High-precision and real-time diagnosis of sucker rod pumping system(SRPS)is important for quickly mastering oil well operations.Deep learning-based method for classifying the dynamometer card(DC)of oil wells is an eff...High-precision and real-time diagnosis of sucker rod pumping system(SRPS)is important for quickly mastering oil well operations.Deep learning-based method for classifying the dynamometer card(DC)of oil wells is an efficient diagnosis method.However,the input of the DC as a two-dimensional image into the deep learning framework suffers from low feature utilization and high computational effort.Additionally,different SRPSs in an oil field have various system parameters,and the same SRPS generates different DCs at different moments.Thus,there is heterogeneity in field data,which can dramatically impair the diagnostic accuracy.To solve the above problems,a working condition recognition method based on 4-segment time-frequency signature matrix(4S-TFSM)and deep learning is presented in this paper.First,the 4-segment time-frequency signature(4S-TFS)method that can reduce the computing power requirements is proposed for feature extraction of DC data.Subsequently,the 4S-TFSM is constructed by relative normalization and matrix calculation to synthesize the features of multiple data and solve the problem of data heterogeneity.Finally,a convolutional neural network(CNN),one of the deep learning frameworks,is used to determine the functioning conditions based on the 4S-TFSM.Experiments on field data verify that the proposed diagnostic method based on 4S-TFSM and CNN(4S-TFSM-CNN)can significantly improve the accuracy of working condition recognition with lower computational cost.To the best of our knowledge,this is the first work to discuss the effect of data heterogeneity on the working condition recognition performance of SRPS.展开更多
In recent years,semantic segmentation on 3D point cloud data has attracted much attention.Unlike 2D images where pixels distribute regularly in the image domain,3D point clouds in non-Euclidean space are irregular and...In recent years,semantic segmentation on 3D point cloud data has attracted much attention.Unlike 2D images where pixels distribute regularly in the image domain,3D point clouds in non-Euclidean space are irregular and inherently sparse.Therefore,it is very difficult to extract long-range contexts and effectively aggregate local features for semantic segmentation in 3D point cloud space.Most current methods either focus on local feature aggregation or long-range context dependency,but fail to directly establish a global-local feature extractor to complete the point cloud semantic segmentation tasks.In this paper,we propose a Transformer-based stratified graph convolutional network(SGT-Net),which enlarges the effective receptive field and builds direct long-range dependency.Specifically,we first propose a novel dense-sparse sampling strategy that provides dense local vertices and sparse long-distance vertices for subsequent graph convolutional network(GCN).Secondly,we propose a multi-key self-attention mechanism based on the Transformer to further weight augmentation for crucial neighboring relationships and enlarge the effective receptive field.In addition,to further improve the efficiency of the network,we propose a similarity measurement module to determine whether the neighborhood near the center point is effective.We demonstrate the validity and superiority of our method on the S3DIS and ShapeNet datasets.Through ablation experiments and segmentation visualization,we verify that the SGT model can improve the performance of the point cloud semantic segmentation.展开更多
We described ferromagnetic film and bilayer films composed of two ferromagnetic layers coupled through antiferromagnetic interfacial interaction by classical Heisenberg model and simulated their magnetization state,ma...We described ferromagnetic film and bilayer films composed of two ferromagnetic layers coupled through antiferromagnetic interfacial interaction by classical Heisenberg model and simulated their magnetization state,magnetic permeability,and Faraday effect at zero and finite temperature by using the Landau–Lifshitz–Gilbert(LLG)equation.The results indicate that in a microwave field with positive circular polarization,the ferromagnetic film has one resonance peak while the bilayer film has two resonance peaks.However,the resonance peak disappears in ferromagnetic film,and only one resonance peak emerges in bilayer film in the negative circularly polarized microwave field.When the microwave field’s frequency exceeds the film’s resonance frequency,the Faraday rotation angle of the ferromagnetic film is the greatest,and it decreases when the thickness of the two halves of the bilayer is reduced.When the microwave field’s frequency remains constant,the Faraday rotation angle fluctuates with temperature in the same manner as spontaneous magnetization does.When a DC magnetic field is applied in the direction of the anisotropic axis of the film,the Faraday rotation angle varies with the DC magnetic field and shows a similar shape of the hysteresis loop.展开更多
We propose a novel image segmentation algorithm to tackle the challenge of limited recognition and segmentation performance in identifying welding seam images during robotic intelligent operations.Initially,to enhance...We propose a novel image segmentation algorithm to tackle the challenge of limited recognition and segmentation performance in identifying welding seam images during robotic intelligent operations.Initially,to enhance the capability of deep neural networks in extracting geometric attributes from depth images,we developed a novel deep geometric convolution operator(DGConv).DGConv is utilized to construct a deep local geometric feature extraction module,facilitating a more comprehensive exploration of the intrinsic geometric information within depth images.Secondly,we integrate the newly proposed deep geometric feature module with the Fully Convolutional Network(FCN8)to establish a high-performance deep neural network algorithm tailored for depth image segmentation.Concurrently,we enhance the FCN8 detection head by separating the segmentation and classification processes.This enhancement significantly boosts the network’s overall detection capability.Thirdly,for a comprehensive assessment of our proposed algorithm and its applicability in real-world industrial settings,we curated a line-scan image dataset featuring weld seams.This dataset,named the Standardized Linear Depth Profile(SLDP)dataset,was collected from actual industrial sites where autonomous robots are in operation.Ultimately,we conducted experiments utilizing the SLDP dataset,achieving an average accuracy of 92.7%.Our proposed approach exhibited a remarkable performance improvement over the prior method on the identical dataset.Moreover,we have successfully deployed the proposed algorithm in genuine industrial environments,fulfilling the prerequisites of unmanned robot operations.展开更多
Time-sensitive networks(TSNs)support not only traditional best-effort communications but also deterministic communications,which send each packet at a deterministic time so that the data transmissions of networked con...Time-sensitive networks(TSNs)support not only traditional best-effort communications but also deterministic communications,which send each packet at a deterministic time so that the data transmissions of networked control systems can be precisely scheduled to guarantee hard real-time constraints.No-wait scheduling is suitable for such TSNs and generates the schedules of deterministic communications with the minimal network resources so that all of the remaining resources can be used to improve the throughput of best-effort communications.However,due to inappropriate message fragmentation,the realtime performance of no-wait scheduling algorithms is reduced.Therefore,in this paper,joint algorithms of message fragmentation and no-wait scheduling are proposed.First,a specification for the joint problem based on optimization modulo theories is proposed so that off-the-shelf solvers can be used to find optimal solutions.Second,to improve the scalability of our algorithm,the worst-case delay of messages is analyzed,and then,based on the analysis,a heuristic algorithm is proposed to construct low-delay schedules.Finally,we conduct extensive test cases to evaluate our proposed algorithms.The evaluation results indicate that,compared to existing algorithms,the proposed joint algorithm improves schedulability by up to 50%.展开更多
