Oscillation detection has been a hot research topic in industries due to the high incidence of oscillation loops and their negative impact on plant profitability.Although numerous automatic detection techniques have b...Oscillation detection has been a hot research topic in industries due to the high incidence of oscillation loops and their negative impact on plant profitability.Although numerous automatic detection techniques have been proposed,most of them can only address part of the practical difficulties.An oscillation is heuristically defined as a visually apparent periodic variation.However,manual visual inspection is labor-intensive and prone to missed detection.Convolutional neural networks(CNNs),inspired by animal visual systems,have been raised with powerful feature extraction capabilities.In this work,an exploration of the typical CNN models for visual oscillation detection is performed.Specifically,we tested MobileNet-V1,ShuffleNet-V2,Efficient Net-B0,and GhostNet models,and found that such a visual framework is well-suited for oscillation detection.The feasibility and validity of this framework are verified utilizing extensive numerical and industrial cases.Compared with state-of-theart oscillation detectors,the suggested framework is more straightforward and more robust to noise and mean-nonstationarity.In addition,this framework generalizes well and is capable of handling features that are not present in the training data,such as multiple oscillations and outliers.展开更多
The deep structure,material circulation,and dynamic processes in the Southeast Asia have long been an elusive scientific puzzle due to the lack of systematic scientific observations and recognized theoretical models.B...The deep structure,material circulation,and dynamic processes in the Southeast Asia have long been an elusive scientific puzzle due to the lack of systematic scientific observations and recognized theoretical models.Based on the deep seismic tomography using long-period natural earthquake data,in this study,the deep structure and material circulation of the curved subduction system in Southeast Asia was studied,and the dynamic processes since 100 million years ago was reconstructed.It is pointed out that challenges still exist in the precise reconstruction of deep mantle structures of the study area,the influence of multi-stage subduction on deep material exchange and shallow magma activity,as well as the spatiotemporal evolution and coupling mechanism of multi-plate convergence.Future work should focus on high-resolution land-sea joint 3-D seismic tomography imaging of the curved subduction system in the Southeast Asia,combined with geochemical analysis and geodynamic modelling works.展开更多
An extreme rainfall event occurred over Hangzhou,China,during the afternoon hours on 24 June 2013.This event occurred under suitable synoptic conditions and the maximum 4-h cumulative rainfall amount was over 150 mm.T...An extreme rainfall event occurred over Hangzhou,China,during the afternoon hours on 24 June 2013.This event occurred under suitable synoptic conditions and the maximum 4-h cumulative rainfall amount was over 150 mm.This rainfall event had two major rainbands.One was caused by a quasi-stationary convective line,and the other by a backbuilding convective line related to the interaction of the outflow boundary from the first rainband and an existing low-level mesoscale convergence line associated with a mei-yu frontal system.The rainfall event lasted 4 h,while the back-building process occurred in 2 h when the extreme rainfall center formed.So far,few studies have examined the back-building processes in the mei-yu season that are caused by the interaction of a mesoscale convergence line and a convective cold pool.The two rainbands are successfully reproduced by the Weather Research and Forecasting(WRF)model with fourlevel,two-way interactive nesting.In the model,new cells repeatedly occur at the west side of older cells,and the backbuilding process occurs in an environment with large CAPE,a low LFC,and plenty of water vapor.Outflows from older cells enhance the low-level convergence that forces new cells.High precipitation efficiency of the back-building training cells leads to accumulated precipitation of over 150 mm.Sensitivity experiments without evaporation of rainwater show that the convective cold pool plays an important role in the organization of the back-building process in the current extreme precipitation case.展开更多
In this study,the vertical components of broadband teleseismic P wave data recorded by China Earthquake Network are used to image the rupture processes of the February 6th,2023 Turkish earthquake doublet via back proj...In this study,the vertical components of broadband teleseismic P wave data recorded by China Earthquake Network are used to image the rupture processes of the February 6th,2023 Turkish earthquake doublet via back projection analysis.Data in two frequency bands(0.5-2 Hz and 1-3 Hz)are used in the imaging processes.The results show that the rupture of the first event extends about 200 km to the northeast and about 150 km to the southwest,lasting~90 s in total.The southwestern rupture is triggered by the northeastern rupture,demonstrating a sequential bidirectional unilateral rupture pattern.The rupture of the second event extends approximately 80 km in both northeast and west directions,lasting~35 s in total and demonstrates a typical bilateral rupture feature.The cascading ruptures on both sides also reflect the occurrence of selective rupture behaviors on bifurcated faults.In addition,we observe super-shear ruptures on certain fault sections with relatively straight fault structures and sparse aftershocks.展开更多
This study aims to identify a natural plant chemical with hypolipidemic effects that can be used to treat high cholesterol without adverse reactions.Through network pharmacology screening,it was found that Rosae Rugos...This study aims to identify a natural plant chemical with hypolipidemic effects that can be used to treat high cholesterol without adverse reactions.Through network pharmacology screening,it was found that Rosae Rugosae Flos(RF)flavonoids had potential therapeutic effects on hyperlipidemia and its mechanism of action was discussed.TCMSP and GeneCards databases were used to obtain active ingredients and disease targets.Venn diagrams were drawn to illustrate the findings.The interaction network diagram was created using Cytoscape 3.8.0 software.The PPI protein network was constructed using String.GO and KEGG enrichment analysis was performed using Metascape.The results revealed 2 active flavonoid ingredients and 60 potential targets in RF.The key targets,including CCL2,PPARG,and PPARA,were found to play a role in multiple pathways such as the AGE-RAGE signaling pathway,lipid and atherosclerosis,and cancer pathway in diabetic complications.The solvent extraction method was optimized for efficient flavonoid extraction based on network pharmacology prediction results.This was achieved through a single factor and orthogonal test,resulting in an optimum process with a reflux time of 1.5 h,a solid-liquid ratio of 1:13 g/mL,and an ethanol concentration of 50%.展开更多
