The original intention of Model Predictive Control(MPC)was to bridge the gap between advanced control theories and the application of complex practical engineering systems.Since its emergence in the 1970s,it has been ...The original intention of Model Predictive Control(MPC)was to bridge the gap between advanced control theories and the application of complex practical engineering systems.Since its emergence in the 1970s,it has been one of the core technologies in the Advanced Process Control(APC)system.Owing to its receding horizon optimization mechanism and capability to handle constraints,MPC has consistently attracted attention from both academia and industry,and has been successfully applied in numerous industries such as petrochemicals,metallurgy,pharmaceuticals,electric power,water treatment,and power electronics.According to statistics from the International Federation of Automatic Control and industrial data,MPC has been deployed in some form in more than 50% of large-scale industrial control systems worldwide.展开更多
To address the limitation problem of existing chemical processes and equipment in micro-chemical systems,this study proposes an array-type integrated distillation unit model.The multivariable collaborative control of ...To address the limitation problem of existing chemical processes and equipment in micro-chemical systems,this study proposes an array-type integrated distillation unit model.The multivariable collaborative control of product concentration and top pressure in the integrated distillation system is achieved via multi-agent technology.To reduce the frequent changes in feed flow and reboiler temperature input control commands,an event-triggered mechanism is introduced to ensure smooth system operation.For obtaining the states of each production variable in the process,an array integrated simulation system for the distillation unit is developed,which successfully collects data sets of key production and control variables through process software simulation.Utilizing subspace system modeling techniques and normalized datasets,the state equations for single-cluster systems within the array-integrated model are constructed.Subsequently,the proposed coordinated control algorithm is applied to the production process of o-chlorotoluene within a limited range of operating variables.Finally,the proposed control methodology is verified via simulations using Aspen Plus process data to ensure the continuity of chemical production process.展开更多
BACKGROUND Patient-specific quality assurance(QA)is an essential component in the safe and precise delivery of radiotherapy,particularly in head and neck cancer cases where anatomical complexity and proximity to criti...BACKGROUND Patient-specific quality assurance(QA)is an essential component in the safe and precise delivery of radiotherapy,particularly in head and neck cancer cases where anatomical complexity and proximity to critical structures increase the risk of delivery deviations.As treatment techniques become more conformal and modulated,conventional gamma analysis often lacks sensitivity to subtle variations,necessitating advanced statistical tools to verify the accuracy of dose delivery.AIM To assess patient-specific QA performance,dosimetric accuracy,and long-term consistency using gamma analysis with statistical process control methods.METHODS A retrospective analysis of intensity-modulated radiation therapy and volumetric modulated arc therapy head and neck treatment plans was performed using electronic portal imaging device-based gamma pass rates under global criteria.Individual and Moving Range charts were generated to assess QA stability and detect trends.Initial control limits were derived from 20 QA plans and validated on 350 QA plans.Tolerance limits(TL)and action limits(AL)were calculated.The resulting charts offered a robust framework for monitoring QA performance and identifying deviations.RESULTS The central line values decreased progressively with stricter gamma criteria,from 98.835%(3%/3 mm)to 94.4%(2%/2 mm).TL and AL narrowed correspondingly,with TL dropping from 97.869%to 93.672%.Exponentially Weighted Moving Average charts provided smoother detection of persistent small deviations compared to Individual and Moving Range charts,enhancing sensitivity.Across all criteria,QA processes remained statistically stable,although tighter thresholds reduced pass rates.CONCLUSION Gamma analysis is a reliable method for patient-specific QA in head and neck radiotherapy.The incorporation of statistical process control adds a valuable layer of continuous monitoring,supports site-specific tolerance/AL,and strengthens treatment accuracy.展开更多
