Deep Learning has been widely used to model soft sensors in modern industrial processes with nonlinear variables and uncertainty.Due to the outstanding ability for high-level feature extraction,stacked autoencoder(SAE...Deep Learning has been widely used to model soft sensors in modern industrial processes with nonlinear variables and uncertainty.Due to the outstanding ability for high-level feature extraction,stacked autoencoder(SAE)has been widely used to improve the model accuracy of soft sensors.However,with the increase of network layers,SAE may encounter serious information loss issues,which affect the modeling performance of soft sensors.Besides,there are typically very few labeled samples in the data set,which brings challenges to traditional neural networks to solve.In this paper,a multi-scale feature fused stacked autoencoder(MFF-SAE)is suggested for feature representation related to hierarchical output,where stacked autoencoder,mutual information(MI)and multi-scale feature fusion(MFF)strategies are integrated.Based on correlation analysis between output and input variables,critical hidden variables are extracted from the original variables in each autoencoder's input layer,which are correspondingly given varying weights.Besides,an integration strategy based on multi-scale feature fusion is adopted to mitigate the impact of information loss with the deepening of the network layers.Then,the MFF-SAE method is designed and stacked to form deep networks.Two practical industrial processes are utilized to evaluate the performance of MFF-SAE.Results from simulations indicate that in comparison to other cutting-edge techniques,the proposed method may considerably enhance the accuracy of soft sensor modeling,where the suggested method reduces the root mean square error(RMSE)by 71.8%,17.1%and 64.7%,15.1%,respectively.展开更多
Coking at the fractionating tower bottom and the decant oil circulation system disrupts the heat balance,leading to unplanned shutdown and destroying the long period stable operation of the Fluid Catalytic Cracking Un...Coking at the fractionating tower bottom and the decant oil circulation system disrupts the heat balance,leading to unplanned shutdown and destroying the long period stable operation of the Fluid Catalytic Cracking Unit(FCCU).The FCCU operates through interconnected subsystems,generating high-dimensional,nonlinear,and non-stationary data characterized by spatiotemporally correlated.The decant oil solid content is the crucial indicator for monitoring catalyst loss from the reactor-regenerator system and coking risk tendency at the fractionating tower bottom that relies on sampling and laboratory testing,which is lagging responsiveness and labor-intensive.Developing the online decant oil solid content soft sensor using industrial data to support operators in conducting predictive maintenance is essential.Therefore,this paper proposes a hybrid deep learning framework for soft sensor development that combines spatiotemporal pattern extraction with interpretability,enabling accurate risk identification in dynamic operational conditions.This framework employs a Filter-Wrapper method for dimensionality reduction,followed by a 2D Convolutional Neural Network(2DCNN)for extracting spatial patterns,and a Bidirectional Gated Recurrent Unit(BiGRU)for capturing long-term temporal dependencies,with an Attention Mechanism(AM)to highlight critical features adaptively.The integration of SHapley Additive exPlanations(SHAP),Complementary Ensemble Empirical Mode Decomposition with Adaptive Noise(CEEMDAN),2DCNN,and expert knowledge precisely quantifies feature contributions and decomposes signals,significantly enhancing the practicality of risk identification.Applied to a China refinery with processing capacity of 2.80×10^(6) t/a,the soft sensor achieved the R^(2) value of 0.93 and five-level risk identification accuracy of 96.42%.These results demonstrate the framework's accuracy,robustness,and suitability for complex industrial scenarios,advancing risk visualization and management.展开更多
Adaptive graph neural networks(AGNNs)have achieved remarkable success in industrial process soft sensing by incorporating explicit features that delineate the relationships between process variables.This article intro...Adaptive graph neural networks(AGNNs)have achieved remarkable success in industrial process soft sensing by incorporating explicit features that delineate the relationships between process variables.This article introduces a novel GNN framework,termed entropy-regularized ensemble adaptive graph(E^(2)AG),aimed at enhancing the predictive accuracy of AGNNs.Specifically,this work pioneers a novel AGNN learning approach based on mirror descent,which is central to ensuring the efficiency of the training procedure and consequently guarantees that the learned graph naturally adheres to the row-normalization requirement intrinsic to the message-passing of GNNs.Subsequently,motivated by multi-head self-attention mechanism,the training of ensembled AGNNs is rigorously examined within this framework,incorporating an entropy regularization term in the learning objective to ensure the diversity of the learned graph.After that,the architecture and training algorithm of the model are then concisely summarized.Finally,to ascertain the efficacy of the proposed E^(2)AG model,extensive experiments are conducted on real-world industrial datasets.The evaluation focuses on prediction accuracy,model efficacy,and sensitivity analysis,demonstrating the superiority of E^(2)AG in industrial soft sensing applications.展开更多
To address the stochasticity and nonlinearity of solar collector power systems,a soft sensor prediction model with a hybrid convolutional neural network(CNN)and long short-term memory network(LSTM)was constructed,and ...To address the stochasticity and nonlinearity of solar collector power systems,a soft sensor prediction model with a hybrid convolutional neural network(CNN)and long short-term memory network(LSTM)was constructed,and the hyperparameter optimization of the hybrid neural network(CNN-LSTM)was carried out by using the sparrow search algorithm(SSA).The model utilized the powerful feature extraction and non-linear mapping capabilities of deep learning to effectively handle the complex relationship between input and target variables.The batch normalization technique was used to speed up the training and improve the stability of the soft-sensing model,and the random discard technique was used to prevent the soft-sensing model from overfitting.Finally,the mean absolute error(MAE)was used to assess the accuracy of the soft sensor model predictions.This study compared the proposed model with soft sensor prediction models like Bp,Elman,CNN,LSTM,and CNN-LSTM,using dynamic thermal performance data from the solar collector field of the molten salt linear Fresnel photovoltaic demonstration power plant.The deep learning-based soft sensor model outperformed the other models according to the experimental data.Its coefficients of determination(namely R^(2))are higher by 6.35%,8.42%,5.69%,6.90%,and 3.67%,respectively.The accuracy and robustness have been significantly improved.展开更多