Deep learning(DL) is progressively popular as a viable alternative to traditional signal processing(SP) based methods for fault diagnosis. However, the lack of explainability makes DL-based fault diagnosis methods dif...Deep learning(DL) is progressively popular as a viable alternative to traditional signal processing(SP) based methods for fault diagnosis. However, the lack of explainability makes DL-based fault diagnosis methods difficult to be trusted and understood by industrial users. In addition, the extraction of weak fault features from signals with heavy noise is imperative in industrial applications. To address these limitations, inspired by the Filterbank-Feature-Decision methodology, we propose a new Signal Processing Informed Neural Network(SPINN) framework by embedding SP knowledge into the DL model. As one of the practical implementations for SPINN, a denoising fault-aware wavelet network(DFAWNet) is developed, which consists of fused wavelet convolution(FWConv), dynamic hard thresholding(DHT),index-based soft filtering(ISF), and a classifier. Taking advantage of wavelet transform, FWConv extracts multiscale features while learning wavelet scales and selecting important wavelet bases automatically;DHT dynamically eliminates noise-related components via point-wise hard thresholding;inspired by index-based filtering, ISF optimizes and selects optimal filters for diagnostic feature extraction. It’s worth noting that SPINN may be readily applied to different deep learning networks by simply adding filterbank and feature modules in front. Experiments results demonstrate a significant diagnostic performance improvement over other explainable or denoising deep learning networks. The corresponding code is available at https://github. com/alber tszg/DFAWn et.展开更多
As a key component of injection molding,multi-cavity hot runner(MCHR)system faces the crucial problem of polymer melt filling imbalance among the cavities.The thermal imbalance in the system has been considered as the...As a key component of injection molding,multi-cavity hot runner(MCHR)system faces the crucial problem of polymer melt filling imbalance among the cavities.The thermal imbalance in the system has been considered as the leading cause.Hence,the solution may rest with the synchronization of those heating processes in MCHR system.This paper proposes a’Master-Slave’generalized predictive synchronization control(MS-GPSC)method with’Mr.Slowest’strategy for preheating stage of MCHR system.The core of the proposed method is choosing the heating process with slowest dynamics as the’Master’to track the setpoint,while the other heating processes are treated as‘Slaves’tracking the output of’Master’.This proposed method is shown to have the good ability of temperature synchronization.The corresponding analysis is conducted on parameters tuning and stability,simulations and experiments show the strategy is effective.展开更多
Object detection plays a vital role in the video surveillance systems.To enhance security,surveillance cameras are now installed in public areas such as traffic signals,roadways,retail malls,train stations,and banks.Ho...Object detection plays a vital role in the video surveillance systems.To enhance security,surveillance cameras are now installed in public areas such as traffic signals,roadways,retail malls,train stations,and banks.However,monitor-ing the video continually at a quicker pace is a challenging job.As a consequence,security cameras are useless and need human monitoring.The primary difficulty with video surveillance is identifying abnormalities such as thefts,accidents,crimes,or other unlawful actions.The anomalous action does not occur at a high-er rate than usual occurrences.To detect the object in a video,first we analyze the images pixel by pixel.In digital image processing,segmentation is the process of segregating the individual image parts into pixels.The performance of segmenta-tion is affected by irregular illumination and/or low illumination.These factors highly affect the real-time object detection process in the video surveillance sys-tem.In this paper,a modified ResNet model(M-Resnet)is proposed to enhance the image which is affected by insufficient light.Experimental results provide the comparison of existing method output and modification architecture of the ResNet model shows the considerable amount improvement in detection objects in the video stream.The proposed model shows better results in the metrics like preci-sion,recall,pixel accuracy,etc.,andfinds a reasonable improvement in the object detection.展开更多
Implementation of artificial neural network(ANN)is very important to theoretical studyand applications of ANN.On the basis of studying existing methods,this paper concentrateson the DSP-based virtual implementation of...Implementation of artificial neural network(ANN)is very important to theoretical studyand applications of ANN.On the basis of studying existing methods,this paper concentrateson the DSP-based virtual implementation of ANN.A parallel processing system composed ofTMS320C30 has been designed and configured,which ean provide a peak speed as high as100 MFLOPS and a parallel efficiency of 90%(during the forward phase of BP),and can heused for sonar signal processing.Scalability of the system is also studied.展开更多
A novel process monitoring method based on convolutional neural network(CNN)is proposed and applied to detect faults in industrial process.By utilizing the CNN algorithm,cross-correlation and autocorrelation among var...A novel process monitoring method based on convolutional neural network(CNN)is proposed and applied to detect faults in industrial process.By utilizing the CNN algorithm,cross-correlation and autocorrelation among variables are captured to establish a prediction model for each process variable to approximate the first-principle of physical/chemical relationships among different variables under normal operating conditions.When the process is operated under pre-set operating conditions,prediction residuals can be assumed as noise if a proper model is employed.Once process faults occur,the residuals will increase due to the changes of correlation among variables.A principal component analysis(PCA)model based on the residuals is established to realize process monitoring.By monitoring the changes in main feature of prediction residuals,the faults can be promptly detected.Case studies on a numerical nonlinear example and data from two industrial processes are presented to validate the performance of process monitoring based on CNN.展开更多
The need for the analysis of modern businesses is rapidly increasing as the supporting enterprise systems generate more and more data.This data can be extremely valuable for executing organizations because the data al...The need for the analysis of modern businesses is rapidly increasing as the supporting enterprise systems generate more and more data.This data can be extremely valuable for executing organizations because the data allows constant monitoring,analyzing,and improving the underlying processes,which leads to the reduction of cost and the improvement of the quality.Process mining is a useful technique for analyzing enterprise systems by using an event log that contains behaviours.This research focuses on the process discovery and refinement using real-life event log data collected from a large multinational organization that deals with coatings and paints.By investigating and analyzing their order handling pro-cesses,this study aims at learning a model that gives insight inspection of the processes and performance analysis.Furthermore,the animation is also performed for the better inspection,diagnostics,and compliance-related questions to specify the system.The configuration of the system and the conformance checking for further enhancement is also addressed in this research.To achieve the objectives,this research uses process mining techniques,i.e.process discovery in the form of formal Petri nets models with the help of process maps,and process refinement through conformance checking and enhancement.Initially,the identified executed process is reconstructed by using the process discovery techniques.Following the reconstruction,we perform a deep analysis for the underlying process to ensure the process improvement and redesigning.Finally,some recommendations are made to improve the enterprise management system processes.展开更多
