPS nano‐amendment displayed the highest immobilization rate for Hg^(2+),possibly because of its chemical reaction with-SH to form sulfide,possessing the lowest solubility product constant among a variety of metal sulfides.
文摘The next generation of synchrotron radiation light sources features extremely low emittance,enabling the generation of synchrotron radiation with significantly higher brilliance,which facilitates the exploration of matter at smaller scales.However,the extremely low emittance results in stronger sextupole magnet strengths,leading to high natural chromaticity.This necessitates the use of sextupole magnets to correct the natural chromaticity.For the Shanghai Synchrotron Radiation Facility Upgrade(SSRF-U),a lattice was designed for the storage ring that can achieve an ultra-low natural emittance of 72.2 pm·rad at the beam energy of 3.5 GeV.However,the significant detuning effects,driven by high second-order resonant driving terms due to strong sextupoles,will degrade the performance of the facility.To resolve this issue,installation of octupoles in the SSRF-U storage ring has been planned.This paper presents the study results on configuration selection and optimization method for the octupoles.An optimal solution for the SSRF-U storage ring was obtained to effectively mitigate the amplitude-dependent tune shift and the second-order chromaticity,consequently leading to an increased dynamic aperture(DA),momentum acceptance(MA),and reduced sensitivity to magnetic field errors.
文摘Multiple tuned mass dampers(MTMDs)reduce dynamic response with multiple specified frequencies of building structures.Many optimization algorithms for placement design exist,though they rarely conform to code-based verification nor produce high quality solutions without high computational effort and high complexity.This study proposes an inverse element exchange method(IEEM)with multi-level programming and compares it to a single tuned mass damper(STMD)and uniform distribution of multiple tuned mass dampers in the frequency and time domains.A ten-story shear building is used for the numerical case study.The results show that the proposed method can offer improvement over the STMD,uniform distribution of multiple tuned mass dampers,and distribution optimized by genetic algorithms(GA)with regard to minimizing the interstory drift ratio(IDR)in both the frequency and time domains and the time consumption for optimization.
文摘Under the perspective of translation aesthetics,the article studies three Chinese-to-English translations of Slow,Slow Tune as research objects,comparing and contrasting the differences between the different translations under the perspective of translation aesthetics.The study finds that:Kenneth Rexroth’s translation is poor in textual understanding,with a lot of errors,and is too shallow in emotional expression,remaining only on the surface of the text;Lin Yutang makes occasional mistakes in understanding individual words,and is able to explore the implicit feelings of the lyricist,and his translation focuses on textual and phonological beauty;Xu Yuanchong’s translation is not only capable of conveying the meaning of the original work,but also of conveying the lyricist’s feelings more vividly and focusing on the phonological beauty.Xu Yuanchong’s translation not only conveys the meaning of the original work,but also vividly conveys the emotion of the lyricist,and pays attention to the beauty of sound.
基金supported by the Natural Science Research Project of Colleges and Universities in Anhui Province (No.KJ2021A0479)the Science Research Program of Anhui University of Finance and Economics (No.ACKYC22082)。
文摘We present a gain adaptive tuning method for fiber Raman amplifier(FRA) using two-stage neural networks(NNs) and double weights updates. After training the connection weights of two-stage NNs separately in training phase, the connection weights of the unified NN are updated again in verification phase according to error between the predicted and target gains to eliminate the inherent error of the NNs. The simulation results show that the mean of root mean square error(RMSE) and maximum error of gains are 0.131 d B and 0.281 d B, respectively. It shows that the method can realize adaptive adjustment function of FRA gain with high accuracy.
基金supported by Senior Talent Fund of Jiangsu University(No.5501310021)China Postdoctoral Science Foundation(No.2023M741419)+1 种基金the Young Elite Scientist Sponsorship Program by ZJAST(No.G301310002)Research Fund for International Scientists(No.22350710187).