The accurate and intelligent identification of the working conditions of a sucker-rod pumping system is necessary. As onshore oil extraction gradually enters its mid-to late-stage, the cost required to train a deep le...The accurate and intelligent identification of the working conditions of a sucker-rod pumping system is necessary. As onshore oil extraction gradually enters its mid-to late-stage, the cost required to train a deep learning working condition recognition model for pumping wells by obtaining enough new working condition samples is expensive. For the few-shot problem and large calculation issues of new working conditions of oil wells, a working condition recognition method for pumping unit wells based on a 4-dimensional time-frequency signature (4D-TFS) and meta-learning convolutional shrinkage neural network (ML-CSNN) is proposed. First, the measured pumping unit well workup data are converted into 4D-TFS data, and the initial feature extraction task is performed while compressing the data. Subsequently, a convolutional shrinkage neural network (CSNN) with a specific structure that can ablate low-frequency features is designed to extract working conditions features. Finally, a meta-learning fine-tuning framework for learning the network parameters that are susceptible to task changes is merged into the CSNN to solve the few-shot issue. The results of the experiments demonstrate that the trained ML-CSNN has good recognition accuracy and generalization ability for few-shot working condition recognition. More specifically, in the case of lower computational complexity, only few-shot samples are needed to fine-tune the network parameters, and the model can be quickly adapted to new classes of well conditions.展开更多
Large-scale solar sails can provide power to spacecraft for deep space exploration.A new type of telescopic tubular mast(TTM)driven by a bistable carbon fiber-reinforced polymer tube was designed in this study to solv...Large-scale solar sails can provide power to spacecraft for deep space exploration.A new type of telescopic tubular mast(TTM)driven by a bistable carbon fiber-reinforced polymer tube was designed in this study to solve the problem of contact between the sail membrane and the spacecraft under light pressure.Compared with the traditional TTM,it has a small size,light weight,high extension ratio,and simple structure.The anti-blossoming and self-unlocking structure of the proposed TTM was described.We aimed to simplify the TTM with a complex structure into a beam model with equal linear mass density,and the simulation results showed good consistency.The dynamic equation was derived based on the equivalent model,and the effects of different factors on the vibration characteristics of the TTM were analyzed.The performance parameters were optimized based on a multiobjective genetic algorithm,and prototype production and load experiments were conducted.The results show that the advantages of the new TTM can complete the deployment of large-scale solar sails,which is valuable for future deep space exploration.展开更多
Cell mechanics is essential to cell development and function,and its dynamics evolution reflects the physiological state of cells.Here,we investigate the dynamical mechanical properties of single cells under various d...Cell mechanics is essential to cell development and function,and its dynamics evolution reflects the physiological state of cells.Here,we investigate the dynamical mechanical properties of single cells under various drug conditions,and present two mathematical approaches to quantitatively characterizing the cell physiological state.It is demonstrated that the cellular mechanical properties upon the drug action increase over time and tend to saturate,and can be mathematically characterized by a linear timeinvariant dynamical model.It is shown that the transition matrices of dynamical cell systems significantly improve the classification accuracies of the cells under different drug actions.Furthermore,it is revealed that there exists a positive linear correlation between the cytoskeleton density and the cellular mechanical properties,and the physiological state of a cell in terms of its cytoskeleton density can be predicted from its mechanical properties by a linear regression model.This study builds a relationship between the cellular mechanical properties and the cellular physiological state,adding information for evaluating drug efficacy.展开更多
Terahertz(THz)imaging has drawn significant attention because THz wave has a unique capability to transient,ultrawide spectrum and low photon energy.However,the low resolution has always been a problem due to its long...Terahertz(THz)imaging has drawn significant attention because THz wave has a unique capability to transient,ultrawide spectrum and low photon energy.However,the low resolution has always been a problem due to its long wavelength,limiting their application of fields practical use.In this paper,we proposed a complex one-shot super-resolution(COSSR)framework based on a complex convolution neural network to restore superior THz images at 0.35 times wavelength by extracting features directly from a reference measured sample and groundtruth without the measured PSF.Compared with real convolution neural network-based approaches and complex zero-shot super-resolution(CZSSR),COSSR delivers at least 6.67,0.003,and 6.96%superior higher imaging efficacy in terms of peak signal to noise ratio(PSNR),mean square error(MSE),and structural similarity index measure(SSIM),respectively,for the analyzed data.Additionally,the proposed method is experimentally demonstrated to have a good generalization and to perform well on measured data.The COSSR provides a new pathway for THz imaging super-resolution(SR)reconstruction below the diffraction limit.展开更多
As 3D acquisition technology develops and 3D sensors become increasingly affordable,large quantities of 3D point cloud data are emerging.How to effectively learn and extract the geometric features from these point clo...As 3D acquisition technology develops and 3D sensors become increasingly affordable,large quantities of 3D point cloud data are emerging.How to effectively learn and extract the geometric features from these point clouds has become an urgent problem to be solved.The point cloud geometric information is hidden in disordered,unstructured points,making point cloud analysis a very challenging problem.To address this problem,we propose a novel network framework,called Tree Graph Network(TGNet),which can sample,group,and aggregate local geometric features.Specifically,we construct a Tree Graph by explicit rules,which consists of curves extending in all directions in point cloud feature space,and then aggregate the features of the graph through a cross-attention mechanism.In this way,we incorporate more point cloud geometric structure information into the representation of local geometric features,which makes our network perform better.Our model performs well on several basic point clouds processing tasks such as classification,segmentation,and normal estimation,demonstrating the effectiveness and superiority of our network.Furthermore,we provide ablation experiments and visualizations to better understand our network.展开更多