This paper develops a deep learning tool based on neural processes(NPs)called the Peri-Net-Pro,to predict the crack patterns in a moving disk and classifies them according to the classification modes with quantified u...This paper develops a deep learning tool based on neural processes(NPs)called the Peri-Net-Pro,to predict the crack patterns in a moving disk and classifies them according to the classification modes with quantified uncertainties.In particular,image classification and regression studies are conducted by means of convolutional neural networks(CNNs)and NPs.First,the amount and quality of the data are enhanced by using peridynamics to theoretically compensate for the problems of the finite element method(FEM)in generating crack pattern images.Second,case studies are conducted with the prototype microelastic brittle(PMB),linear peridynamic solid(LPS),and viscoelastic solid(VES)models obtained by using the peridynamic theory.The case studies are performed to classify the images by using CNNs and determine the suitability of the PMB,LBS,and VES models.Finally,a regression analysis is performed on the crack pattern images with NPs to predict the crack patterns.The regression analysis results confirm that the variance decreases when the number of epochs increases by using the NPs.The training results gradually improve,and the variance ranges decrease to less than 0.035.The main finding of this study is that the NPs enable accurate predictions,even with missing or insufficient training data.The results demonstrate that if the context points are set to the 10th,100th,300th,and 784th,the training information is deliberately omitted for the context points of the 10th,100th,and 300th,and the predictions are different when the context points are significantly lower.However,the comparison of the results of the 100th and 784th context points shows that the predicted results are similar because of the Gaussian processes in the NPs.Therefore,if the NPs are employed for training,the missing information of the training data can be supplemented to predict the results.展开更多
Intelligent process planning(PP)is one of the most important components in an intelligent manufacturing system and acts as a bridge between product designing and practical manufacturing.PP is a nondeterministic polyno...Intelligent process planning(PP)is one of the most important components in an intelligent manufacturing system and acts as a bridge between product designing and practical manufacturing.PP is a nondeterministic polynomial-time(NP)-hard problem and,as existing mathematical models are not formulated in linear forms,they cannot be solved well to achieve exact solutions for PP problems.This paper proposes a novel mixed-integer linear programming(MILP)mathematical model by considering the network topology structure and the OR nodes that represent a type of OR logic inside the network.Precedence relationships between operations are discussed by raising three types of precedence relationship matrices.Furthermore,the proposed model can be programmed in commonly-used mathematical programming solvers,such as CPLEX,Gurobi,and so forth,to search for optimal solutions for most open problems.To verify the effectiveness and generality of the proposed model,five groups of numerical experiments are conducted on well-known benchmarks.The results show that the proposed model can solve PP problems effectively and can obtain better solutions than those obtained by the state-ofthe-art algorithms.展开更多
A correct and timely fault diagnosis is important for improving the safety and reliability of chemical processes. With the advancement of big data technology, data-driven fault diagnosis methods are being extensively ...A correct and timely fault diagnosis is important for improving the safety and reliability of chemical processes. With the advancement of big data technology, data-driven fault diagnosis methods are being extensively used and still have considerable potential. In recent years, methods based on deep neural networks have made significant breakthroughs, and fault diagnosis methods for industrial processes based on deep learning have attracted considerable research attention. Therefore, we propose a fusion deeplearning algorithm based on a fully convolutional neural network(FCN) to extract features and build models to correctly diagnose all types of faults. We use long short-term memory(LSTM) units to expand our proposed FCN so that our proposed deep learning model can better extract the time-domain features of chemical process data. We also introduce the attention mechanism into the model, aimed at highlighting the importance of features, which is significant for the fault diagnosis of chemical processes with many features. When applied to the benchmark Tennessee Eastman process, our proposed model exhibits impressive performance, demonstrating the effectiveness of the attention-based LSTM FCN in chemical process fault diagnosis.展开更多
The prognostics health management(PHM)fromthe systematic viewis critical to the healthy continuous operation of processmanufacturing systems(PMS),with different kinds of dynamic interference events.This paper proposes...The prognostics health management(PHM)fromthe systematic viewis critical to the healthy continuous operation of processmanufacturing systems(PMS),with different kinds of dynamic interference events.This paper proposes a three leveled digital twinmodel for the systematic PHMof PMSs.The unit-leveled digital twinmodel of each basic device unit of PMSs is constructed based on edge computing,which can provide real-time monitoring and analysis of the device status.The station-leveled digital twin models in the PMSs are designed to optimize and control the process parameters,which are deployed for the manufacturing execution on the fog server.The shop-leveled digital twin maintenancemodel is designed for production planning,which gives production instructions fromthe private industrial cloud server.To cope with the dynamic disturbances of a PMS,a big data-driven framework is proposed to control the three-level digital twin models,which contains indicator prediction,influence evaluation,and decisionmaking.Finally,a case study with a real chemical fiber system is introduced to illustrate the effectiveness of the digital twin model with edge-fog-cloud computing for the systematic PHM of PMSs.The result demonstrates that the three-leveled digital twin model for the systematic PHM in PMSs works well in the system’s respects.展开更多
The recent developments in Multimedia Internet of Things(MIoT)devices,empowered with Natural Language Processing(NLP)model,seem to be a promising future of smart devices.It plays an important role in industrial models...The recent developments in Multimedia Internet of Things(MIoT)devices,empowered with Natural Language Processing(NLP)model,seem to be a promising future of smart devices.It plays an important role in industrial models such as speech understanding,emotion detection,home automation,and so on.If an image needs to be captioned,then the objects in that image,its actions and connections,and any silent feature that remains under-projected or missing from the images should be identified.The aim of the image captioning process is to generate a caption for image.In next step,the image should be provided with one of the most significant and detailed descriptions that is syntactically as well as semantically correct.In this scenario,computer vision model is used to identify the objects and NLP approaches are followed to describe the image.The current study develops aNatural Language Processing with Optimal Deep Learning Enabled Intelligent Image Captioning System(NLPODL-IICS).The aim of the presented NLPODL-IICS model is to produce a proper description for input image.To attain this,the proposed NLPODL-IICS follows two stages such as encoding and decoding processes.Initially,at the encoding side,the proposed NLPODL-IICS model makes use of Hunger Games Search(HGS)with Neural Search Architecture Network(NASNet)model.This model represents the input data appropriately by inserting it into a predefined length vector.Besides,during decoding phase,Chimp Optimization Algorithm(COA)with deeper Long Short Term Memory(LSTM)approach is followed to concatenate the description sentences 4436 CMC,2023,vol.74,no.2 produced by the method.The application of HGS and COA algorithms helps in accomplishing proper parameter tuning for NASNet and LSTM models respectively.The proposed NLPODL-IICS model was experimentally validated with the help of two benchmark datasets.Awidespread comparative analysis confirmed the superior performance of NLPODL-IICS model over other models.展开更多