As a concealed project,the bearing characteristics and construction quality of highway bridge pile foundations directly determine the safety and durability of bridge structures.However,there are still problems in the ...As a concealed project,the bearing characteristics and construction quality of highway bridge pile foundations directly determine the safety and durability of bridge structures.However,there are still problems in the current pile foundation construction,such as poor geological adaptability,lax control of process parameters,and outdated detection methods,resulting in insufficient bearing capacity and frequent quality hazards.This article refers to the standardized academic paper architecture,systematically analyzes the load transfer mechanism and bearing characteristics evolution law of pile foundations,constructs a quality control system guided by bearing mechanisms,with process control as the core and dynamic detection as the guarantee,proposes the implementation path from"experience construction"to"data-driven construction",studies the control methods of key links such as geological exploration,pore forming technology,concrete pouring,integrity testing,etc.,aiming to provide theoretical and practical support for improving the construction quality and optimizing the bearing performance of highway bridge pile foundations.展开更多
Municipal projects are core components of urban infrastructure construction, characterized by wide‑ranging construction contents, complex on‑site conditions, numerous influencing factors and high risk of cost overrun....Municipal projects are core components of urban infrastructure construction, characterized by wide‑ranging construction contents, complex on‑site conditions, numerous influencing factors and high risk of cost overrun. Traditional cost control mostly focuses on post‑audit and static cost accounting, separating technical management from cost management, which easily leads to design defects, frequent engineering changes and out‑of‑control investment. Starting from the perspective of engineering technology management, this paper takes the whole‑life cycle of municipal projects as the main line, sorts out prominent cost‑control problems existing in decision‑making, design, bidding‑tendering, construction and completion‑acceptance stages. Combined with technical management means such as limit design, BIM digital technology, construction scheme optimization and technical‑economic demonstration, systematic whole‑process cost‑control strategies are put forward. The study indicates that deep integration of engineering technology management and cost control can realize pre‑risk prevention, in‑process dynamic monitoring and post‑event standardized settlement, effectively reduce invalid investment, restrain cost overrun risk and improve comprehensive benefit of government‑invested municipal projects.展开更多
Experiments were conducted to evaluate the microstructure and tensile properties of a medium carbon Cr-Ni-W-Mo steel processedthermo-mechanical controlled processing(TMCP)with cooling at different conditions in water,...Experiments were conducted to evaluate the microstructure and tensile properties of a medium carbon Cr-Ni-W-Mo steel processedthermo-mechanical controlled processing(TMCP)with cooling at different conditions in water,oil,air or lime followedlow tempering.Compared to normal heat-treatment processing,TMCP with water-cooling after deformation enhances the yield strength and tensile strength of the steelabout 323 MPa and about 251 MPa,respectively,due to higher dislocation strengthening and grain boundary strengthening.Meanwhile,it increases the elongation by ;about 1.76%attributed to the increase in volume percentage of the retained austenite and the refined laths of tempered martensite.Slowing the cooling rate after deformation during TMCP leads to a decrease in the strength.This results the coupling effectsthe reduction in dislocation density and volume fraction of tempered martensite together with the coarseness in martensite sizes.However,cooling rate decreasing has less influences on ductility becathe improved elongation the increase in the volume fractions of both retained austenite and lower bainite together with dislocation density decreasing is compensatedthe reduced elongation coarsened grains.展开更多
Abstract Most papers in scheduling research have treated individual job processing times as fixed parameters. However, in many practical situations, a manager may control processing time by reallocating resources. In ...Abstract Most papers in scheduling research have treated individual job processing times as fixed parameters. However, in many practical situations, a manager may control processing time by reallocating resources. In this paper, authors consider a machine scheduling problem with controllable processing times. In the first part of this paper, a special case where the processing times and compression costs are uniform among jobs is discussed. Theoretical results are derived that aid in developing an O(n 2) algorithm to slove the problem optimally. In the second part of this paper, authors generalize the discussion to general case. An effective heuristic to the general problem will be presented.展开更多