The widespread usage of Cyber Physical Systems(CPSs)generates a vast volume of time series data,and precisely determining anomalies in the data is critical for practical production.Autoencoder is the mainstream method...The widespread usage of Cyber Physical Systems(CPSs)generates a vast volume of time series data,and precisely determining anomalies in the data is critical for practical production.Autoencoder is the mainstream method for time series anomaly detection,and the anomaly is judged by reconstruction error.However,due to the strong generalization ability of neural networks,some abnormal samples close to normal samples may be judged as normal,which fails to detect the abnormality.In addition,the dataset rarely provides sufficient anomaly labels.This research proposes an unsupervised anomaly detection approach based on adversarial memory autoencoders for multivariate time series to solve the above problem.Firstly,an encoder encodes the input data into low-dimensional space to acquire a feature vector.Then,a memory module is used to learn the feature vector’s prototype patterns and update the feature vectors.The updating process allows partial forgetting of information to prevent model overgeneralization.After that,two decoders reconstruct the input data.Finally,this research uses the Peak Over Threshold(POT)method to calculate the threshold to determine anomalous samples from normal samples.This research uses a two-stage adversarial training strategy during model training to enlarge the gap between the reconstruction error of normal and abnormal samples.The proposed method achieves significant anomaly detection results on synthetic and real datasets from power systems,water treatment plants,and computer clusters.The F1 score reached an average of 0.9196 on the five datasets,which is 0.0769 higher than the best baseline method.展开更多
Feature extraction plays an important role in constructing artificial intel-ligence(AI)models of industrial control systems(ICSs).Three challenges in this field are learning effective representation from high-dimensio...Feature extraction plays an important role in constructing artificial intel-ligence(AI)models of industrial control systems(ICSs).Three challenges in this field are learning effective representation from high-dimensional features,data heterogeneity,and data noise due to the diversity of data dimensions,formats and noise of sensors,controllers and actuators.Hence,a novel unsupervised learn-ing autoencoder model is proposed for ICS data in this paper.Although traditional methods only capture the linear correlations of ICS features,our deep industrial representation learning model(DIRL)based on a convolutional neural network can mine high-order features,thus solving the problem of high-dimensional and heterogeneous ICS data.In addition,an unsupervised denoising autoencoder is introduced for noisy ICS data in DIRL.Training the denoising autoencoder allows the model to better mitigate the sensor noise problem.In this way,the represen-tative features learned by DIRL could help to evaluate the safety state of ICSs more effectively.We tested our model with absolute and relative accuracy experi-ments on two large-scale ICS datasets.Compared with other popular methods,DIRL showed advantages in four common indicators of AI algorithms:accuracy,precision,recall,and F1-score.This study contributes to the effective analysis of large-scale ICS data,which promotes the stable operation of ICSs.展开更多
Rapid online analysis of liquid slag is essential for optimizing the quality and energy efficiency of steel production. To investigate the key factors that affect the online measurement of refined slag using laser-ind...Rapid online analysis of liquid slag is essential for optimizing the quality and energy efficiency of steel production. To investigate the key factors that affect the online measurement of refined slag using laser-induced breakdown spectroscopy(LIBS), this study examined the effects of slag composition and temperature on the intensity and stability of the LIBS spectra. The experimental temperature was controlled at three levels: 1350℃, 1400℃, and 1450℃. The results showed that slag composition and temperature significantly affected the intensity and stability of the LIBS spectra. Increasing the Fe content and temperature in the slag reduces its viscosity, resulting in an enhanced intensity and stability of the LIBS spectra. Additionally, 42 refined slag samples were quantitatively analyzed for Fe, Si, Ca, Mg, Al, and Mn at 1350℃, 1400℃, and 1450℃.The normalized full spectrum combined with partial least squares(PLS) quantification modeling was used, using the Ca Ⅱ 317.91 nm spectral line as an internal standard. The results show that using the internal standard normalization method can significantly reduce the influence of spectral fluctuations. Meanwhile, a temperature of 1450℃ has been found to yield superior results compared to both 1350℃ and 1400℃, and it is advantageous to conduct a quantitative analysis of the slag when it is in a “water-like” state with low viscosity.展开更多
This paper investigates a new vortex wave imaging approach to improve the imaging quality of small metal targets of size less than 1.5 mm.Antennas with different spiral phase plates are designed to efficiently transmi...This paper investigates a new vortex wave imaging approach to improve the imaging quality of small metal targets of size less than 1.5 mm.Antennas with different spiral phase plates are designed to efficiently transmit vortex beams with orbital angular momentums(OAMs).By analyzing the OAM spectrum of the target,it was discovered that the predominant reflection contains a particular OAM mode that carries abundant azimuthal information.This can be explained by the OAM selectivity of the target and the guidance of the vortex transmitting beam.A simple reflection vortex imaging system was designed to capture the phase information.Measurement results show that the high image contrast reaches 14.9%,which is twice as high as that of the imaging without OAM.Both of simulations and experiments demonstrate that the vortex phase imaging approach proposed in this paper can effectively improve the imaging quality at 80 GHz.This approach is suitable for other millimeter wave imaging systems and is helpful to improve the resolution in anti-terrorism security checks.展开更多
A long history has passed since electromyography(EMG)signals have been explored in human-centered robots for intuitive interaction.However,it still has a gap between scientific research and real-life applications.Prev...A long history has passed since electromyography(EMG)signals have been explored in human-centered robots for intuitive interaction.However,it still has a gap between scientific research and real-life applications.Previous studies mainly focused on EMG decoding algorithms,leaving a dynamic relationship between the human,robot,and uncertain environment in real-life scenarios seldomly concerned.To fill this gap,this paper presents a comprehensive review of EMG-based techniques in human-robot-environment interaction(HREI)systems.The general processing framework is summarized,and three interaction paradigms,including direct control,sensory feedback,and partial autonomous control,are introduced.EMG-based intention decoding is treated as a module of the proposed paradigms.Five key issues involving precision,stability,user attention,compliance,and environmental awareness in this field are discussed.Several important directions,including EMG decomposition,robust algorithms,HREI dataset,proprioception feedback,reinforcement learning,and embodied intelligence,are proposed to pave the way for future research.To the best of what we know,this is the first time that a review of EMG-based methods in the HREI system is summarized.It provides a novel and broader perspective to improve the practicability of current myoelectric interaction systems,in which factors in human-robot interaction,robot-environment interaction,and state perception by human sensations are considered,which has never been done by previous studies.展开更多