文摘Lanthanide-sensitized upconverting nanoparticles(UCNPs)are widely studied because of their unusual optical characteristics,such as large antenna-generated anti-Stokes shifts,high photostability,and narrow emission bandwidths,which can be harnessed for a variety of applications including bioimaging,sensing,information security and high-level anticounterfeiting.The diverse requirements of these applications typically require precise control over upconversion luminescence(UCL).Recently,the concept of energy migration upconversion has emerged as an effective approach to modulate UCL for various lanthanide ions.Moreover,it provides valuable insights into the fundamental comprehension of energy transfer mechanisms on the nanoscale,thereby contributing to the design of efficient lanthanide-sensitized UCNPs and their practical applications.Here we present a comprehensive overview of the latest developments in energy migration upconversion in lanthanide-sensitized nanoparticles for photon upconversion tuning,encompassing design strategies,mechanistic investigations and applications.Additionally,some future prospects in the field of energy migration upconversion are also discussed.
文摘Grooved tuning forks with hierarchical structures have become some of the most widely used piezoelectric quartz microelectromechanical system devices;however,fabricating these devices requires multi-step processes due to the complexity of etching of quartz,particularly in specific orientations of the crystal lattice.This paper proposes a one-step fabrication strategy that can form a complete hierarchical structure with only a single etching process using novel lithography patterns.The core principle of this strategy is based on the effect of the size of the groove patterns on quartz etching,whereby trenches of varying depths can be created in a fixed etching time by adjusting the width of the hard mask.Specifically,the device outline and grooved structure can be completed using a seamlessly designed etching pattern and optimized time.Furthermore,the etching structure itself influences the etching results.It was found that dividing a wide trench by including a wall to separate it into two narrow trenches significantly reduces the etching rate,allowing for predictable tuning of the etching rate for wider grooves.This effectively increases the usability and flexibility of the one-step strategy.This was applied to the manufacture of an ultra-small quartz grooved tuning fork resonator with a frequency of 32.768 kHz in a single step,increasing production efficiency by almost 45%and reducing costs by almost 30%compared to current methods.This has great potential for improving the productivity of grooved tuning fork devices.It can also be extended to the fabrication of other quartz crystal devices requiring hierarchical structures.
基金supported by the National Natural Science Foundation of China(52272295,52071137,51977071,51802040,and 21802020)the Science and Technology Innovation Program of Hunan Province(2021RC3066 and 2021RC3067)+2 种基金the Natural Science Foundation of Hunan Province(2020JJ3004 and 2020JJ4192)Graduate Research Innovation Project of Hunan Province(CX20240456 and CX20240405)N.Zhang and X.Xie also acknowledge the financial support of the Fundamental Research Funds for the Central。
文摘P2-type layered oxide Na_(2/3)Ni_(1/3)Mn_(2/3)O_(2)(NM)is a promising cathode material for sodium-ion batteries(SIBs).However,the severe irreversible phase transition,sluggish Na+diffusion kinetics,and interfacial side reactions at high-voltage result in grievous capacity degradation and inferior electrochemical performance.Herein,a dual-function strategy of entropy tuning and artificial cathode electrolyte interface(CEI)layer construction is reported to generate a novel P2-type medium-entropy Na_(0.75)Li_(0.1)Mg_(0.05)Ni_(0.18)Mn_(0.66)Ta_(0.01)O_(2)with NaTaO_(3)surface modification(LMNMT)to address the aforementioned issues.In situ X-ray diffraction reveals that LMNMT exhibits a near zero-strain phase transition with a volume change of only 1.4%,which is significantly lower than that of NM(20.9%),indicating that entropy tuning effectively suppresses irreversible phase transitions and enhances ion diffusion.Kinetic analysis and post-cycling interfacial characterization further confirm that the artificial CEI layer promotes the formation of a stable,thin NaF-rich CEI and reduces interfacial side reactions,thereby further enhancing ion transport kinetics and surface/interface stability.Consequently,the LMNMT electrode exhibits outstanding rate capability(46 mA h g^(−1)at 20 C)and cycling stability(89.5%capacity retention after 200 cycles at 2 C)within the voltage range of 2–4.35 V.The LMNMT also exhibits superior all-climate performance and air stability.This study provides a novel path for the design of high-voltage cathode materials for SIBs.