Prosthetic devices designed to assist individuals with damaged or missing body parts have made significant strides,particularly with advancements in machine intelligence and bioengineering.Initially focused on movemen...Prosthetic devices designed to assist individuals with damaged or missing body parts have made significant strides,particularly with advancements in machine intelligence and bioengineering.Initially focused on movement assistance,the field has shifted towards developing prosthetics that function as seamless extensions of the human body.During this progress,a key challenge remains the reduction of interface artifacts between prosthetic components and biological tissues.Soft electronics offer a promising solution due to their structural flexibility and enhanced tissue adaptability.However,achieving full integration of prosthetics with the human body requires both artificial perception and efficient transmission of physical signals.In this context,synaptic devices have garnered attention as next-generation neuromorphic computing elements because of their low power consumption,ability to enable hardware-based learning,and high compatibility with sensing units.These devices have the potential to create artificial pathways for sensory recognition and motor responses,forming a“sensory-neuromorphic system”that emulates synaptic junctions in biological neurons,thereby connecting with impaired biological tissues.Here,we discuss recent developments in prosthetic components and neuromorphic applications with a focus on sensory perception and sensorimotor actuation.Initially,we explore a prosthetic system with advanced sensory units,mechanical softness,and artificial intelligence,followed by the hardware implementation of memory devices that combine calculation and learning functions.We then highlight the importance and mechanisms of soft-form synaptic devices that are compatible with sensing units.Furthermore,we review an artificial sensory-neuromorphic perception system that replicates various biological senses and facilitates sensorimotor loops from sensory receptors,the spinal cord,and motor neurons.Finally,we propose insights into the future of closed-loop neuroprosthetics through the technical integration of soft electronics,including bio-integrated sensors and synaptic devices,into prosthetic systems.展开更多
Monitoring biogenic amines,which are metabolic byproducts of shrimp spoilage,is crucial for assessing food quality.Currently,most detection methods for biogenic amines suffer from limitations such as time-consuming pr...Monitoring biogenic amines,which are metabolic byproducts of shrimp spoilage,is crucial for assessing food quality.Currently,most detection methods for biogenic amines suffer from limitations such as time-consuming procedures,complex operations,and delayed results.Colorimetric analysis techniques have gained attention in recent years due to their advantages of short analysis time,simple operation,and suitability for on-site testing.This study successfully developed a series of colorimetric sensor platforms for biogenic amines by loading the natural active ingredient curcumin(CUR)and its derivative of Boron complex BFCUR onto filter paper and electrospun nanofibre films(ENFs),respectively.By analyzing the color response differences of these sensors upon contact with biogenic amines,the colorimetric sensors with superior detection performance were selected and further applied to the visual monitoring and indication of shrimp spoilage processes.展开更多
Flexible fiber sensors,However,traditional methods face challenges in fabricating low-cost,large-scale fiber sensors.In recent years,the thermal drawing process has rapidly advanced,offering a novel approach to flexib...Flexible fiber sensors,However,traditional methods face challenges in fabricating low-cost,large-scale fiber sensors.In recent years,the thermal drawing process has rapidly advanced,offering a novel approach to flexible fiber sensors.Through the preform-tofiber manufacturing technique,a variety of fiber sensors with complex functionalities spanning from the nanoscale to kilometer scale can be automated in a short time.Examples include temperature,acoustic,mechanical,chemical,biological,optoelectronic,and multifunctional sensors,which operate on diverse sensing principles such as resistance,capacitance,piezoelectricity,triboelectricity,photoelectricity,and thermoelectricity.This review outlines the principles of the thermal drawing process and provides a detailed overview of the latest advancements in various thermally drawn fiber sensors.Finally,the future developments of thermally drawn fiber sensors are discussed.展开更多
Developing effective,versatile,and high-precision sensing interfaces remains a crucial challenge in human-machine-environment interaction applications.Despite progress in interaction-oriented sensing skins,limitations...Developing effective,versatile,and high-precision sensing interfaces remains a crucial challenge in human-machine-environment interaction applications.Despite progress in interaction-oriented sensing skins,limitations remain in unit-level reconfiguration,multiaxial force and motion sensing,and robust operation across dynamically changing or irregular surfaces.Herein,we develop a reconfigurable omnidirectional triboelectric whisker sensor array(RO-TWSA)comprising multiple sensing units that integrate a triboelectric whisker structure(TWS)with an untethered hydro-sealing vacuum sucker(UHSVS),enabling reversibly portable deployment and omnidirectional perception across diverse surfaces.Using a simple dual-triangular electrode layout paired with MXene/silicone nanocomposite dielectric layer,the sensor unit achieves precise omnidirectional force and motion sensing with a detection threshold as low as 0.024 N and an angular resolution of 5°,while the UHSVS provides reliable and reversible multi-surface anchoring for the sensor units by involving a newly designed hydrogel combining high mechanical robustness and superior water absorption.Extensive experiments demonstrate the effectiveness of RO-TWSA across various interactive scenarios,including teleoperation,tactile diagnostics,and robotic autonomous exploration.Overall,RO-TWSA presents a versatile and high-resolution tactile interface,offering new avenues for intelligent perception and interaction in complex real-world environments.展开更多
Wearable sensors integrated with deep learning techniques have the potential to revolutionize seamless human-machine interfaces for real-time health monitoring,clinical diagnosis,and robotic applications.Nevertheless,...Wearable sensors integrated with deep learning techniques have the potential to revolutionize seamless human-machine interfaces for real-time health monitoring,clinical diagnosis,and robotic applications.Nevertheless,it remains a critical challenge to simultaneously achieve desirable mechanical and electrical performance along with biocompatibility,adhesion,self-healing,and environmental robustness with excellent sensing metrics.Herein,we report a multifunctional,anti-freezing,selfadhesive,and self-healable organogel pressure sensor composed of cobalt nanoparticle encapsulated nitrogen-doped carbon nanotubes(CoN CNT)embedded in a polyvinyl alcohol-gelatin(PVA/GLE)matrix.Fabricated using a binary solvent system of water and ethylene glycol(EG),the CoN CNT/PVA/GLE organogel exhibits excellent flexibility,biocompatibility,and temperature tolerance with remarkable environmental stability.Electrochemical impedance spectroscopy confirms near-stable performance across a broad humidity range(40%-95%RH).Freeze-tolerant conductivity under sub-zero conditions(-20℃)is attributed to the synergistic role of CoN CNT and EG,preserving mobility and network integrity.The Co N CNT/PVA/GLE organogel sensor exhibits high sensitivity of 5.75 k Pa^(-1)in the detection range from 0 to 20 k Pa,ideal for subtle biomechanical motion detection.A smart human-machine interface for English letter recognition using deep learning achieved 98%accuracy.The organogel sensor utility was extended to detect human gestures like finger bending,wrist motion,and throat vibration during speech.展开更多
As emerging two-dimensional(2D)materials,carbides and nitrides(MXenes)could be solid solutions or organized structures made up of multi-atomic layers.With remarkable and adjustable electrical,optical,mechanical,and el...As emerging two-dimensional(2D)materials,carbides and nitrides(MXenes)could be solid solutions or organized structures made up of multi-atomic layers.With remarkable and adjustable electrical,optical,mechanical,and electrochemical characteristics,MXenes have shown great potential in brain-inspired neuromorphic computing electronics,including neuromorphic gas sensors,pressure sensors and photodetectors.This paper provides a forward-looking review of the research progress regarding MXenes in the neuromorphic sensing domain and discussed the critical challenges that need to be resolved.Key bottlenecks such as insufficient long-term stability under environmental exposure,high costs,scalability limitations in large-scale production,and mechanical mismatch in wearable integration hinder their practical deployment.Furthermore,unresolved issues like interfacial compatibility in heterostructures and energy inefficiency in neu-romorphic signal conversion demand urgent attention.The review offers insights into future research directions enhance the fundamental understanding of MXene properties and promote further integration into neuromorphic computing applications through the convergence with various emerging technologies.展开更多