Electromyography(EMG)has already been broadly used in human-machine interaction(HMI)applications.Determining how to decode the information inside EMG signals robustly and accurately is a key problem for which we urgen...Electromyography(EMG)has already been broadly used in human-machine interaction(HMI)applications.Determining how to decode the information inside EMG signals robustly and accurately is a key problem for which we urgently need a solution.Recently,many EMG pattern recognition tasks have been addressed using deep learning methods.In this paper,we analyze recent papers and present a literature review describing the role that deep learning plays in EMG-based HMI.An overview of typical network structures and processing schemes will be provided.Recent progress in typical tasks such as movement classification,joint angle prediction,and force/torque estimation will be introduced.New issues,including multimodal sensing,inter-subject/inter-session,and robustness toward disturbances will be discussed.We attempt to provide a comprehensive analysis of current research by discussing the advantages,challenges,and opportunities brought by deep learning.We hope that deep learning can aid in eliminating factors that hinder the development of EMG-based HMI systems.Furthermore,possible future directions will be presented to pave the way for future research.展开更多
Supernumerary robotic limbs(SRLs) are a new type of wearable human auxiliary equipment, which is currently a hot research topic in the world. SRLs have broad applications in many fields, and will provide a reference a...Supernumerary robotic limbs(SRLs) are a new type of wearable human auxiliary equipment, which is currently a hot research topic in the world. SRLs have broad applications in many fields, and will provide a reference and technical support for the realization of human-robot collaboration and integration,while playing an important role in improving social security and public services. In this paper, representative SRLs are summarized from the aspects of related literature analysis,research status, ontology structure design, control and driving,sensing and perception, and application fields. This paper also analyzes and summarizes the current technical challenges faced by SRLs, and reviews development progress and key technologies,thus giving a prospect of future technical development trends.展开更多
In this work,the nickel-based powder metallurgy superalloy FGH95 was selected as experimental material,and the experimental parameters in multiple overlap laser shock processing(LSP)treatment were selected based on or...In this work,the nickel-based powder metallurgy superalloy FGH95 was selected as experimental material,and the experimental parameters in multiple overlap laser shock processing(LSP)treatment were selected based on orthogonal experimental design.The experimental data of residual stress and microhardness were measured in the same depth.The residual stress and microhardness laws were investigated and analyzed.Artificial neural network(ANN)with four layers(4-N-(N-1)-2)was applied to predict the residual stress and microhardness of FGH95 subjected to multiple overlap LSP.The experimental data were divided as training-testing sets in pairs.Laser energy,overlap rate,shocked times and depth were set as inputs,while residual stress and microhardness were set as outputs.The prediction performances with different network configuration of developed ANN models were compared and analyzed.The developed ANN model with network configuration of 4-7-6-2 showed the best predict performance.The predicted values showed a good agreement with the experimental values.In addition,the correlation coefficients among all the parameters and the effect of LSP parameters on materials response were studied.It can be concluded that ANN is a useful method to predict residual stress and microhardness of material subjected to LSP when with limited experimental data.展开更多
In the spectral analysis of laser-induced breakdown spectroscopy,abundant characteristic spectral lines and severe interference information exist simultaneously in the original spectral data.Here,a feature selection m...In the spectral analysis of laser-induced breakdown spectroscopy,abundant characteristic spectral lines and severe interference information exist simultaneously in the original spectral data.Here,a feature selection method called recursive feature elimination based on ridge regression(Ridge-RFE)for the original spectral data is recommended to make full use of the valid information of spectra.In the Ridge-RFE method,the absolute value of the ridge regression coefficient was used as a criterion to screen spectral characteristic,the feature with the absolute value of minimum weight in the input subset features was removed by recursive feature elimination(RFE),and the selected features were used as inputs of the partial least squares regression(PLS)model.The Ridge-RFE method based PLS model was used to measure the Fe,Si,Mg,Cu,Zn and Mn for 51 aluminum alloy samples,and the results showed that the root mean square error of prediction decreased greatly compared to the PLS model with full spectrum as input.The overall results demonstrate that the Ridge-RFE method is more efficient to extract the redundant features,make PLS model for better quantitative analysis results and improve model generalization ability.展开更多
Range is an important factor to the design of autonomous underwater vehicles (AUVs), while drag reduction efforts are pursued, the investigation of body-propeller interaction is another vital consideration. We present...Range is an important factor to the design of autonomous underwater vehicles (AUVs), while drag reduction efforts are pursued, the investigation of body-propeller interaction is another vital consideration. We present a numerical and experimental study of the hull-propeller interaction for deeply submerged underwater vehicles, using a proportional-integral- derivative (PID) controller method to estimate self-propulsion point in CFD environment. The hydrodynamic performance of hull and propeller at the balance state when the AUV sails at a fixed depth is investigated, using steady RANS solver of Star-CCM+. The proposed steady RANS solver takes only hours to reach a reasonable solution. It is more time efficient than unsteady simulations which takes days or weeks, as well as huge consumption of computing resources. Explorer 1000, a long range AUV developed by Shenyang Institute of Automation, Chinese Academy of Sciences, was studied as an object, and self-propulsion point, thrust deduction, wake fraction and hull efficiency were analyzed by using the proposed RANS method. Behind-hull performance of the selected propeller MAU4-40, as well as the hull-propeller interaction, was obtained from the computed hydrodynamic forces. The numerical results are in good qualitative and quantitative agreement with the experimental results obtained in the Qiandao Lake of Zhejiang province, China.展开更多
基金We would like to thank the associate editor and the reviewers for their constructive comments.This work was supported in part by the National Natural Science Foundation of China under Grant 62203234in part by the State Key Laboratory of Robotics of China under Grant 2023-Z03+1 种基金in part by the Natural Science Foundation of Liaoning Province under Grant 2023-BS-025in part by the Research Program of Liaoning Liaohe Laboratory under Grant LLL23ZZ-02-02.