Thermomechanical controlled processing (TMCP) of low carbon cold heading steel in different austenite conditions were conducted by a laboratory hot rolling mill. Effect of various processing parameters on the mechan...Thermomechanical controlled processing (TMCP) of low carbon cold heading steel in different austenite conditions were conducted by a laboratory hot rolling mill. Effect of various processing parameters on the mechanical properties of the steel was investigated. The results showed that the mechanical properties of the low carbon cold heading steel could be significantly improved by TMCP without heat treatment. The improvement of mechanical properties can be attributed mainly to the ferrite grain refinement due to low temperature rolling. In the experiments the better ultimate tensile strength and ductility are obtained by lowering finishing cooling temperature within the temperature range from 650 ℃ to 550 ℃ since the interlamellar space in pearlite colonies become smaller. Good mechanical properties can be obtained in a proper austenite condition and thermomechanical processing parameter. The ferrite morphology has a more pronounced effect on the mechanical behavior than refinement of the microstructure. It is possible to realize the replacement of medium-carbon by low-carbon for 490 MPa grade cold heading steel with TMCP.展开更多
During the sintering process of iron ore,a large amount of nitrogen oxides is generated,for which there is currently no efficient and economical treatment process.Therefore,it is necessary to implement process control...During the sintering process of iron ore,a large amount of nitrogen oxides is generated,for which there is currently no efficient and economical treatment process.Therefore,it is necessary to implement process control in sintering production to keep the mass concentration of NOxin sintering flue gas at a low level.Through industrial trials at sintering sites,methods such as correlation analysis,path analysis,and multiple linear regression were applied to analyze the influence of various factors on NO emissions during the sintering process.The results indicate that negative correlations exist between nitrogen monoxide(NO)emissions and negative pressure,permeability index,O2 concentration,CO concentration,and flue gas temperature.Conversely,positive correlations exist between NO emissions and dust concentration,water vapor volume fraction,and sintering bed speed.Among these factors,O2 concentration and dust concentration are identified as the most significant influencing factors on NO emissions.By analyzing the masses and modes of influence of different factors,the mechanisms of action of each factor were obtained.Specifically,O2 concentration,dust concentration,permeability index,CO concentration,and flue gas temperature play a direct dominant role in NO emissions during the sintering process,while water vapor volume fraction,sintering trolley speed,and negative pressure have an indirect effect.A predictive model for NO mass concentration in flue gas was established with an accuracy rate of 91.6%,showing consistent overall trends with actual values.Finally,denitrification strategies for sintering industrial production were proposed,along with prospects for preliminary denitrification of sintering flue gas using fluidized bed conditions in the duct.展开更多
A mathematical model of the decarburization reaction zone was established for the Ruhrstahl–Heraeus (RH) forced oxygen blowing decarburization process by Matlab R2022b software. For the problem of inaccurate predicti...A mathematical model of the decarburization reaction zone was established for the Ruhrstahl–Heraeus (RH) forced oxygen blowing decarburization process by Matlab R2022b software. For the problem of inaccurate prediction due to the large variation range of oxygen absorption rate under different process conditions, we statistically analyzed the main factors affecting the oxygen absorption rate. The backpropagation neural network was used to train and predict the oxygen absorption rate and was used to calculate the RH decarburization reaction zone model. We designed and developed a mathematical modeling software with process control of decarburization in RH degasser, which can realize the change of operating process parameters in the dynamic prediction process. The optimized mathematical model has more than 95% of the furnaces whose absolute error in calculation of carbon content is within ± 5 × 10−6, more than 90% of the heats whose relative error in calculation of oxygen content is within ± 15%, and the average absolute error of calculation of oxygen content is 26.4 × 10−6. Finally, we studied the influence of oxygen blowing timing, oxygen blowing volume and initial oxygen content on the forced decarburization process.展开更多