A variety of neural networks have been presented to deal with issues in deep learning in the last decades.Despite the prominent success achieved by the neural network,it still lacks theoretical guidance to design an e...A variety of neural networks have been presented to deal with issues in deep learning in the last decades.Despite the prominent success achieved by the neural network,it still lacks theoretical guidance to design an efficient neural network model,and verifying the performance of a model needs excessive resources.Previous research studies have demonstrated that many existing models can be regarded as different numerical discretizations of differential equations.This connection sheds light on designing an effective recurrent neural network(RNN)by resorting to numerical analysis.Simple RNN is regarded as a discretisation of the forward Euler scheme.Considering the limited solution accuracy of the forward Euler methods,a Taylor‐type discrete scheme is presented with lower truncation error and a Taylor‐type RNN(T‐RNN)is designed with its guidance.Extensive experiments are conducted to evaluate its performance on statistical language models and emotion analysis tasks.The noticeable gains obtained by T‐RNN present its superiority and the feasibility of designing the neural network model using numerical methods.展开更多
Micro-mobile heat pipe-cooled nuclear power plants are promising candidates for distributed energy resource power genera-tors and can be flexibly deployed in remote places to meet increasing electric power demands.How...Micro-mobile heat pipe-cooled nuclear power plants are promising candidates for distributed energy resource power genera-tors and can be flexibly deployed in remote places to meet increasing electric power demands.However,previous steady-state simulations and experiments have deviated significantly from actual micronuclear system operations.Hence,a transient analysis is required for performance optimization and safety assessment.In this study,a hardware-in-the-loop(HIL)approach was used to investigate the dynamic behavior of scaled-down heat pipe-cooled systems.The real-time features of the HIL architecture were interpreted and validated,and an optimal time step of 500 ms was selected for the thermal transient.The power transient was modeled using point kinetic equations,and a scaled-down thermal prototype was set up to avoid mod-eling unpredictable heat transfer behaviors and feeding temperature samples into the main program running on a desktop PC.A series of dynamic test results showed significant power and temperature oscillations during the transient process,owing to the inconsistency of the rapid nuclear reaction rate and large thermal inertia.The proposed HIL approach is stable and effective for further studying of the dynamic characteristics and control optimization of solid-state small nuclear-powered systems at an early prototyping stage.展开更多
In the graph signal processing(GSP)framework,distributed algorithms are highly desirable in processing signals defined on large-scale networks.However,in most existing distributed algorithms,all nodes homogeneously pe...In the graph signal processing(GSP)framework,distributed algorithms are highly desirable in processing signals defined on large-scale networks.However,in most existing distributed algorithms,all nodes homogeneously perform the local computation,which calls for heavy computational and communication costs.Moreover,in many real-world networks,such as those with straggling nodes,the homogeneous manner may result in serious delay or even failure.To this end,we propose active network decomposition algorithms to select non-straggling nodes(normal nodes)that perform the main computation and communication across the network.To accommodate the decomposition in different kinds of networks,two different approaches are developed,one is centralized decomposition that leverages the adjacency of the network and the other is distributed decomposition that employs the indicator message transmission between neighboring nodes,which constitutes the main contribution of this paper.By incorporating the active decomposition scheme,a distributed Newton method is employed to solve the least squares problem in GSP,where the Hessian inverse is approximately evaluated by patching a series of inverses of local Hessian matrices each of which is governed by one normal node.The proposed algorithm inherits the fast convergence of the second-order algorithms while maintains low computational and communication cost.Numerical examples demonstrate the effectiveness of the proposed algorithm.展开更多
Depression has become one of the most common mental illnesses in the world.For better prediction and diagnosis,methods of automatic depression recognition based on speech signal are constantly proposed and updated,wit...Depression has become one of the most common mental illnesses in the world.For better prediction and diagnosis,methods of automatic depression recognition based on speech signal are constantly proposed and updated,with a transition from the early traditional methods based on hand‐crafted features to the application of architectures of deep learning.This paper systematically and precisely outlines the most prominent and up‐to‐date research of automatic depression recognition by intelligent speech signal processing so far.Furthermore,methods for acoustic feature extraction,algorithms for classification and regression,as well as end to end deep models are investigated and analysed.Finally,general trends are summarised and key unresolved issues are identified to be considered in future studies of automatic speech depression recognition.展开更多
基金the National Natural Science Foundation of China(62003298,62163036)the Major Project of Science and Technology of Yunnan Province(202202AD080005,202202AH080009)the Yunnan University Professional Degree Graduate Practice Innovation Fund Project(ZC-22222770)。
文摘Oscillation detection has been a hot research topic in industries due to the high incidence of oscillation loops and their negative impact on plant profitability.Although numerous automatic detection techniques have been proposed,most of them can only address part of the practical difficulties.An oscillation is heuristically defined as a visually apparent periodic variation.However,manual visual inspection is labor-intensive and prone to missed detection.Convolutional neural networks(CNNs),inspired by animal visual systems,have been raised with powerful feature extraction capabilities.In this work,an exploration of the typical CNN models for visual oscillation detection is performed.Specifically,we tested MobileNet-V1,ShuffleNet-V2,Efficient Net-B0,and GhostNet models,and found that such a visual framework is well-suited for oscillation detection.The feasibility and validity of this framework are verified utilizing extensive numerical and industrial cases.Compared with state-of-theart oscillation detectors,the suggested framework is more straightforward and more robust to noise and mean-nonstationarity.In addition,this framework generalizes well and is capable of handling features that are not present in the training data,such as multiple oscillations and outliers.