文摘High-precision and real-time diagnosis of sucker rod pumping system(SRPS)is important for quickly mastering oil well operations.Deep learning-based method for classifying the dynamometer card(DC)of oil wells is an efficient diagnosis method.However,the input of the DC as a two-dimensional image into the deep learning framework suffers from low feature utilization and high computational effort.Additionally,different SRPSs in an oil field have various system parameters,and the same SRPS generates different DCs at different moments.Thus,there is heterogeneity in field data,which can dramatically impair the diagnostic accuracy.To solve the above problems,a working condition recognition method based on 4-segment time-frequency signature matrix(4S-TFSM)and deep learning is presented in this paper.First,the 4-segment time-frequency signature(4S-TFS)method that can reduce the computing power requirements is proposed for feature extraction of DC data.Subsequently,the 4S-TFSM is constructed by relative normalization and matrix calculation to synthesize the features of multiple data and solve the problem of data heterogeneity.Finally,a convolutional neural network(CNN),one of the deep learning frameworks,is used to determine the functioning conditions based on the 4S-TFSM.Experiments on field data verify that the proposed diagnostic method based on 4S-TFSM and CNN(4S-TFSM-CNN)can significantly improve the accuracy of working condition recognition with lower computational cost.To the best of our knowledge,this is the first work to discuss the effect of data heterogeneity on the working condition recognition performance of SRPS.
基金supported in part by the National Natural Science Foundation of China under Grant Nos.U20A20197,62306187the Foundation of Ministry of Industry and Information Technology TC220H05X-04.
文摘In recent years,semantic segmentation on 3D point cloud data has attracted much attention.Unlike 2D images where pixels distribute regularly in the image domain,3D point clouds in non-Euclidean space are irregular and inherently sparse.Therefore,it is very difficult to extract long-range contexts and effectively aggregate local features for semantic segmentation in 3D point cloud space.Most current methods either focus on local feature aggregation or long-range context dependency,but fail to directly establish a global-local feature extractor to complete the point cloud semantic segmentation tasks.In this paper,we propose a Transformer-based stratified graph convolutional network(SGT-Net),which enlarges the effective receptive field and builds direct long-range dependency.Specifically,we first propose a novel dense-sparse sampling strategy that provides dense local vertices and sparse long-distance vertices for subsequent graph convolutional network(GCN).Secondly,we propose a multi-key self-attention mechanism based on the Transformer to further weight augmentation for crucial neighboring relationships and enlarge the effective receptive field.In addition,to further improve the efficiency of the network,we propose a similarity measurement module to determine whether the neighborhood near the center point is effective.We demonstrate the validity and superiority of our method on the S3DIS and ShapeNet datasets.Through ablation experiments and segmentation visualization,we verify that the SGT model can improve the performance of the point cloud semantic segmentation.
基金the Research Program of Shenyang Institute of Science and Technology(Grant No.ZD-2024-05).
文摘We described ferromagnetic film and bilayer films composed of two ferromagnetic layers coupled through antiferromagnetic interfacial interaction by classical Heisenberg model and simulated their magnetization state,magnetic permeability,and Faraday effect at zero and finite temperature by using the Landau–Lifshitz–Gilbert(LLG)equation.The results indicate that in a microwave field with positive circular polarization,the ferromagnetic film has one resonance peak while the bilayer film has two resonance peaks.However,the resonance peak disappears in ferromagnetic film,and only one resonance peak emerges in bilayer film in the negative circularly polarized microwave field.When the microwave field’s frequency exceeds the film’s resonance frequency,the Faraday rotation angle of the ferromagnetic film is the greatest,and it decreases when the thickness of the two halves of the bilayer is reduced.When the microwave field’s frequency remains constant,the Faraday rotation angle fluctuates with temperature in the same manner as spontaneous magnetization does.When a DC magnetic field is applied in the direction of the anisotropic axis of the film,the Faraday rotation angle varies with the DC magnetic field and shows a similar shape of the hysteresis loop.
基金This work was supported by the National Natural Science Foundation of China(Grant No.U20A20197).
文摘We propose a novel image segmentation algorithm to tackle the challenge of limited recognition and segmentation performance in identifying welding seam images during robotic intelligent operations.Initially,to enhance the capability of deep neural networks in extracting geometric attributes from depth images,we developed a novel deep geometric convolution operator(DGConv).DGConv is utilized to construct a deep local geometric feature extraction module,facilitating a more comprehensive exploration of the intrinsic geometric information within depth images.Secondly,we integrate the newly proposed deep geometric feature module with the Fully Convolutional Network(FCN8)to establish a high-performance deep neural network algorithm tailored for depth image segmentation.Concurrently,we enhance the FCN8 detection head by separating the segmentation and classification processes.This enhancement significantly boosts the network’s overall detection capability.Thirdly,for a comprehensive assessment of our proposed algorithm and its applicability in real-world industrial settings,we curated a line-scan image dataset featuring weld seams.This dataset,named the Standardized Linear Depth Profile(SLDP)dataset,was collected from actual industrial sites where autonomous robots are in operation.Ultimately,we conducted experiments utilizing the SLDP dataset,achieving an average accuracy of 92.7%.Our proposed approach exhibited a remarkable performance improvement over the prior method on the identical dataset.Moreover,we have successfully deployed the proposed algorithm in genuine industrial environments,fulfilling the prerequisites of unmanned robot operations.
基金partially supported by National Key Research and Development Program of China(2018YFB1700200)National Natural Science Foundation of China(61972389,61903356,61803368,U1908212)+2 种基金Youth Innovation Promotion Association of the Chinese Academy of Sciences,National Science and Technology Major Project(2017ZX02101007-004)Liaoning Provincial Natural Science Foundation of China(2020-MS-034,2019-YQ-09)China Postdoctoral Science Foundation(2019M661156)。
文摘Time-sensitive networks(TSNs)support not only traditional best-effort communications but also deterministic communications,which send each packet at a deterministic time so that the data transmissions of networked control systems can be precisely scheduled to guarantee hard real-time constraints.No-wait scheduling is suitable for such TSNs and generates the schedules of deterministic communications with the minimal network resources so that all of the remaining resources can be used to improve the throughput of best-effort communications.However,due to inappropriate message fragmentation,the realtime performance of no-wait scheduling algorithms is reduced.Therefore,in this paper,joint algorithms of message fragmentation and no-wait scheduling are proposed.First,a specification for the joint problem based on optimization modulo theories is proposed so that off-the-shelf solvers can be used to find optimal solutions.Second,to improve the scalability of our algorithm,the worst-case delay of messages is analyzed,and then,based on the analysis,a heuristic algorithm is proposed to construct low-delay schedules.Finally,we conduct extensive test cases to evaluate our proposed algorithms.The evaluation results indicate that,compared to existing algorithms,the proposed joint algorithm improves schedulability by up to 50%.