This study presents an innovative development of the exponentially weighted moving average(EWMA)control chart,explicitly adapted for the examination of time series data distinguished by seasonal autoregressive moving ...This study presents an innovative development of the exponentially weighted moving average(EWMA)control chart,explicitly adapted for the examination of time series data distinguished by seasonal autoregressive moving average behavior—SARMA(1,1)L under exponential white noise.Unlike previous works that rely on simplified models such as AR(1)or assume independence,this research derives for the first time an exact two-sided Average Run Length(ARL)formula for theModified EWMAchart under SARMA(1,1)L conditions,using a mathematically rigorous Fredholm integral approach.The derived formulas are validated against numerical integral equation(NIE)solutions,showing strong agreement and significantly reduced computational burden.Additionally,a performance comparison index(PCI)is introduced to assess the chart’s detection capability.Results demonstrate that the proposed method exhibits superior sensitivity to mean shifts in autocorrelated environments,outperforming existing approaches.The findings offer a new,efficient framework for real-time quality control in complex seasonal processes,with potential applications in environmental monitoring and intelligent manufacturing systems.展开更多
L2 reading is not only an important channel for people to obtain information and knowledge,but also the main way for people to learn a foreign language.Reading information processing can be divided into controlled pro...L2 reading is not only an important channel for people to obtain information and knowledge,but also the main way for people to learn a foreign language.Reading information processing can be divided into controlled processing and automatic processing.Controlled information processing is a conscious and resource-intensive processing model,while automatic information processing is an unconscious and automatic processing model.This study investigates the characteristics and interactivity of controlled and automatic information processing in L2 reading,and explores the roles of controlled and automatic information processing strategies in improving L2 reading ability.The findings are as follows:(a)controlled and automatic information processing is interactive in L2 reading;and(b)the uses of controlled and automatic information processing strategies are beneficial to the improvement of the reading ability of L2 learners.This study has important theoretical and practical value in improving the efficiency of L2 reading teaching and learning.展开更多
Directed energy deposition arc welding(DED-arc)encounters challenges such as low forming precision and compromised process stability during the deposition process owing to its intricate molten pool dynamics.The measur...Directed energy deposition arc welding(DED-arc)encounters challenges such as low forming precision and compromised process stability during the deposition process owing to its intricate molten pool dynamics.The measurement and control of the molten-pool size are essential to addressing these problems.However,most contemporary research focuses on controlling a single variable(e.g.,molten-pool width or height)in either open-or closed-loop configurations.This approach is inherently limited in addressing the dynamic nonlinear characteristics and high-precision requirements of the DED-arc process.To address this challenge,we designed a binocular passive vision sensing system that utilizes an image processing algorithm to facilitate the real-time measurement and size extraction of the molten-pool width and height.A dual fuzzy proportional–integral–derivative controller was also developed to control the width and height of the molten pool by adjusting the arc current and voltage separately.This controller was then compared with open-loop control experiments under constant parameters.The results demonstrate that the proposed bivariate control method can effectively regulate the error range of the actual and detected values of the molten-pool width and height within 0.5 and 0.3 mm,respectively,thereby significantly enhancing the control accuracy and optimizing the forming appearance of thin-walled metal parts.展开更多
The adaptive feedback control of stability with circumferential inlet distortion has been experimentally investigated in a low-speed,axial compressor.The flat-baffles with different span heights are used to simulate d...The adaptive feedback control of stability with circumferential inlet distortion has been experimentally investigated in a low-speed,axial compressor.The flat-baffles with different span heights are used to simulate different distorted inflow cases.Compared with auto-correlation and root-mean-square analysis,cross-correlation analysis used to predict early stall warning does not depend on the distortion position.Hence,the cross-correlation coefficient was used to monitor the stable status of the compressor and provide the feedback signal in the active control strategy when suffering from different distortions.Based on the stall margin improvement of tip air injection obtained under different distorted inflow cases and the sensitivity analysis of cross-correlation coefficients to injected momentum ratios,tip air injection was adopted as the actuator for adaptive feedback control.The digital signal processing controller was designed and applied to achieve adaptive feedback control in distorted inflow conditions.The results show that the adaptive feedback control of air injection nearly achieves the same stall margin improvement as steady air injection under different distortion intensities with a reduced injection mass flow.Thus,the proposed adaptive feedback control method is ideal for the engine operation with circumferential distorted inflow,which frequently occurs in flight.展开更多