Conductive elastomers combining micromechanical sensitivity,lightweight adaptability,and environmental sustainability are critically needed for advanced flexible electronics requiring precise responsiveness and long-t...Conductive elastomers combining micromechanical sensitivity,lightweight adaptability,and environmental sustainability are critically needed for advanced flexible electronics requiring precise responsiveness and long-term wearability;however,the integration of these properties remains a significant challenge.Here,we present a biomass-derived conductive elastomer featuring a rationally engineered dynamic crosslinked network integrated with a tunable microporous architecture.This structural design imparts pronounced micromechanical sensitivity,an ultralow density(~0.25 g cm^(−3)),and superior mechanical compliance for adaptive deformation.Moreover,the unique micro-spring effect derived from the porous architecture ensures exceptional stretchability(>500%elongation at break)and superior resilience,delivering immediate and stable electrical response under both subtle(<1%)and large(>200%)mechanical stimuli.Intrinsic dynamic interactions endow the elastomer with efficient room temperature self-healing and complete recyclability without compromising performance.First-principles simulations clarify the mechanisms behind micropore formation and the resulting functionality.Beyond its facile and mild fabrication process,this work establishes a scalable route toward high-performance,sustainable conductive elastomers tailored for next-generation soft electronics.展开更多
Fiber-structured ion sensors have gained traction in health monitoring and medical diagnostics owing to their structural flexibility,enhanced sensitivity,and suitability for integration into wearable devices.This stud...Fiber-structured ion sensors have gained traction in health monitoring and medical diagnostics owing to their structural flexibility,enhanced sensitivity,and suitability for integration into wearable devices.This study employed a simple and efficient solutionbased process to fabricate nanofibers containing aggregation-induced emission(AIE)dyes.The resulting AIE nanofibers exhibited stable and intense fluorescence,nanosecond fluorescence lifetime,and low-loss light transport when functioning as active waveguides.Additionally,crossed nanofiber intersections exhibited diffraction-limited emission spots.The AIE nanofibers demonstrate efficient and ionspecific fluorescence quenching in response to Ag^(+).These results support the development of sensing units capable of operating in liquid environments or in direct contact with skin or tissues,facilitating real-time monitoring of ion concentrations for personalized healthcare management.展开更多
The advancement of wearable sensing technologies demands multifunctional materials that integrate high sensitivity,environmental resilience,and intelligent signal processing.In this work,a flexible hydrophobic conduct...The advancement of wearable sensing technologies demands multifunctional materials that integrate high sensitivity,environmental resilience,and intelligent signal processing.In this work,a flexible hydrophobic conductive yarn(FCB@SY)featuring a controllable microcrack structure is developed via a synergistic approach combining ultrasonic swelling and non-solvent induced phase separation(NIPS).By embedding a robust conductive network and engineering microcrack morphology,the resulting sensor achieves an ultrahigh gauge factor(GF≈12,670),an ultrabroad working range(0%-547%),a low detection limit(0.5%),rapid response/recovery time(140 ms/140 ms),and outstanding durability over 10,000 cycles.Furthermore,the hydrophobic surface endowed by conductive coatings imparts exceptional chemical stability against acidic and alkaline environments,as well as reliable waterproof performance.This enables consistent functionality under harsh conditions,including underwater operation.Integrated with machine learning algorithms,the FCB@SY-based intelligent sensing system demonstrates dualmode capabilities in human motion tracking and gesture recognition,offering significant potential for applications in wearable electronics,human-machine interfaces,and soft robotics.展开更多
In the context of the rapid development of artificial intelligence and robotics,their application scenarios are continuously expanding to a variety of complex environments,with increasing attention being paid to the u...In the context of the rapid development of artificial intelligence and robotics,their application scenarios are continuously expanding to a variety of complex environments,with increasing attention being paid to the use of flexible sensors in lowtemperature environments.In this study,an ionic hydrogel was synthesized using acrylamide(AM),hydroxyethyl cellulose(HEC),and lithium chloride(LiCl)as composites.This hydrogel exhibits high adhesion,excellent sensitivity(gauge factor(GF)=2.84),rapid response time(100 ms),exceptional stretch ability(>1776%),high toughness(2.5 MJ/m^(3)),and the ability to maintain detectability at low temperatures(-60℃).HEC imparts reliable mechanical properties to the sensor through hydrogen bonding interactions of its hydroxyl groups.LiCl ensures that the sensor has outstanding antifreezing properties,maintains good conductivity and mechanical performance.Used for robotic attitude detection,the sensor demonstrated accurate recognition of various joint movements at both 20 and -20℃.This technology was extended to industrial operations and maintenance,where a mechanical claw was used to grasp parts at both room temperature and low temperature.A convolutional neural network deep learning algorithm was employed to identify and classify eight types of parts,achieving an impressive recognition accuracy of 98.8%.The polyacrylamide(PAM)/HEC/LiCl hydrogel sensor demonstrates the capability for wide-temperature range detection in flexible robotics,holding significant potential for future applications in human-machine interaction,tactile perception,and related fields.展开更多
Diabetes mellitus represents a major global health issue,driving the need for noninvasive alternatives to traditional blood glucose monitoring methods.Recent advancements in wearable technology have introduced skin-in...Diabetes mellitus represents a major global health issue,driving the need for noninvasive alternatives to traditional blood glucose monitoring methods.Recent advancements in wearable technology have introduced skin-interfaced biosensors capable of analyzing sweat and skin biomarkers,providing innovative solutions for diabetes diagnosis and monitoring.This review comprehensively discusses the current developments in noninvasive wearable biosensors,emphasizing simultaneous detection of biochemical biomarkers(such as glucose,cortisol,lactate,branched-chain amino acids,and cytokines)and physiological signals(including heart rate,blood pressure,and sweat rate)for accurate,personalized diabetes management.We explore innovations in multimodal sensor design,materials science,biorecognition elements,and integration techniques,highlighting the importance of advanced data analytics,artificial intelligence-driven predictive algorithms,and closed-loop therapeutic systems.Additionally,the review addresses ongoing challenges in biomarker validation,sensor stability,user compliance,data privacy,and regulatory considerations.A holistic,multimodal approach enabled by these next-generation wearable biosensors holds significant potential for improving patient outcomes and facilitating proactive healthcare interventions in diabetes management.展开更多