基金supported in part by the National Natural Science Foundation of China under Grant U1908212,62203432 and 92067205in part by the State Key Laboratory of Robotics of China under Grant 2023-Z03 and 2023-Z15in part by the Natural Science Foundation of Liaoning Province under Grant 2020-KF-11-02.
文摘The accurate and intelligent identification of the working conditions of a sucker-rod pumping system is necessary. As onshore oil extraction gradually enters its mid-to late-stage, the cost required to train a deep learning working condition recognition model for pumping wells by obtaining enough new working condition samples is expensive. For the few-shot problem and large calculation issues of new working conditions of oil wells, a working condition recognition method for pumping unit wells based on a 4-dimensional time-frequency signature (4D-TFS) and meta-learning convolutional shrinkage neural network (ML-CSNN) is proposed. First, the measured pumping unit well workup data are converted into 4D-TFS data, and the initial feature extraction task is performed while compressing the data. Subsequently, a convolutional shrinkage neural network (CSNN) with a specific structure that can ablate low-frequency features is designed to extract working conditions features. Finally, a meta-learning fine-tuning framework for learning the network parameters that are susceptible to task changes is merged into the CSNN to solve the few-shot issue. The results of the experiments demonstrate that the trained ML-CSNN has good recognition accuracy and generalization ability for few-shot working condition recognition. More specifically, in the case of lower computational complexity, only few-shot samples are needed to fine-tune the network parameters, and the model can be quickly adapted to new classes of well conditions.
基金Supported by National Key R&D Program of China (Grant No.2018YFB1304600)National Natural Science Foundation of China (Grant No.51905527)+1 种基金CAS Interdisciplinary Innovation Team of China (Grant No.JCTD-2018-11)State Key Laboratory of Robotics Foundation of China (Grant No.Y91Z0303)。
文摘Large-scale solar sails can provide power to spacecraft for deep space exploration.A new type of telescopic tubular mast(TTM)driven by a bistable carbon fiber-reinforced polymer tube was designed in this study to solve the problem of contact between the sail membrane and the spacecraft under light pressure.Compared with the traditional TTM,it has a small size,light weight,high extension ratio,and simple structure.The anti-blossoming and self-unlocking structure of the proposed TTM was described.We aimed to simplify the TTM with a complex structure into a beam model with equal linear mass density,and the simulation results showed good consistency.The dynamic equation was derived based on the equivalent model,and the effects of different factors on the vibration characteristics of the TTM were analyzed.The performance parameters were optimized based on a multiobjective genetic algorithm,and prototype production and load experiments were conducted.The results show that the advantages of the new TTM can complete the deployment of large-scale solar sails,which is valuable for future deep space exploration.
基金This work was supported by the National Natural Science Foundation of China(Grant Nos:U1908215,61925307,62003338,and 61933008)CAS Project for Young Scientists in Basic Research(Grant No:YSBR-041)+2 种基金Liaoning Revitalization Talents Program(Grant No:XLYC2002014)Natural Science Foundation of Liaoning Province of China(Grant No:2020-ZLLH-47)Joint fund of Science&Technology Department of Liaoning Province and State Key Laboratory of Robotics,China(Grant No:2019-KF-01-01).
文摘Cell mechanics is essential to cell development and function,and its dynamics evolution reflects the physiological state of cells.Here,we investigate the dynamical mechanical properties of single cells under various drug conditions,and present two mathematical approaches to quantitatively characterizing the cell physiological state.It is demonstrated that the cellular mechanical properties upon the drug action increase over time and tend to saturate,and can be mathematically characterized by a linear timeinvariant dynamical model.It is shown that the transition matrices of dynamical cell systems significantly improve the classification accuracies of the cells under different drug actions.Furthermore,it is revealed that there exists a positive linear correlation between the cytoskeleton density and the cellular mechanical properties,and the physiological state of a cell in terms of its cytoskeleton density can be predicted from its mechanical properties by a linear regression model.This study builds a relationship between the cellular mechanical properties and the cellular physiological state,adding information for evaluating drug efficacy.
基金"XingLiaoYingCai"Talents of Liaoning Province,China(Grant No.XLYC2007074)Shenyang Young and Middle-aged Science and Technology Innovation Talent Support Program(Grant No.RC200512)+1 种基金Project supported by“XingLiaoYingCai"Talents of Liaoning Province,China(Grant No.XLYC2007074)Shenyang Young and Middle-aged Science and Technology Innovation Talent Support Program(Grant No.RC200512),。
文摘Terahertz(THz)imaging has drawn significant attention because THz wave has a unique capability to transient,ultrawide spectrum and low photon energy.However,the low resolution has always been a problem due to its long wavelength,limiting their application of fields practical use.In this paper,we proposed a complex one-shot super-resolution(COSSR)framework based on a complex convolution neural network to restore superior THz images at 0.35 times wavelength by extracting features directly from a reference measured sample and groundtruth without the measured PSF.Compared with real convolution neural network-based approaches and complex zero-shot super-resolution(CZSSR),COSSR delivers at least 6.67,0.003,and 6.96%superior higher imaging efficacy in terms of peak signal to noise ratio(PSNR),mean square error(MSE),and structural similarity index measure(SSIM),respectively,for the analyzed data.Additionally,the proposed method is experimentally demonstrated to have a good generalization and to perform well on measured data.The COSSR provides a new pathway for THz imaging super-resolution(SR)reconstruction below the diffraction limit.
基金supported by the National Natural Science Foundation of China (Grant Nos.91948203,52075532).
文摘As 3D acquisition technology develops and 3D sensors become increasingly affordable,large quantities of 3D point cloud data are emerging.How to effectively learn and extract the geometric features from these point clouds has become an urgent problem to be solved.The point cloud geometric information is hidden in disordered,unstructured points,making point cloud analysis a very challenging problem.To address this problem,we propose a novel network framework,called Tree Graph Network(TGNet),which can sample,group,and aggregate local geometric features.Specifically,we construct a Tree Graph by explicit rules,which consists of curves extending in all directions in point cloud feature space,and then aggregate the features of the graph through a cross-attention mechanism.In this way,we incorporate more point cloud geometric structure information into the representation of local geometric features,which makes our network perform better.Our model performs well on several basic point clouds processing tasks such as classification,segmentation,and normal estimation,demonstrating the effectiveness and superiority of our network.Furthermore,we provide ablation experiments and visualizations to better understand our network.