基金Support by the National Natural Science Foundation of China(No.92258303)the Project of Donghai Laboratory(No.DH-2022ZY0005)。
文摘The deep structure,material circulation,and dynamic processes in the Southeast Asia have long been an elusive scientific puzzle due to the lack of systematic scientific observations and recognized theoretical models.Based on the deep seismic tomography using long-period natural earthquake data,in this study,the deep structure and material circulation of the curved subduction system in Southeast Asia was studied,and the dynamic processes since 100 million years ago was reconstructed.It is pointed out that challenges still exist in the precise reconstruction of deep mantle structures of the study area,the influence of multi-stage subduction on deep material exchange and shallow magma activity,as well as the spatiotemporal evolution and coupling mechanism of multi-plate convergence.Future work should focus on high-resolution land-sea joint 3-D seismic tomography imaging of the curved subduction system in the Southeast Asia,combined with geochemical analysis and geodynamic modelling works.
基金supported by the National Natural Science Foundation of China (Grant Nos.41730965, U2242204, and 41175047)the National Key Basic Research and Development Project of China (Grant No.2013CB430104)+2 种基金the Key Project of the Joint Funds of the Natural Science Foundation of Zhejiang Province (Grant No.LZJMZ23D050003financial support from the China Scholarship Council for her visit to CAPSUniversity of Oklahoma
文摘An extreme rainfall event occurred over Hangzhou,China,during the afternoon hours on 24 June 2013.This event occurred under suitable synoptic conditions and the maximum 4-h cumulative rainfall amount was over 150 mm.This rainfall event had two major rainbands.One was caused by a quasi-stationary convective line,and the other by a backbuilding convective line related to the interaction of the outflow boundary from the first rainband and an existing low-level mesoscale convergence line associated with a mei-yu frontal system.The rainfall event lasted 4 h,while the back-building process occurred in 2 h when the extreme rainfall center formed.So far,few studies have examined the back-building processes in the mei-yu season that are caused by the interaction of a mesoscale convergence line and a convective cold pool.The two rainbands are successfully reproduced by the Weather Research and Forecasting(WRF)model with fourlevel,two-way interactive nesting.In the model,new cells repeatedly occur at the west side of older cells,and the backbuilding process occurs in an environment with large CAPE,a low LFC,and plenty of water vapor.Outflows from older cells enhance the low-level convergence that forces new cells.High precipitation efficiency of the back-building training cells leads to accumulated precipitation of over 150 mm.Sensitivity experiments without evaporation of rainwater show that the convective cold pool plays an important role in the organization of the back-building process in the current extreme precipitation case.
基金supported by the National Key R&D Program of China(No.2022YFF0800601)National Scientific Foundation of China(Nos.41930103 and 41774047).
文摘In this study,the vertical components of broadband teleseismic P wave data recorded by China Earthquake Network are used to image the rupture processes of the February 6th,2023 Turkish earthquake doublet via back projection analysis.Data in two frequency bands(0.5-2 Hz and 1-3 Hz)are used in the imaging processes.The results show that the rupture of the first event extends about 200 km to the northeast and about 150 km to the southwest,lasting~90 s in total.The southwestern rupture is triggered by the northeastern rupture,demonstrating a sequential bidirectional unilateral rupture pattern.The rupture of the second event extends approximately 80 km in both northeast and west directions,lasting~35 s in total and demonstrates a typical bilateral rupture feature.The cascading ruptures on both sides also reflect the occurrence of selective rupture behaviors on bifurcated faults.In addition,we observe super-shear ruptures on certain fault sections with relatively straight fault structures and sparse aftershocks.
文摘This study aims to identify a natural plant chemical with hypolipidemic effects that can be used to treat high cholesterol without adverse reactions.Through network pharmacology screening,it was found that Rosae Rugosae Flos(RF)flavonoids had potential therapeutic effects on hyperlipidemia and its mechanism of action was discussed.TCMSP and GeneCards databases were used to obtain active ingredients and disease targets.Venn diagrams were drawn to illustrate the findings.The interaction network diagram was created using Cytoscape 3.8.0 software.The PPI protein network was constructed using String.GO and KEGG enrichment analysis was performed using Metascape.The results revealed 2 active flavonoid ingredients and 60 potential targets in RF.The key targets,including CCL2,PPARG,and PPARA,were found to play a role in multiple pathways such as the AGE-RAGE signaling pathway,lipid and atherosclerosis,and cancer pathway in diabetic complications.The solvent extraction method was optimized for efficient flavonoid extraction based on network pharmacology prediction results.This was achieved through a single factor and orthogonal test,resulting in an optimum process with a reflux time of 1.5 h,a solid-liquid ratio of 1:13 g/mL,and an ethanol concentration of 50%.