Human action recognition(HAR)is crucial for the development of efficient computer vision,where bioinspired neuromorphic perception visual systems have emerged as a vital solution to address transmission bottlenecks ac...Human action recognition(HAR)is crucial for the development of efficient computer vision,where bioinspired neuromorphic perception visual systems have emerged as a vital solution to address transmission bottlenecks across sensor-processor interfaces.However,the absence of interactions among versatile biomimicking functionalities within a single device,which was developed for specific vision tasks,restricts the computational capacity,practicality,and scalability of in-sensor vision computing.Here,we propose a bioinspired vision sensor composed of a Ga N/Al N-based ultrathin quantum-disks-in-nanowires(QD-NWs)array to mimic not only Parvo cells for high-contrast vision and Magno cells for dynamic vision in the human retina but also the synergistic activity between the two cells for in-sensor vision computing.By simply tuning the applied bias voltage on each QD-NW-array-based pixel,we achieve two biosimilar photoresponse characteristics with slow and fast reactions to light stimuli that enhance the in-sensor image quality and HAR efficiency,respectively.Strikingly,the interplay and synergistic interaction of the two photoresponse modes within a single device markedly increased the HAR recognition accuracy from 51.4%to 81.4%owing to the integrated artificial vision system.The demonstration of an intelligent vision sensor offers a promising device platform for the development of highly efficient HAR systems and future smart optoelectronics.展开更多
The growing prevalence of exercise-induced tibial stress fractures demands wearable sensors capable of monitoring dynamic musculoskeletal loads with medical-grade precision.While flexible pressure-sensing insoles show...The growing prevalence of exercise-induced tibial stress fractures demands wearable sensors capable of monitoring dynamic musculoskeletal loads with medical-grade precision.While flexible pressure-sensing insoles show clinical potential,their development has been hindered by the intrinsic trade-off between high sensitivity and full-range linearity(R^(2)>0.99 up to 1 MPa)in conventional designs.Inspired by the tactile sensing mechanism of human skin,where dermal stratification enables wide-range pressure adaptation and ion-channelregulated signaling maintains linear electrical responses,we developed a dual-mechanism flexible iontronic pressure sensor(FIPS).This innovative design synergistically combines two bioinspired components:interdigitated fabric microstructures enabling pressure-proportional contact area expansion(αP1/3)and iontronic film facilitating self-adaptive ion concentration modulation(αP^(2/3)),which together generate a linear capacitance-pressure response(CαP).The FIPS achieves breakthrough performance:242 kPa^(-1)sensitivity with 0.997linearity across 0-1 MPa,yielding a record linear sensing factor(LSF=242,000).The design is validated across various substrates and ionic materials,demonstrating its versatility.Finally,the FIPS-driven design enables a smart insole demonstrating 1.8%error in tibial load assessment during gait analysis,outperforming nonlinear counterparts(6.5%error)in early fracture-risk prediction.The biomimetic design framework establishes a universal approach for developing high-performance linear sensors,establishing generalized principles for medical-grade wearable devices.展开更多
Soft X-ray detectors play a vital role in materials science,high-energy physics and medical imaging.Cs_(2)AgBiBr_(6),a lead-free double perovskite,has gained attention for its excellent optoelectronic properties,stabi...Soft X-ray detectors play a vital role in materials science,high-energy physics and medical imaging.Cs_(2)AgBiBr_(6),a lead-free double perovskite,has gained attention for its excellent optoelectronic properties,stability,and nontoxicity.However,its fast crystallization and requirement for high-temperature annealing(>250℃)often lead to inferior film quality,limiting its application in flexible devices.This study introduces an alloying strategy that significantly improves the quality of Cs_(2)AgBiBr_(6)thin films annealed at a reduced temperature of 150℃.Devices based on the alloyed thin films exhibit an ultra-low dark current of 0.32 nA·cm^(-2)and a quantum efficiency of 725%.Furthermore,the first successful integration of Cs_(2)AgBiBr_(6)with a thinfilm transistor backplane demonstrates its superior imaging performance,indicating that Cs_(2)AgBiBr_(6)is a promising material for next-generation soft X-ray sensors.展开更多
In order to solve the problem of the invalidation of thermal parameters andoptimal running, we present an efficient soft sensor approach based on sparse online Gaussianprocesses( GP), which is based on a combination o...In order to solve the problem of the invalidation of thermal parameters andoptimal running, we present an efficient soft sensor approach based on sparse online Gaussianprocesses( GP), which is based on a combination of a Bayesian online algorithm together with asequential construction of a relevant subsample of the data to specify the prediction of the GPmodel. By an appealing parameterization and projection techniques that use the reproducing kernelHubert space (RKHS) norm, recursions for the effective parameters and a sparse Gaussianapproximation of the posterior process are obtained. The sparse representation of Gaussian processesmakes the GP-based soft sensor practical in a large dataset and real-time application. And theproposed thermalparameter soft sensor is of importance for the economical running of the powerplant.展开更多
近年来,综合能源系统作为一种以多种能源形态和设备相互交互的能源系统方案得到了广泛应用和研究.然而,在面对动态复杂的多能源系统时,传统的优化调度方法往往无法满足其实时性和精准度需求.因此,本文设计了一种软深度确定性策略梯度(So...近年来,综合能源系统作为一种以多种能源形态和设备相互交互的能源系统方案得到了广泛应用和研究.然而,在面对动态复杂的多能源系统时,传统的优化调度方法往往无法满足其实时性和精准度需求.因此,本文设计了一种软深度确定性策略梯度(Soft Deep Deterministic Policy Gradient,Soft-DDPG)算法驱动的综合能源系统优化调度方法,以最小化调度周期内系统总运行成本为目标,建立设备运行综合能效评估模型,再采用Soft-DDPG算法对每个能源设备的能效调度动作进行优化控制.Soft-DDPG算法将softmax算子引入到动作值函数的计算中,有效降低了Q值高估问题.与此同时,该算法在动作选择策略中加入了随机噪声,提高了算法的学习效率.实验结果显示,本文所提出的方法解决了综合能源系统能效调度实时性差、精准度低的瓶颈问题,实现了系统的高效灵活调度,降低了系统的总运行成本.展开更多
基金supported by the National Key Research and Development Program of China(2023YFB3307800)National Natural Science Foundation of China(62394343,62373155)+2 种基金Major Science and Technology Project of Xinjiang(No.2022A01006-4)State Key Laboratory of Industrial Control Technology,China(Grant No.ICT2024A26)Fundamental Research Funds for the Central Universities.
文摘Deep Learning has been widely used to model soft sensors in modern industrial processes with nonlinear variables and uncertainty.Due to the outstanding ability for high-level feature extraction,stacked autoencoder(SAE)has been widely used to improve the model accuracy of soft sensors.However,with the increase of network layers,SAE may encounter serious information loss issues,which affect the modeling performance of soft sensors.Besides,there are typically very few labeled samples in the data set,which brings challenges to traditional neural networks to solve.In this paper,a multi-scale feature fused stacked autoencoder(MFF-SAE)is suggested for feature representation related to hierarchical output,where stacked autoencoder,mutual information(MI)and multi-scale feature fusion(MFF)strategies are integrated.Based on correlation analysis between output and input variables,critical hidden variables are extracted from the original variables in each autoencoder's input layer,which are correspondingly given varying weights.Besides,an integration strategy based on multi-scale feature fusion is adopted to mitigate the impact of information loss with the deepening of the network layers.Then,the MFF-SAE method is designed and stacked to form deep networks.Two practical industrial processes are utilized to evaluate the performance of MFF-SAE.Results from simulations indicate that in comparison to other cutting-edge techniques,the proposed method may considerably enhance the accuracy of soft sensor modeling,where the suggested method reduces the root mean square error(RMSE)by 71.8%,17.1%and 64.7%,15.1%,respectively.