Fiber quality measurement in spinning preparation is crucial for optimizing waste and meeting yarn quality specifications.The brand-new Uster AFIS 6–the next-generation laboratory instrument from Uster Technologies–...Fiber quality measurement in spinning preparation is crucial for optimizing waste and meeting yarn quality specifications.The brand-new Uster AFIS 6–the next-generation laboratory instrument from Uster Technologies–uniquely tests man-made fiber properties in addition to cotton.It provides critical data to optimize fiber process control for cotton,man-made fibers,and blended yarns.展开更多
Traditional tools have limited adaptability in complex machining environments due to their lack of working-condition perception and autonomous regulation.With advances in sensors,materials,and data-processing technolo...Traditional tools have limited adaptability in complex machining environments due to their lack of working-condition perception and autonomous regulation.With advances in sensors,materials,and data-processing technologies,tool design is shifting from a single-function‘mechanical arm’for cutting towards integrated intelligent terminals.This paper systematically reviews progress in intelligent tool technology from two perspectives:design and regulation.For intelligent design,the fundamental principles of condition-perception tools equipped with built-in multi-type sensors are discussed,enabling in situ,real-time monitoring of multidimensional parameters such as cutting force,temperature,and vibration.Force monitoring is achieved through elastic deformation or dynamic charge response,temperature monitoring through the thermoelectric effect,and vibration monitoring through micro-displacement and intensity detection.The design focus emphasises sensor miniaturisation and integration,balancing measurement accuracy with tool stiffness while minimising machining interference.In regulation,key technologies for constructing closed-loop control systems(CLCS)are summarised,which dynamically adjust cutting speed,feed rate,and other parameters based on sensed data,achieving precise control of force,temperature,and vibration via feedback mechanisms and driving units.Breakthroughs in tool wear compensation(TWC)mechanisms are introduced.Multi-source signal fusion combined with deep learning algorithms is further examined for improving monitoring accuracy and remaining useful life(RUL)prediction.Through model predictive control,intelligent regulation of cutting parameters within process flows is realised.Finally,challenges such as sensor reliability,multi-source coupling,and balancing cost with industrial applicability are analysed.Future directions highlight novel structural designs,high-performance material development,and multi-technology integration,aiming to establish a fully intelligent machining system through the integrated design of‘perception-decision-execution’.展开更多
Multi-beam scanning electron microscope(MBSEM)reconciles the inherent contradiction between“resolution and throughput”of traditional scanning electron microscopes(SEMs)through parallel electron beam manipulation,eme...Multi-beam scanning electron microscope(MBSEM)reconciles the inherent contradiction between“resolution and throughput”of traditional scanning electron microscopes(SEMs)through parallel electron beam manipulation,emerging as a key technology to address the bottlenecks in large-volume,high-resolution imaging and advanced industrial inspection.This article provides a comprehensive overview of MBSEM,covering its evolutionary course of technology,core design fundamentals,and interdisciplinary applications.First,it sorts out the evolutionary process from conceptualization in the early 21 st century to commercialization in the 2010s,clarifying the core logic of breaking through the physical limitations of single-beam systems via“multi-beam parallelism”.Subsequently,focusing on the core technological chain of“beam generation–optical focusing–signal detection–data processing”,it conducts an in-depth analysis of the design concepts,technical characteristics,and applicable scenarios of three mainstream architectures:the single-source single-column,split optical system,and semiconductor-specific multi-beam inspection(MBI).Combined with representative research and product data from teams such as Delft University of Technology,Zeiss,and ASML/HMI,it reveals the differentiated advantages of each architecture in beam uniformity(the relative deviation percentage of the current