基金supported by the National Natural Science Foundation of China(62203431)。
文摘The widespread usage of Cyber Physical Systems(CPSs)generates a vast volume of time series data,and precisely determining anomalies in the data is critical for practical production.Autoencoder is the mainstream method for time series anomaly detection,and the anomaly is judged by reconstruction error.However,due to the strong generalization ability of neural networks,some abnormal samples close to normal samples may be judged as normal,which fails to detect the abnormality.In addition,the dataset rarely provides sufficient anomaly labels.This research proposes an unsupervised anomaly detection approach based on adversarial memory autoencoders for multivariate time series to solve the above problem.Firstly,an encoder encodes the input data into low-dimensional space to acquire a feature vector.Then,a memory module is used to learn the feature vector’s prototype patterns and update the feature vectors.The updating process allows partial forgetting of information to prevent model overgeneralization.After that,two decoders reconstruct the input data.Finally,this research uses the Peak Over Threshold(POT)method to calculate the threshold to determine anomalous samples from normal samples.This research uses a two-stage adversarial training strategy during model training to enlarge the gap between the reconstruction error of normal and abnormal samples.The proposed method achieves significant anomaly detection results on synthetic and real datasets from power systems,water treatment plants,and computer clusters.The F1 score reached an average of 0.9196 on the five datasets,which is 0.0769 higher than the best baseline method.
基金This study is supported by The National Key Research and Development Program of China:“Key measurement and control equipment with built-in information security functions”(Grant No.2018YFB2004200)Independent Subject of State Key Laboratory of Robotics“Research on security industry network construction technology for 5G communication”(No.2022-Z13).
文摘Feature extraction plays an important role in constructing artificial intel-ligence(AI)models of industrial control systems(ICSs).Three challenges in this field are learning effective representation from high-dimensional features,data heterogeneity,and data noise due to the diversity of data dimensions,formats and noise of sensors,controllers and actuators.Hence,a novel unsupervised learn-ing autoencoder model is proposed for ICS data in this paper.Although traditional methods only capture the linear correlations of ICS features,our deep industrial representation learning model(DIRL)based on a convolutional neural network can mine high-order features,thus solving the problem of high-dimensional and heterogeneous ICS data.In addition,an unsupervised denoising autoencoder is introduced for noisy ICS data in DIRL.Training the denoising autoencoder allows the model to better mitigate the sensor noise problem.In this way,the represen-tative features learned by DIRL could help to evaluate the safety state of ICSs more effectively.We tested our model with absolute and relative accuracy experi-ments on two large-scale ICS datasets.Compared with other popular methods,DIRL showed advantages in four common indicators of AI algorithms:accuracy,precision,recall,and F1-score.This study contributes to the effective analysis of large-scale ICS data,which promotes the stable operation of ICSs.
基金financially supported by the National Key R&D Program Projects of China (No.2021YFB3202402)National Natural Science Foundation of China (No.62173321)。
文摘Rapid online analysis of liquid slag is essential for optimizing the quality and energy efficiency of steel production. To investigate the key factors that affect the online measurement of refined slag using laser-induced breakdown spectroscopy(LIBS), this study examined the effects of slag composition and temperature on the intensity and stability of the LIBS spectra. The experimental temperature was controlled at three levels: 1350℃, 1400℃, and 1450℃. The results showed that slag composition and temperature significantly affected the intensity and stability of the LIBS spectra. Increasing the Fe content and temperature in the slag reduces its viscosity, resulting in an enhanced intensity and stability of the LIBS spectra. Additionally, 42 refined slag samples were quantitatively analyzed for Fe, Si, Ca, Mg, Al, and Mn at 1350℃, 1400℃, and 1450℃.The normalized full spectrum combined with partial least squares(PLS) quantification modeling was used, using the Ca Ⅱ 317.91 nm spectral line as an internal standard. The results show that using the internal standard normalization method can significantly reduce the influence of spectral fluctuations. Meanwhile, a temperature of 1450℃ has been found to yield superior results compared to both 1350℃ and 1400℃, and it is advantageous to conduct a quantitative analysis of the slag when it is in a “water-like” state with low viscosity.
基金Science,Technology and Innovation Project of Xiongan New Area (Grant No.2022XAGG0181)LiaoNing Revitalization Talents Program (Grant No.XLYC2007074)+1 种基金Shenyang Young and Middle-aged Science and Technology Innovation Talent Support Program (Grant No.RC220523)Natural Science Foundation of Liaoning Province of China (Grant Nos.2022-YGJC-03 and 2022-MS-034)to provide fund for conducting experiments。
文摘This paper investigates a new vortex wave imaging approach to improve the imaging quality of small metal targets of size less than 1.5 mm.Antennas with different spiral phase plates are designed to efficiently transmit vortex beams with orbital angular momentums(OAMs).By analyzing the OAM spectrum of the target,it was discovered that the predominant reflection contains a particular OAM mode that carries abundant azimuthal information.This can be explained by the OAM selectivity of the target and the guidance of the vortex transmitting beam.A simple reflection vortex imaging system was designed to capture the phase information.Measurement results show that the high image contrast reaches 14.9%,which is twice as high as that of the imaging without OAM.Both of simulations and experiments demonstrate that the vortex phase imaging approach proposed in this paper can effectively improve the imaging quality at 80 GHz.This approach is suitable for other millimeter wave imaging systems and is helpful to improve the resolution in anti-terrorism security checks.
基金supported by the National Key Research and Development Program of China(2022YFF1202500,2022YFF1202502,2022YFB4703200,2023YFB4704700,2023YFB4704702)the National Natural Science Foundation of China(U22A2067,U20A20197,61773369,61903360,92048302,62203430)+1 种基金the Self-Planned Project of the State Key Laboratory of Robotics(2023-Z05)China Postdoctoral Science Foundation funded project(2022M723312)。
文摘A long history has passed since electromyography(EMG)signals have been explored in human-centered robots for intuitive interaction.However,it still has a gap between scientific research and real-life applications.Previous studies mainly focused on EMG decoding algorithms,leaving a dynamic relationship between the human,robot,and uncertain environment in real-life scenarios seldomly concerned.To fill this gap,this paper presents a comprehensive review of EMG-based techniques in human-robot-environment interaction(HREI)systems.The general processing framework is summarized,and three interaction paradigms,including direct control,sensory feedback,and partial autonomous control,are introduced.EMG-based intention decoding is treated as a module of the proposed paradigms.Five key issues involving precision,stability,user attention,compliance,and environmental awareness in this field are discussed.Several important directions,including EMG decomposition,robust algorithms,HREI dataset,proprioception feedback,reinforcement learning,and embodied intelligence,are proposed to pave the way for future research.To the best of what we know,this is the first time that a review of EMG-based methods in the HREI system is summarized.It provides a novel and broader perspective to improve the practicability of current myoelectric interaction systems,in which factors in human-robot interaction,robot-environment interaction,and state perception by human sensations are considered,which has never been done by previous studies.