基金supported by the Innovative Research Group Project of the National Natural Science Foundation of China(22021004)Sinopec Major Science and Technology Projects(321123-1)。
文摘Coking at the fractionating tower bottom and the decant oil circulation system disrupts the heat balance,leading to unplanned shutdown and destroying the long period stable operation of the Fluid Catalytic Cracking Unit(FCCU).The FCCU operates through interconnected subsystems,generating high-dimensional,nonlinear,and non-stationary data characterized by spatiotemporally correlated.The decant oil solid content is the crucial indicator for monitoring catalyst loss from the reactor-regenerator system and coking risk tendency at the fractionating tower bottom that relies on sampling and laboratory testing,which is lagging responsiveness and labor-intensive.Developing the online decant oil solid content soft sensor using industrial data to support operators in conducting predictive maintenance is essential.Therefore,this paper proposes a hybrid deep learning framework for soft sensor development that combines spatiotemporal pattern extraction with interpretability,enabling accurate risk identification in dynamic operational conditions.This framework employs a Filter-Wrapper method for dimensionality reduction,followed by a 2D Convolutional Neural Network(2DCNN)for extracting spatial patterns,and a Bidirectional Gated Recurrent Unit(BiGRU)for capturing long-term temporal dependencies,with an Attention Mechanism(AM)to highlight critical features adaptively.The integration of SHapley Additive exPlanations(SHAP),Complementary Ensemble Empirical Mode Decomposition with Adaptive Noise(CEEMDAN),2DCNN,and expert knowledge precisely quantifies feature contributions and decomposes signals,significantly enhancing the practicality of risk identification.Applied to a China refinery with processing capacity of 2.80×10^(6) t/a,the soft sensor achieved the R^(2) value of 0.93 and five-level risk identification accuracy of 96.42%.These results demonstrate the framework's accuracy,robustness,and suitability for complex industrial scenarios,advancing risk visualization and management.
基金supported in part by the National Natural Science Foundation of China(NSFC)(62473103,62203169,62473121)the Postdoctoral Science Foundation of Zhejiang Province(ZJ2023011).
文摘Adaptive graph neural networks(AGNNs)have achieved remarkable success in industrial process soft sensing by incorporating explicit features that delineate the relationships between process variables.This article introduces a novel GNN framework,termed entropy-regularized ensemble adaptive graph(E^(2)AG),aimed at enhancing the predictive accuracy of AGNNs.Specifically,this work pioneers a novel AGNN learning approach based on mirror descent,which is central to ensuring the efficiency of the training procedure and consequently guarantees that the learned graph naturally adheres to the row-normalization requirement intrinsic to the message-passing of GNNs.Subsequently,motivated by multi-head self-attention mechanism,the training of ensembled AGNNs is rigorously examined within this framework,incorporating an entropy regularization term in the learning objective to ensure the diversity of the learned graph.After that,the architecture and training algorithm of the model are then concisely summarized.Finally,to ascertain the efficacy of the proposed E^(2)AG model,extensive experiments are conducted on real-world industrial datasets.The evaluation focuses on prediction accuracy,model efficacy,and sensitivity analysis,demonstrating the superiority of E^(2)AG in industrial soft sensing applications.
基金supported by National Natural Science Foundation of China(No.52266012)Gansu Province College Industry Support Plan Project(No.2022CYZC-34)+1 种基金Gansu Province Major Science and Technology Special Project(Nos.20ZD7GF011,22ZD6GA063)Jiuquan City Science and Technology Programme Project(No.2023CA3058).
文摘To address the stochasticity and nonlinearity of solar collector power systems,a soft sensor prediction model with a hybrid convolutional neural network(CNN)and long short-term memory network(LSTM)was constructed,and the hyperparameter optimization of the hybrid neural network(CNN-LSTM)was carried out by using the sparrow search algorithm(SSA).The model utilized the powerful feature extraction and non-linear mapping capabilities of deep learning to effectively handle the complex relationship between input and target variables.The batch normalization technique was used to speed up the training and improve the stability of the soft-sensing model,and the random discard technique was used to prevent the soft-sensing model from overfitting.Finally,the mean absolute error(MAE)was used to assess the accuracy of the soft sensor model predictions.This study compared the proposed model with soft sensor prediction models like Bp,Elman,CNN,LSTM,and CNN-LSTM,using dynamic thermal performance data from the solar collector field of the molten salt linear Fresnel photovoltaic demonstration power plant.The deep learning-based soft sensor model outperformed the other models according to the experimental data.Its coefficients of determination(namely R^(2))are higher by 6.35%,8.42%,5.69%,6.90%,and 3.67%,respectively.The accuracy and robustness have been significantly improved.
基金supported by the National Research Foundation of Korea(NRF)grant funded by the Korean government(MSIT)(No.2020R1C1C1005567)supported by the NAVER Digital Bio Innovation Research Fund,funded by NAVER Corporation(Grant No.[37-2023-0040])+3 种基金supported by Institute of Information&communications Technology Planning&Evaluation(IITP)grant funded by the Korea government(MSIT)(No.2020-0-00261,Development of low power/low delay/self-power suppliable RF simultaneous information and power transfer system and stretchable electronic epineurium for wireless nerve bypass implementation)supported by Institute for Basic Science(IBS-R015-D1,IBSR015-D2)supported by a grant of the Korea-US Collaborative Research Fund(KUCRF)funded by the Ministry of Science and ICT and Ministry of Health&Welfare,Republic of Korea(Grant Number.RS-2024-00467213)。
文摘Prosthetic devices designed to assist individuals with damaged or missing body parts have made significant strides,particularly with advancements in machine intelligence and bioengineering.Initially focused on movement assistance,the field has shifted towards developing prosthetics that function as seamless extensions of the human body.During this progress,a key challenge remains the reduction of interface artifacts between prosthetic components and biological tissues.Soft electronics offer a promising solution due to their structural flexibility and enhanced tissue adaptability.However,achieving full integration of prosthetics with the human body requires both artificial perception and efficient transmission of physical signals.In this context,synaptic devices have garnered attention as next-generation neuromorphic computing elements because of their low power consumption,ability to enable hardware-based learning,and high compatibility with sensing units.These devices have the potential to create artificial pathways for sensory recognition and motor responses,forming a“sensory-neuromorphic system”that emulates synaptic junctions in biological neurons,thereby connecting with impaired biological tissues.Here,we discuss recent developments in prosthetic components and neuromorphic applications with a focus on sensory perception and sensorimotor actuation.Initially,we explore a prosthetic system with advanced sensory units,mechanical softness,and artificial intelligence,followed by the hardware implementation of memory devices that combine calculation and learning functions.We then highlight the importance and mechanisms of soft-form synaptic devices that are compatible with sensing units.Furthermore,we review an artificial sensory-neuromorphic perception system that replicates various biological senses and facilitates sensorimotor loops from sensory receptors,the spinal cord,and motor neurons.Finally,we propose insights into the future of closed-loop neuroprosthetics through the technical integration of soft electronics,including bio-integrated sensors and synaptic devices,into prosthetic systems.