基金National Natural Science Foundation of China (Grant Nos. 51835009, 52105116)China Postdoctoral Science Foundation (Grant Nos. 2021M692557, 2021TQ0263)。
文摘Deep learning(DL) is progressively popular as a viable alternative to traditional signal processing(SP) based methods for fault diagnosis. However, the lack of explainability makes DL-based fault diagnosis methods difficult to be trusted and understood by industrial users. In addition, the extraction of weak fault features from signals with heavy noise is imperative in industrial applications. To address these limitations, inspired by the Filterbank-Feature-Decision methodology, we propose a new Signal Processing Informed Neural Network(SPINN) framework by embedding SP knowledge into the DL model. As one of the practical implementations for SPINN, a denoising fault-aware wavelet network(DFAWNet) is developed, which consists of fused wavelet convolution(FWConv), dynamic hard thresholding(DHT),index-based soft filtering(ISF), and a classifier. Taking advantage of wavelet transform, FWConv extracts multiscale features while learning wavelet scales and selecting important wavelet bases automatically;DHT dynamically eliminates noise-related components via point-wise hard thresholding;inspired by index-based filtering, ISF optimizes and selects optimal filters for diagnostic feature extraction. It’s worth noting that SPINN may be readily applied to different deep learning networks by simply adding filterbank and feature modules in front. Experiments results demonstrate a significant diagnostic performance improvement over other explainable or denoising deep learning networks. The corresponding code is available at https://github. com/alber tszg/DFAWn et.
基金supported in part by National Natural Science Foundation of China(62203127)Basic and Applied Basic Research Project of Guangzhou City(2023A04J1712)+1 种基金The Foshan-HKUST Projects Program(FSUST19-FYTRI01)GDAS’Project of Science and Technology Development(2020GDASYL-20200202001).
文摘As a key component of injection molding,multi-cavity hot runner(MCHR)system faces the crucial problem of polymer melt filling imbalance among the cavities.The thermal imbalance in the system has been considered as the leading cause.Hence,the solution may rest with the synchronization of those heating processes in MCHR system.This paper proposes a’Master-Slave’generalized predictive synchronization control(MS-GPSC)method with’Mr.Slowest’strategy for preheating stage of MCHR system.The core of the proposed method is choosing the heating process with slowest dynamics as the’Master’to track the setpoint,while the other heating processes are treated as‘Slaves’tracking the output of’Master’.This proposed method is shown to have the good ability of temperature synchronization.The corresponding analysis is conducted on parameters tuning and stability,simulations and experiments show the strategy is effective.
文摘Object detection plays a vital role in the video surveillance systems.To enhance security,surveillance cameras are now installed in public areas such as traffic signals,roadways,retail malls,train stations,and banks.However,monitor-ing the video continually at a quicker pace is a challenging job.As a consequence,security cameras are useless and need human monitoring.The primary difficulty with video surveillance is identifying abnormalities such as thefts,accidents,crimes,or other unlawful actions.The anomalous action does not occur at a high-er rate than usual occurrences.To detect the object in a video,first we analyze the images pixel by pixel.In digital image processing,segmentation is the process of segregating the individual image parts into pixels.The performance of segmenta-tion is affected by irregular illumination and/or low illumination.These factors highly affect the real-time object detection process in the video surveillance sys-tem.In this paper,a modified ResNet model(M-Resnet)is proposed to enhance the image which is affected by insufficient light.Experimental results provide the comparison of existing method output and modification architecture of the ResNet model shows the considerable amount improvement in detection objects in the video stream.The proposed model shows better results in the metrics like preci-sion,recall,pixel accuracy,etc.,andfinds a reasonable improvement in the object detection.
文摘Implementation of artificial neural network(ANN)is very important to theoretical studyand applications of ANN.On the basis of studying existing methods,this paper concentrateson the DSP-based virtual implementation of ANN.A parallel processing system composed ofTMS320C30 has been designed and configured,which ean provide a peak speed as high as100 MFLOPS and a parallel efficiency of 90%(during the forward phase of BP),and can heused for sonar signal processing.Scalability of the system is also studied.
文摘A novel process monitoring method based on convolutional neural network(CNN)is proposed and applied to detect faults in industrial process.By utilizing the CNN algorithm,cross-correlation and autocorrelation among variables are captured to establish a prediction model for each process variable to approximate the first-principle of physical/chemical relationships among different variables under normal operating conditions.When the process is operated under pre-set operating conditions,prediction residuals can be assumed as noise if a proper model is employed.Once process faults occur,the residuals will increase due to the changes of correlation among variables.A principal component analysis(PCA)model based on the residuals is established to realize process monitoring.By monitoring the changes in main feature of prediction residuals,the faults can be promptly detected.Case studies on a numerical nonlinear example and data from two industrial processes are presented to validate the performance of process monitoring based on CNN.
文摘The need for the analysis of modern businesses is rapidly increasing as the supporting enterprise systems generate more and more data.This data can be extremely valuable for executing organizations because the data allows constant monitoring,analyzing,and improving the underlying processes,which leads to the reduction of cost and the improvement of the quality.Process mining is a useful technique for analyzing enterprise systems by using an event log that contains behaviours.This research focuses on the process discovery and refinement using real-life event log data collected from a large multinational organization that deals with coatings and paints.By investigating and analyzing their order handling pro-cesses,this study aims at learning a model that gives insight inspection of the processes and performance analysis.Furthermore,the animation is also performed for the better inspection,diagnostics,and compliance-related questions to specify the system.The configuration of the system and the conformance checking for further enhancement is also addressed in this research.To achieve the objectives,this research uses process mining techniques,i.e.process discovery in the form of formal Petri nets models with the help of process maps,and process refinement through conformance checking and enhancement.Initially,the identified executed process is reconstructed by using the process discovery techniques.Following the reconstruction,we perform a deep analysis for the underlying process to ensure the process improvement and redesigning.Finally,some recommendations are made to improve the enterprise management system processes.