文摘Hydrogenation catalysts frequently impose a compromise between activity and selectivity,where maximizing one property inevitably diminishes the other.Researchers from the Dalian Institute of Chemical Physics(DICP)of the Chinese Academy of Sciences,in collaboration with scholars from University of Science and Technology of China and the Karlsruhe Institute of Technology in Germany,cracked this dilemma by engineering bimetallic catalysts with atomic precision-a breakthrough that boosts hydrogenation efficiency by 35-fold while maintaining pinpoint accuracy,resolving the stubborn activity-selectivity paradox.
基金supported by the MSIT(Ministry of Science and ICT),Republic of Korea,under the ITRC(Information Technology Research Center)Support Program(IITP-2024-RS-2022-00156354)supervised by the IITP(Institute for Information&Communications Technology Planning&Evaluation)supported by the Technology Development Program(RS-2023-00264489)funded by the Ministry of SMEs and Startups(MSS,Republic of Korea).
文摘Fire can cause significant damage to the environment,economy,and human lives.If fire can be detected early,the damage can be minimized.Advances in technology,particularly in computer vision powered by deep learning,have enabled automated fire detection in images and videos.Several deep learning models have been developed for object detection,including applications in fire and smoke detection.This study focuses on optimizing the training hyperparameters of YOLOv8 andYOLOv10models usingBayesianTuning(BT).Experimental results on the large-scale D-Fire dataset demonstrate that this approach enhances detection performance.Specifically,the proposed approach improves the mean average precision at an Intersection over Union(IoU)threshold of 0.5(mAP50)of the YOLOv8s,YOLOv10s,YOLOv8l,and YOLOv10lmodels by 0.26,0.21,0.84,and 0.63,respectively,compared tomodels trainedwith the default hyperparameters.The performance gains are more pronounced in larger models,YOLOv8l and YOLOv10l,than in their smaller counterparts,YOLOv8s and YOLOv10s.Furthermore,YOLOv8 models consistently outperform YOLOv10,with mAP50 improvements of 0.26 for YOLOv8s over YOLOv10s and 0.65 for YOLOv8l over YOLOv10l when trained with BT.These results establish YOLOv8 as the preferred model for fire detection applications where detection performance is prioritized.
基金supported by National Natural Science Foundation of China(22090061,22375221)Fundamental Research Program of Shanxi Province(No.202203021223004)+1 种基金Program for Guangdong Introducing Innovative and Entrepreneurial Teams(2017ZT07C069)Hundred Talents Program of Sun Yat-Sen University.
文摘Separation of ternary C_(4) olefins(n-butene,iso-butene and 1,3-butadiene)is very challenging but crucial in the petrol-chemical industry due to their similar molecular sizes and properties.Herein,to optimize the separation efficiency for separation of C_(4) olefins,a new Hofmann-type MOF,[Ni(piz)Ni(CN)_(4)](piz=piperazine)-isostructural to the typical one[Ni(pyz)Ni(CN)_(4)](pyz=pyrazine),has been synthesized by a facile method from aqueous solution.The pore size reduction of[Ni(piz)Ni(CN)_(4)](3.62A,in contrast to 3.85A in[Ni(pyz)Ni(CN)_(4)])results in negligible iso-butene(i-C_(4)H_(8))uptake(from 2.92 to 0.04 mmol g^(-1))whereas retaining significant uptake for 1,3-butadiene(1,3-C_(4)H_(6),1.96 mmol g^(-1))and n-butene(n-C_(4)H_(8),1.47 mmol g^(-1)),showing much higher uptake ratios of 1,3-C_(4)H_(6)/i-C_(4)H_(8)(47)and n-C_(4)H_(8)/i-C_(4)H_(8)(35)that outperform most of the benchmark porous materials for separating C_(4) olefins.Breakthrough experiments demonstrate successful separation of high-purity(99.9999%)i-C_(4)H_(8) and 1,3-C_(4)H_(6) from equimolar 1,3-C_(4)H_(6)/i-C_(4)H_(8),n-C_(4)H_(8)/i-C_(4)H_(8) and 1,3-C_(4)H_(6)/n-C_(4)H_(8)/i-C_(4)H_(8) mixtures.