基金Supported by the Guangdong-Hong Kong-Macao Joint Laboratory on Micro-Nano Manufacturing Technology,China(No.2021LSYS004)Guangdong Provincial Key Laboratory of Sustainable Biomimetic Materials and Green Energy,China(No.2024B1212010003)。
文摘Monitoring biogenic amines,which are metabolic byproducts of shrimp spoilage,is crucial for assessing food quality.Currently,most detection methods for biogenic amines suffer from limitations such as time-consuming procedures,complex operations,and delayed results.Colorimetric analysis techniques have gained attention in recent years due to their advantages of short analysis time,simple operation,and suitability for on-site testing.This study successfully developed a series of colorimetric sensor platforms for biogenic amines by loading the natural active ingredient curcumin(CUR)and its derivative of Boron complex BFCUR onto filter paper and electrospun nanofibre films(ENFs),respectively.By analyzing the color response differences of these sensors upon contact with biogenic amines,the colorimetric sensors with superior detection performance were selected and further applied to the visual monitoring and indication of shrimp spoilage processes.
基金supported by the National Key Research and Development Program of China(2023YFB3809800)the National Natural Science Foundation of China(52172249,52525601)+2 种基金the Chinese Academy of Sciences Talents Program(E2290701)the Jiangsu Province Talents Program(JSSCRC2023545)the Special Fund Project of Carbon Peaking Carbon Neutrality Science and Technology Innovation of Jiangsu Province(BE2022011).
文摘Flexible fiber sensors,However,traditional methods face challenges in fabricating low-cost,large-scale fiber sensors.In recent years,the thermal drawing process has rapidly advanced,offering a novel approach to flexible fiber sensors.Through the preform-tofiber manufacturing technique,a variety of fiber sensors with complex functionalities spanning from the nanoscale to kilometer scale can be automated in a short time.Examples include temperature,acoustic,mechanical,chemical,biological,optoelectronic,and multifunctional sensors,which operate on diverse sensing principles such as resistance,capacitance,piezoelectricity,triboelectricity,photoelectricity,and thermoelectricity.This review outlines the principles of the thermal drawing process and provides a detailed overview of the latest advancements in various thermally drawn fiber sensors.Finally,the future developments of thermally drawn fiber sensors are discussed.
基金supported by the National Natural Science Foundation of China(General Program)under Grant 52571385National Key R&D Program of China(Grant No.2024YFC2815000 and No.2024YFB3816000)+12 种基金Open Fund of State Key Laboratory of Deep-sea Manned Vehicles(Grant No.2025SKLDMV07)Shenzhen Science and Technology Program(WDZC20231128114452001,JCYJ20240813112107010 and JCYJ20240813111910014)the Tsinghua SIGS Scientific Research Startup Fund(QD2022021C)the Dreams Foundation of Jianghuai Advance Technology Center(2023-ZM 01 Z006)the Ocean Decade International Cooperation Center(ODCC)(GHZZ3702840002024020000026)Shenzhen Key Laboratory of Advanced Technology for Marine Ecology(ZDSYS20230626091459009)Shenzhen Science and Technology Program(No.KJZD20240903100905008)the National Natural Science Foundation of China(No.22305141)Pearl River Talent Program(No.2023QN10C114)General Program of Guangdong Province(No.2025A1515011700)the Guangdong Innovative and Entrepreneurial Research Team Program(2023ZT10C040)Scientific Research Foundation from Shenzhen Finance Bureau(No.GJHZ20240218113600002)Tsinghua University(JC2023001).
文摘Developing effective,versatile,and high-precision sensing interfaces remains a crucial challenge in human-machine-environment interaction applications.Despite progress in interaction-oriented sensing skins,limitations remain in unit-level reconfiguration,multiaxial force and motion sensing,and robust operation across dynamically changing or irregular surfaces.Herein,we develop a reconfigurable omnidirectional triboelectric whisker sensor array(RO-TWSA)comprising multiple sensing units that integrate a triboelectric whisker structure(TWS)with an untethered hydro-sealing vacuum sucker(UHSVS),enabling reversibly portable deployment and omnidirectional perception across diverse surfaces.Using a simple dual-triangular electrode layout paired with MXene/silicone nanocomposite dielectric layer,the sensor unit achieves precise omnidirectional force and motion sensing with a detection threshold as low as 0.024 N and an angular resolution of 5°,while the UHSVS provides reliable and reversible multi-surface anchoring for the sensor units by involving a newly designed hydrogel combining high mechanical robustness and superior water absorption.Extensive experiments demonstrate the effectiveness of RO-TWSA across various interactive scenarios,including teleoperation,tactile diagnostics,and robotic autonomous exploration.Overall,RO-TWSA presents a versatile and high-resolution tactile interface,offering new avenues for intelligent perception and interaction in complex real-world environments.
基金supported by the Basic Science Research Program(2023R1A2C3004336,RS-202300243807)&Regional Leading Research Center(RS-202400405278)through the National Research Foundation of Korea(NRF)grant funded by the Korea Government(MSIT)。
文摘Wearable sensors integrated with deep learning techniques have the potential to revolutionize seamless human-machine interfaces for real-time health monitoring,clinical diagnosis,and robotic applications.Nevertheless,it remains a critical challenge to simultaneously achieve desirable mechanical and electrical performance along with biocompatibility,adhesion,self-healing,and environmental robustness with excellent sensing metrics.Herein,we report a multifunctional,anti-freezing,selfadhesive,and self-healable organogel pressure sensor composed of cobalt nanoparticle encapsulated nitrogen-doped carbon nanotubes(CoN CNT)embedded in a polyvinyl alcohol-gelatin(PVA/GLE)matrix.Fabricated using a binary solvent system of water and ethylene glycol(EG),the CoN CNT/PVA/GLE organogel exhibits excellent flexibility,biocompatibility,and temperature tolerance with remarkable environmental stability.Electrochemical impedance spectroscopy confirms near-stable performance across a broad humidity range(40%-95%RH).Freeze-tolerant conductivity under sub-zero conditions(-20℃)is attributed to the synergistic role of CoN CNT and EG,preserving mobility and network integrity.The Co N CNT/PVA/GLE organogel sensor exhibits high sensitivity of 5.75 k Pa^(-1)in the detection range from 0 to 20 k Pa,ideal for subtle biomechanical motion detection.A smart human-machine interface for English letter recognition using deep learning achieved 98%accuracy.The organogel sensor utility was extended to detect human gestures like finger bending,wrist motion,and throat vibration during speech.