基金Project supported by the National Science Foundation of U.S.A.(Nos.DMS-1555072,DMS-2053746DMS-2134209)+1 种基金the Brookhaven National Laboratory of U.S.A.(No.382247)U.S.Department of Energy(DOE)Office of Science Advanced Scientific Computing Research Program(Nos.DESC0021142 and DE-SC0023161)。
文摘This paper develops a deep learning tool based on neural processes(NPs)called the Peri-Net-Pro,to predict the crack patterns in a moving disk and classifies them according to the classification modes with quantified uncertainties.In particular,image classification and regression studies are conducted by means of convolutional neural networks(CNNs)and NPs.First,the amount and quality of the data are enhanced by using peridynamics to theoretically compensate for the problems of the finite element method(FEM)in generating crack pattern images.Second,case studies are conducted with the prototype microelastic brittle(PMB),linear peridynamic solid(LPS),and viscoelastic solid(VES)models obtained by using the peridynamic theory.The case studies are performed to classify the images by using CNNs and determine the suitability of the PMB,LBS,and VES models.Finally,a regression analysis is performed on the crack pattern images with NPs to predict the crack patterns.The regression analysis results confirm that the variance decreases when the number of epochs increases by using the NPs.The training results gradually improve,and the variance ranges decrease to less than 0.035.The main finding of this study is that the NPs enable accurate predictions,even with missing or insufficient training data.The results demonstrate that if the context points are set to the 10th,100th,300th,and 784th,the training information is deliberately omitted for the context points of the 10th,100th,and 300th,and the predictions are different when the context points are significantly lower.However,the comparison of the results of the 100th and 784th context points shows that the predicted results are similar because of the Gaussian processes in the NPs.Therefore,if the NPs are employed for training,the missing information of the training data can be supplemented to predict the results.
基金supported in part by the National Natural Science Foundation of China(51825502,51775216)in part by the Program for Huazhong University of Science and Technology(HUST)Academic Frontier Youth Team(2017QYTD04).
文摘Intelligent process planning(PP)is one of the most important components in an intelligent manufacturing system and acts as a bridge between product designing and practical manufacturing.PP is a nondeterministic polynomial-time(NP)-hard problem and,as existing mathematical models are not formulated in linear forms,they cannot be solved well to achieve exact solutions for PP problems.This paper proposes a novel mixed-integer linear programming(MILP)mathematical model by considering the network topology structure and the OR nodes that represent a type of OR logic inside the network.Precedence relationships between operations are discussed by raising three types of precedence relationship matrices.Furthermore,the proposed model can be programmed in commonly-used mathematical programming solvers,such as CPLEX,Gurobi,and so forth,to search for optimal solutions for most open problems.To verify the effectiveness and generality of the proposed model,five groups of numerical experiments are conducted on well-known benchmarks.The results show that the proposed model can solve PP problems effectively and can obtain better solutions than those obtained by the state-ofthe-art algorithms.
文摘A correct and timely fault diagnosis is important for improving the safety and reliability of chemical processes. With the advancement of big data technology, data-driven fault diagnosis methods are being extensively used and still have considerable potential. In recent years, methods based on deep neural networks have made significant breakthroughs, and fault diagnosis methods for industrial processes based on deep learning have attracted considerable research attention. Therefore, we propose a fusion deeplearning algorithm based on a fully convolutional neural network(FCN) to extract features and build models to correctly diagnose all types of faults. We use long short-term memory(LSTM) units to expand our proposed FCN so that our proposed deep learning model can better extract the time-domain features of chemical process data. We also introduce the attention mechanism into the model, aimed at highlighting the importance of features, which is significant for the fault diagnosis of chemical processes with many features. When applied to the benchmark Tennessee Eastman process, our proposed model exhibits impressive performance, demonstrating the effectiveness of the attention-based LSTM FCN in chemical process fault diagnosis.
基金supported by the Fundamental Research Funds for The Central Universities(Grant No.2232021A-08)National Natural Science Foundation of China(GrantNo.51905091)Shanghai Sailing Program(Grand No.19YF1401500).
文摘The prognostics health management(PHM)fromthe systematic viewis critical to the healthy continuous operation of processmanufacturing systems(PMS),with different kinds of dynamic interference events.This paper proposes a three leveled digital twinmodel for the systematic PHMof PMSs.The unit-leveled digital twinmodel of each basic device unit of PMSs is constructed based on edge computing,which can provide real-time monitoring and analysis of the device status.The station-leveled digital twin models in the PMSs are designed to optimize and control the process parameters,which are deployed for the manufacturing execution on the fog server.The shop-leveled digital twin maintenancemodel is designed for production planning,which gives production instructions fromthe private industrial cloud server.To cope with the dynamic disturbances of a PMS,a big data-driven framework is proposed to control the three-level digital twin models,which contains indicator prediction,influence evaluation,and decisionmaking.Finally,a case study with a real chemical fiber system is introduced to illustrate the effectiveness of the digital twin model with edge-fog-cloud computing for the systematic PHM of PMSs.The result demonstrates that the three-leveled digital twin model for the systematic PHM in PMSs works well in the system’s respects.
基金Princess Nourah bint Abdulrahman University Researchers Supporting Project number(PNURSP2022R161)PrincessNourah bint Abdulrahman University,Riyadh,Saudi Arabia.The authors would like to thank the|Deanship of Scientific Research at Umm Al-Qura University|for supporting this work by Grant Code:(22UQU4310373DSR33).