基金supported in part by the National Natural Science Foundation of China(U181321461773369+2 种基金61903360)the Selfplanned Project of the State Key Laboratory of Robotics(2020-Z12)China Postdoctoral Science Foundation funded project(2019M661155)。
文摘Electromyography(EMG)has already been broadly used in human-machine interaction(HMI)applications.Determining how to decode the information inside EMG signals robustly and accurately is a key problem for which we urgently need a solution.Recently,many EMG pattern recognition tasks have been addressed using deep learning methods.In this paper,we analyze recent papers and present a literature review describing the role that deep learning plays in EMG-based HMI.An overview of typical network structures and processing schemes will be provided.Recent progress in typical tasks such as movement classification,joint angle prediction,and force/torque estimation will be introduced.New issues,including multimodal sensing,inter-subject/inter-session,and robustness toward disturbances will be discussed.We attempt to provide a comprehensive analysis of current research by discussing the advantages,challenges,and opportunities brought by deep learning.We hope that deep learning can aid in eliminating factors that hinder the development of EMG-based HMI systems.Furthermore,possible future directions will be presented to pave the way for future research.
基金supported in part by theNational Key R&D Program of China (2018YFB1304600)the Natural Science Foundation of China (51775541)CAS Interdisciplinary Innovation Team (JCTD-2018-11)。
文摘Supernumerary robotic limbs(SRLs) are a new type of wearable human auxiliary equipment, which is currently a hot research topic in the world. SRLs have broad applications in many fields, and will provide a reference and technical support for the realization of human-robot collaboration and integration,while playing an important role in improving social security and public services. In this paper, representative SRLs are summarized from the aspects of related literature analysis,research status, ontology structure design, control and driving,sensing and perception, and application fields. This paper also analyzes and summarizes the current technical challenges faced by SRLs, and reviews development progress and key technologies,thus giving a prospect of future technical development trends.
基金Projects(51875558,51471176)supported by the National Natural Science Foundation of ChinaProject(2017YFB1302802)supported by the National Key R&D Program of China。
文摘In this work,the nickel-based powder metallurgy superalloy FGH95 was selected as experimental material,and the experimental parameters in multiple overlap laser shock processing(LSP)treatment were selected based on orthogonal experimental design.The experimental data of residual stress and microhardness were measured in the same depth.The residual stress and microhardness laws were investigated and analyzed.Artificial neural network(ANN)with four layers(4-N-(N-1)-2)was applied to predict the residual stress and microhardness of FGH95 subjected to multiple overlap LSP.The experimental data were divided as training-testing sets in pairs.Laser energy,overlap rate,shocked times and depth were set as inputs,while residual stress and microhardness were set as outputs.The prediction performances with different network configuration of developed ANN models were compared and analyzed.The developed ANN model with network configuration of 4-7-6-2 showed the best predict performance.The predicted values showed a good agreement with the experimental values.In addition,the correlation coefficients among all the parameters and the effect of LSP parameters on materials response were studied.It can be concluded that ANN is a useful method to predict residual stress and microhardness of material subjected to LSP when with limited experimental data.
基金supported by National Key Research and Development Program of China(No.2016YFF0102502)the Key Research Program of Frontier Sciences,CAS(No.QYZDJ-SSW-JSC037)the Youth Innovation Promotion Association,CAS,Liao Ning Revitalization Talents Program(No.XLYC1807110)。
文摘In the spectral analysis of laser-induced breakdown spectroscopy,abundant characteristic spectral lines and severe interference information exist simultaneously in the original spectral data.Here,a feature selection method called recursive feature elimination based on ridge regression(Ridge-RFE)for the original spectral data is recommended to make full use of the valid information of spectra.In the Ridge-RFE method,the absolute value of the ridge regression coefficient was used as a criterion to screen spectral characteristic,the feature with the absolute value of minimum weight in the input subset features was removed by recursive feature elimination(RFE),and the selected features were used as inputs of the partial least squares regression(PLS)model.The Ridge-RFE method based PLS model was used to measure the Fe,Si,Mg,Cu,Zn and Mn for 51 aluminum alloy samples,and the results showed that the root mean square error of prediction decreased greatly compared to the PLS model with full spectrum as input.The overall results demonstrate that the Ridge-RFE method is more efficient to extract the redundant features,make PLS model for better quantitative analysis results and improve model generalization ability.
基金financially supported by the National Natural Science Foundation of China(Grant No.41806122)the Strategic Priority Research Program of the Chinese Academy of Sciences(Grant No.XDA11040102)+2 种基金the State Key Laboratory of Robotics of China(Grant No.2017-Z08)Youth Innovation Promotion Association,CASJiang Xinsong Innovation Fund
文摘Range is an important factor to the design of autonomous underwater vehicles (AUVs), while drag reduction efforts are pursued, the investigation of body-propeller interaction is another vital consideration. We present a numerical and experimental study of the hull-propeller interaction for deeply submerged underwater vehicles, using a proportional-integral- derivative (PID) controller method to estimate self-propulsion point in CFD environment. The hydrodynamic performance of hull and propeller at the balance state when the AUV sails at a fixed depth is investigated, using steady RANS solver of Star-CCM+. The proposed steady RANS solver takes only hours to reach a reasonable solution. It is more time efficient than unsteady simulations which takes days or weeks, as well as huge consumption of computing resources. Explorer 1000, a long range AUV developed by Shenyang Institute of Automation, Chinese Academy of Sciences, was studied as an object, and self-propulsion point, thrust deduction, wake fraction and hull efficiency were analyzed by using the proposed RANS method. Behind-hull performance of the selected propeller MAU4-40, as well as the hull-propeller interaction, was obtained from the computed hydrodynamic forces. The numerical results are in good qualitative and quantitative agreement with the experimental results obtained in the Qiandao Lake of Zhejiang province, China.