基金supported by the NSFC(12474071)Natural Science Foundation of Shandong Province(ZR2024YQ051,ZR2025QB50)+6 种基金Guangdong Basic and Applied Basic Research Foundation(2025A1515011191)the Shanghai Sailing Program(23YF1402200,23YF1402400)funded by Basic Research Program of Jiangsu(BK20240424)Open Research Fund of State Key Laboratory of Crystal Materials(KF2406)Taishan Scholar Foundation of Shandong Province(tsqn202408006,tsqn202507058)Young Talent of Lifting engineering for Science and Technology in Shandong,China(SDAST2024QTB002)the Qilu Young Scholar Program of Shandong University。
文摘As emerging two-dimensional(2D)materials,carbides and nitrides(MXenes)could be solid solutions or organized structures made up of multi-atomic layers.With remarkable and adjustable electrical,optical,mechanical,and electrochemical characteristics,MXenes have shown great potential in brain-inspired neuromorphic computing electronics,including neuromorphic gas sensors,pressure sensors and photodetectors.This paper provides a forward-looking review of the research progress regarding MXenes in the neuromorphic sensing domain and discussed the critical challenges that need to be resolved.Key bottlenecks such as insufficient long-term stability under environmental exposure,high costs,scalability limitations in large-scale production,and mechanical mismatch in wearable integration hinder their practical deployment.Furthermore,unresolved issues like interfacial compatibility in heterostructures and energy inefficiency in neu-romorphic signal conversion demand urgent attention.The review offers insights into future research directions enhance the fundamental understanding of MXene properties and promote further integration into neuromorphic computing applications through the convergence with various emerging technologies.
基金supported by National Natural Science Foundation of China(No.52103044)Double First-Class Initiative University of Science and Technology of China(KY2400000037)the Young Talent Programme(GG2400007009).
文摘Conductive elastomers combining micromechanical sensitivity,lightweight adaptability,and environmental sustainability are critically needed for advanced flexible electronics requiring precise responsiveness and long-term wearability;however,the integration of these properties remains a significant challenge.Here,we present a biomass-derived conductive elastomer featuring a rationally engineered dynamic crosslinked network integrated with a tunable microporous architecture.This structural design imparts pronounced micromechanical sensitivity,an ultralow density(~0.25 g cm^(−3)),and superior mechanical compliance for adaptive deformation.Moreover,the unique micro-spring effect derived from the porous architecture ensures exceptional stretchability(>500%elongation at break)and superior resilience,delivering immediate and stable electrical response under both subtle(<1%)and large(>200%)mechanical stimuli.Intrinsic dynamic interactions endow the elastomer with efficient room temperature self-healing and complete recyclability without compromising performance.First-principles simulations clarify the mechanisms behind micropore formation and the resulting functionality.Beyond its facile and mild fabrication process,this work establishes a scalable route toward high-performance,sustainable conductive elastomers tailored for next-generation soft electronics.
基金partially supported by the National Natural Science Foundation of China(Nos.11804120,61827822,and 22275072)Guangdong Basic and Applied Basic Research Foundation(No.2023A1515030209)+1 种基金Research Projects from Guangzhou(Nos.2023A03J0018 and 2024A04J3712)Fundamental Research Funds for the Central Universities(No.21623412).
文摘Fiber-structured ion sensors have gained traction in health monitoring and medical diagnostics owing to their structural flexibility,enhanced sensitivity,and suitability for integration into wearable devices.This study employed a simple and efficient solutionbased process to fabricate nanofibers containing aggregation-induced emission(AIE)dyes.The resulting AIE nanofibers exhibited stable and intense fluorescence,nanosecond fluorescence lifetime,and low-loss light transport when functioning as active waveguides.Additionally,crossed nanofiber intersections exhibited diffraction-limited emission spots.The AIE nanofibers demonstrate efficient and ionspecific fluorescence quenching in response to Ag^(+).These results support the development of sensing units capable of operating in liquid environments or in direct contact with skin or tissues,facilitating real-time monitoring of ion concentrations for personalized healthcare management.
基金the financial support of this work by the National Natural Science Foundation of China(No.52373093)Excellent Youth Found of Natural Science Foundation of Henan Province(No.242300421062)+1 种基金Central Plains Youth Top notch Talent Program of Henan Provincethe 111 project(No.D18023).
文摘The advancement of wearable sensing technologies demands multifunctional materials that integrate high sensitivity,environmental resilience,and intelligent signal processing.In this work,a flexible hydrophobic conductive yarn(FCB@SY)featuring a controllable microcrack structure is developed via a synergistic approach combining ultrasonic swelling and non-solvent induced phase separation(NIPS).By embedding a robust conductive network and engineering microcrack morphology,the resulting sensor achieves an ultrahigh gauge factor(GF≈12,670),an ultrabroad working range(0%-547%),a low detection limit(0.5%),rapid response/recovery time(140 ms/140 ms),and outstanding durability over 10,000 cycles.Furthermore,the hydrophobic surface endowed by conductive coatings imparts exceptional chemical stability against acidic and alkaline environments,as well as reliable waterproof performance.This enables consistent functionality under harsh conditions,including underwater operation.Integrated with machine learning algorithms,the FCB@SY-based intelligent sensing system demonstrates dualmode capabilities in human motion tracking and gesture recognition,offering significant potential for applications in wearable electronics,human-machine interfaces,and soft robotics.
基金supported by the National Natural Science Foundation of China(No.52475580)the Special Foundation of the Taishan Scholar Project(No.tsqn202211077)+3 种基金the Shandong Provincial Natural Science Foundation(No.ZR2023ME118)the Open Project of State Key Laboratory of Chemical Safety(No.SKLCS-2024020)the Fundamental Research Funds for the Central Universities(No.24CX02014A)the Fund of State Key Laboratory of Deep Oil and Gas,China University of Petroleum(East China).
文摘In the context of the rapid development of artificial intelligence and robotics,their application scenarios are continuously expanding to a variety of complex environments,with increasing attention being paid to the use of flexible sensors in lowtemperature environments.In this study,an ionic hydrogel was synthesized using acrylamide(AM),hydroxyethyl cellulose(HEC),and lithium chloride(LiCl)as composites.This hydrogel exhibits high adhesion,excellent sensitivity(gauge factor(GF)=2.84),rapid response time(100 ms),exceptional stretch ability(>1776%),high toughness(2.5 MJ/m^(3)),and the ability to maintain detectability at low temperatures(-60℃).HEC imparts reliable mechanical properties to the sensor through hydrogen bonding interactions of its hydroxyl groups.LiCl ensures that the sensor has outstanding antifreezing properties,maintains good conductivity and mechanical performance.Used for robotic attitude detection,the sensor demonstrated accurate recognition of various joint movements at both 20 and -20℃.This technology was extended to industrial operations and maintenance,where a mechanical claw was used to grasp parts at both room temperature and low temperature.A convolutional neural network deep learning algorithm was employed to identify and classify eight types of parts,achieving an impressive recognition accuracy of 98.8%.The polyacrylamide(PAM)/HEC/LiCl hydrogel sensor demonstrates the capability for wide-temperature range detection in flexible robotics,holding significant potential for future applications in human-machine interaction,tactile perception,and related fields.