文摘The recent developments in Multimedia Internet of Things(MIoT)devices,empowered with Natural Language Processing(NLP)model,seem to be a promising future of smart devices.It plays an important role in industrial models such as speech understanding,emotion detection,home automation,and so on.If an image needs to be captioned,then the objects in that image,its actions and connections,and any silent feature that remains under-projected or missing from the images should be identified.The aim of the image captioning process is to generate a caption for image.In next step,the image should be provided with one of the most significant and detailed descriptions that is syntactically as well as semantically correct.In this scenario,computer vision model is used to identify the objects and NLP approaches are followed to describe the image.The current study develops aNatural Language Processing with Optimal Deep Learning Enabled Intelligent Image Captioning System(NLPODL-IICS).The aim of the presented NLPODL-IICS model is to produce a proper description for input image.To attain this,the proposed NLPODL-IICS follows two stages such as encoding and decoding processes.Initially,at the encoding side,the proposed NLPODL-IICS model makes use of Hunger Games Search(HGS)with Neural Search Architecture Network(NASNet)model.This model represents the input data appropriately by inserting it into a predefined length vector.Besides,during decoding phase,Chimp Optimization Algorithm(COA)with deeper Long Short Term Memory(LSTM)approach is followed to concatenate the description sentences 4436 CMC,2023,vol.74,no.2 produced by the method.The application of HGS and COA algorithms helps in accomplishing proper parameter tuning for NASNet and LSTM models respectively.The proposed NLPODL-IICS model was experimentally validated with the help of two benchmark datasets.Awidespread comparative analysis confirmed the superior performance of NLPODL-IICS model over other models.
基金supported in part by the National Natural Science Foundation of China under Grant 62176109in part by the Tibetan Information Processing and Machine Translation Key Laboratory of Qinghai Province under Grant 2021‐Z‐003+3 种基金in part by the Natural Science Foundation of Gansu Province under Grant 21JR7RA531 and Grant 22JR5RA487in part by the Fundamental Research Funds for the Central Universities under Grant lzujbky‐2022‐23in part by the CAAI‐Huawei MindSpore Open Fund under Grant CAAIXSJLJJ‐2022‐020Ain part by the Supercomputing Center of Lanzhou University,in part by Sichuan Science and Technology Program No.2022nsfsc0916.
文摘A variety of neural networks have been presented to deal with issues in deep learning in the last decades.Despite the prominent success achieved by the neural network,it still lacks theoretical guidance to design an efficient neural network model,and verifying the performance of a model needs excessive resources.Previous research studies have demonstrated that many existing models can be regarded as different numerical discretizations of differential equations.This connection sheds light on designing an effective recurrent neural network(RNN)by resorting to numerical analysis.Simple RNN is regarded as a discretisation of the forward Euler scheme.Considering the limited solution accuracy of the forward Euler methods,a Taylor‐type discrete scheme is presented with lower truncation error and a Taylor‐type RNN(T‐RNN)is designed with its guidance.Extensive experiments are conducted to evaluate its performance on statistical language models and emotion analysis tasks.The noticeable gains obtained by T‐RNN present its superiority and the feasibility of designing the neural network model using numerical methods.
基金This work was financially supported by the National Key R&D Program of China(No.2020YFB1901900)National Natural Science Foundation of China(No.12275175)+2 种基金Special Fund for Strengthening Industry of Shanghai(No.GYQJ-2018-2-02)Shanghai Rising Star Program(No.21QA1404200)the LingChuang Research Project of the China National Nuclear Corporation.
文摘Micro-mobile heat pipe-cooled nuclear power plants are promising candidates for distributed energy resource power genera-tors and can be flexibly deployed in remote places to meet increasing electric power demands.However,previous steady-state simulations and experiments have deviated significantly from actual micronuclear system operations.Hence,a transient analysis is required for performance optimization and safety assessment.In this study,a hardware-in-the-loop(HIL)approach was used to investigate the dynamic behavior of scaled-down heat pipe-cooled systems.The real-time features of the HIL architecture were interpreted and validated,and an optimal time step of 500 ms was selected for the thermal transient.The power transient was modeled using point kinetic equations,and a scaled-down thermal prototype was set up to avoid mod-eling unpredictable heat transfer behaviors and feeding temperature samples into the main program running on a desktop PC.A series of dynamic test results showed significant power and temperature oscillations during the transient process,owing to the inconsistency of the rapid nuclear reaction rate and large thermal inertia.The proposed HIL approach is stable and effective for further studying of the dynamic characteristics and control optimization of solid-state small nuclear-powered systems at an early prototyping stage.
基金supported by National Natural Science Foundation of China(Grant No.61761011)Natural Science Foundation of Guangxi(Grant No.2020GXNSFBA297078).
文摘In the graph signal processing(GSP)framework,distributed algorithms are highly desirable in processing signals defined on large-scale networks.However,in most existing distributed algorithms,all nodes homogeneously perform the local computation,which calls for heavy computational and communication costs.Moreover,in many real-world networks,such as those with straggling nodes,the homogeneous manner may result in serious delay or even failure.To this end,we propose active network decomposition algorithms to select non-straggling nodes(normal nodes)that perform the main computation and communication across the network.To accommodate the decomposition in different kinds of networks,two different approaches are developed,one is centralized decomposition that leverages the adjacency of the network and the other is distributed decomposition that employs the indicator message transmission between neighboring nodes,which constitutes the main contribution of this paper.By incorporating the active decomposition scheme,a distributed Newton method is employed to solve the least squares problem in GSP,where the Hessian inverse is approximately evaluated by patching a series of inverses of local Hessian matrices each of which is governed by one normal node.The proposed algorithm inherits the fast convergence of the second-order algorithms while maintains low computational and communication cost.Numerical examples demonstrate the effectiveness of the proposed algorithm.
基金supported by the National Natural Science Foundation of China(NSFC,no.61701243,71771125)the Major Project of Natural Science Foundation of Jiangsu Education Department(no.19KJA180002).
文摘Depression has become one of the most common mental illnesses in the world.For better prediction and diagnosis,methods of automatic depression recognition based on speech signal are constantly proposed and updated,with a transition from the early traditional methods based on hand‐crafted features to the application of architectures of deep learning.This paper systematically and precisely outlines the most prominent and up‐to‐date research of automatic depression recognition by intelligent speech signal processing so far.Furthermore,methods for acoustic feature extraction,algorithms for classification and regression,as well as end to end deep models are investigated and analysed.Finally,general trends are summarised and key unresolved issues are identified to be considered in future studies of automatic speech depression recognition.