文摘Diabetes mellitus represents a major global health issue,driving the need for noninvasive alternatives to traditional blood glucose monitoring methods.Recent advancements in wearable technology have introduced skin-interfaced biosensors capable of analyzing sweat and skin biomarkers,providing innovative solutions for diabetes diagnosis and monitoring.This review comprehensively discusses the current developments in noninvasive wearable biosensors,emphasizing simultaneous detection of biochemical biomarkers(such as glucose,cortisol,lactate,branched-chain amino acids,and cytokines)and physiological signals(including heart rate,blood pressure,and sweat rate)for accurate,personalized diabetes management.We explore innovations in multimodal sensor design,materials science,biorecognition elements,and integration techniques,highlighting the importance of advanced data analytics,artificial intelligence-driven predictive algorithms,and closed-loop therapeutic systems.Additionally,the review addresses ongoing challenges in biomarker validation,sensor stability,user compliance,data privacy,and regulatory considerations.A holistic,multimodal approach enabled by these next-generation wearable biosensors holds significant potential for improving patient outcomes and facilitating proactive healthcare interventions in diabetes management.
基金funded by the National Natural Science Foundation of China(Grant Nos.62322410,52272168,624B2135,61804047)the Fundamental Research Funds for the Central Universities(No.WK2030000103)。
文摘Human action recognition(HAR)is crucial for the development of efficient computer vision,where bioinspired neuromorphic perception visual systems have emerged as a vital solution to address transmission bottlenecks across sensor-processor interfaces.However,the absence of interactions among versatile biomimicking functionalities within a single device,which was developed for specific vision tasks,restricts the computational capacity,practicality,and scalability of in-sensor vision computing.Here,we propose a bioinspired vision sensor composed of a Ga N/Al N-based ultrathin quantum-disks-in-nanowires(QD-NWs)array to mimic not only Parvo cells for high-contrast vision and Magno cells for dynamic vision in the human retina but also the synergistic activity between the two cells for in-sensor vision computing.By simply tuning the applied bias voltage on each QD-NW-array-based pixel,we achieve two biosimilar photoresponse characteristics with slow and fast reactions to light stimuli that enhance the in-sensor image quality and HAR efficiency,respectively.Strikingly,the interplay and synergistic interaction of the two photoresponse modes within a single device markedly increased the HAR recognition accuracy from 51.4%to 81.4%owing to the integrated artificial vision system.The demonstration of an intelligent vision sensor offers a promising device platform for the development of highly efficient HAR systems and future smart optoelectronics.
基金supported by the National Natural Science Foundation of China(NSFC 52175281,52475315)Youth Innovation Promotion Association of CAS(2021382)。
文摘The growing prevalence of exercise-induced tibial stress fractures demands wearable sensors capable of monitoring dynamic musculoskeletal loads with medical-grade precision.While flexible pressure-sensing insoles show clinical potential,their development has been hindered by the intrinsic trade-off between high sensitivity and full-range linearity(R^(2)>0.99 up to 1 MPa)in conventional designs.Inspired by the tactile sensing mechanism of human skin,where dermal stratification enables wide-range pressure adaptation and ion-channelregulated signaling maintains linear electrical responses,we developed a dual-mechanism flexible iontronic pressure sensor(FIPS).This innovative design synergistically combines two bioinspired components:interdigitated fabric microstructures enabling pressure-proportional contact area expansion(αP1/3)and iontronic film facilitating self-adaptive ion concentration modulation(αP^(2/3)),which together generate a linear capacitance-pressure response(CαP).The FIPS achieves breakthrough performance:242 kPa^(-1)sensitivity with 0.997linearity across 0-1 MPa,yielding a record linear sensing factor(LSF=242,000).The design is validated across various substrates and ionic materials,demonstrating its versatility.Finally,the FIPS-driven design enables a smart insole demonstrating 1.8%error in tibial load assessment during gait analysis,outperforming nonlinear counterparts(6.5%error)in early fracture-risk prediction.The biomimetic design framework establishes a universal approach for developing high-performance linear sensors,establishing generalized principles for medical-grade wearable devices.
基金supported by the NSFC under Grant No.62474169the National Key Research and Development Program of China under Grant No.2024YFB3212200the funding from USTC under Grant Nos.WK2100000025,KY2190000003,and KY2190000006。
文摘Soft X-ray detectors play a vital role in materials science,high-energy physics and medical imaging.Cs_(2)AgBiBr_(6),a lead-free double perovskite,has gained attention for its excellent optoelectronic properties,stability,and nontoxicity.However,its fast crystallization and requirement for high-temperature annealing(>250℃)often lead to inferior film quality,limiting its application in flexible devices.This study introduces an alloying strategy that significantly improves the quality of Cs_(2)AgBiBr_(6)thin films annealed at a reduced temperature of 150℃.Devices based on the alloyed thin films exhibit an ultra-low dark current of 0.32 nA·cm^(-2)and a quantum efficiency of 725%.Furthermore,the first successful integration of Cs_(2)AgBiBr_(6)with a thinfilm transistor backplane demonstrates its superior imaging performance,indicating that Cs_(2)AgBiBr_(6)is a promising material for next-generation soft X-ray sensors.
文摘In order to solve the problem of the invalidation of thermal parameters andoptimal running, we present an efficient soft sensor approach based on sparse online Gaussianprocesses( GP), which is based on a combination of a Bayesian online algorithm together with asequential construction of a relevant subsample of the data to specify the prediction of the GPmodel. By an appealing parameterization and projection techniques that use the reproducing kernelHubert space (RKHS) norm, recursions for the effective parameters and a sparse Gaussianapproximation of the posterior process are obtained. The sparse representation of Gaussian processesmakes the GP-based soft sensor practical in a large dataset and real-time application. And theproposed thermalparameter soft sensor is of importance for the economical running of the powerplant.
文摘近年来,综合能源系统作为一种以多种能源形态和设备相互交互的能源系统方案得到了广泛应用和研究.然而,在面对动态复杂的多能源系统时,传统的优化调度方法往往无法满足其实时性和精准度需求.因此,本文设计了一种软深度确定性策略梯度(Soft Deep Deterministic Policy Gradient,Soft-DDPG)算法驱动的综合能源系统优化调度方法,以最小化调度周期内系统总运行成本为目标,建立设备运行综合能效评估模型,再采用Soft-DDPG算法对每个能源设备的能效调度动作进行优化控制.Soft-DDPG算法将softmax算子引入到动作值函数的计算中,有效降低了Q值高估问题.与此同时,该算法在动作选择策略中加入了随机噪声,提高了算法的学习效率.实验结果显示,本文所提出的方法解决了综合能源系统能效调度实时性差、精准度低的瓶颈问题,实现了系统的高效灵活调度,降低了系统的总运行成本.