density and probe size of each sub-beam in the multi-beam array from the central beam,with a smaller deviation value indicating better uniformity)and signal crosstalk(the percentage of the interference signal intensity to the target signal intensity when the detection signal of a single beam in the multi-beam array interferes with the detection channel of adjacent beams)control,and throughput(the percentage of the interference signal intensity to the target signal intensity when the detection signal of a single beam in the multi-beam array interferes with the detection channel of adjacent beams)improvement.Finally,integrating the practical demands of neuroscience connectomics,advanced semiconductor manufacturing processes,and biomedicine,it elaborates on the application breakthroughs of MBSEM in the three-dimensional(3D)reconstruction of large-volume brain tissue,wafer defect screening for 7 nm and smaller nodes,and low-damage imaging of thin biological tissues,and compares its disruptive value relative to traditional technologies(single-beam SEM,optical inspection).Unlike previous reviews,this work systematically integrates both academic prototypes and industrial systems for the first time,providing an in-depth analysis of the differentiated trade-offs among imaging speed,resolution,and sample adaptability across different architectures,offering a systematic reference for MBSEM R&D and interdisciplinary applications.展开更多
Ethylene yield serves as a key metric in petrochemical production,where optimizing its energy efficiency remains a critical challenge for sustainable production.Meanwhile,manual hyperparameter tuning of deep learning ...Ethylene yield serves as a key metric in petrochemical production,where optimizing its energy efficiency remains a critical challenge for sustainable production.Meanwhile,manual hyperparameter tuning of deep learning based yield prediction models often results in suboptimal configurations,which reduces prediction accuracy and reliability due to extreme operating conditions generating outliers in ethylene production processes.Therefore,a novel neural network automatic design method(NNADM)is proposed,which incorporates the neural network parameters automatic optimization and loss function adaptive construction.An innovative adaptive loss formulation is proposed to strategically integrate the complementary strengths of the mean squared error(MSE)and the Log-Cosh functions,featuring dynamic outlier resistance through self-adjusting weight coefficients.Then,the Bayesian optimization search algorithm is utilized to discover optimal hyperparameters of the neural network,including hidden layer unit,epoch,batch size,and the loss function.Finally,the NNADM is integrated with several classical neural networks for ethylene yield prediction.Experimental results show that several classical neural networks realize an average of 16.86%decrease in MSE index after integrating the NNADM.In addition,the proposed model offers direction and development blueprints for ethylene production facilities that have low energy efficiency.By implementing this model,it is possible to cut approximately 8376.4 tons of carbon emissions and simultaneously secure an extra 499 tons of ethylene output.展开更多
Recent advances in artificial intelligence(AI)have significantly enhanced quality management,enabling more effective handling of complex,high-dimensional,and multi-modal data.AI methods,including machine learning(ML)a...Recent advances in artificial intelligence(AI)have significantly enhanced quality management,enabling more effective handling of complex,high-dimensional,and multi-modal data.AI methods,including machine learning(ML)and deep learning(DL),have been pivotal in advancing key areas such as quality optimization,monitoring,and diagnosis.These methods have increased adaptability,efficiency,and scalability,making them particularly suitable for modern industrial applications.This review provides a comprehensive examination of AI methods in quality management,covering the integration of surrogate models,Bayesian optimization(BO),intelligent control charts,change-point detection(CPD),and interpretable quality diagnosis.The review concludes with proposed directions for future research aimed at overcoming existing challenges and enhancing the deployment of AI in real-world quality management implementation.展开更多
摘要The original intention of Model Predictive Control(MPC)was to bridge the gap between advanced control theories and the application of complex practical engineering systems.Since its emergence in the 1970s,it has been one of the core technologies in the Advanced Process Control(APC)system.Owing to its receding horizon optimization mechanism and capability to handle constraints,MPC has consistently attracted attention from both academia and industry,and has been successfully applied in numerous industries such as petrochemicals,metallurgy,pharmaceuticals,electric power,water treatment,and power electronics.According to statistics from the International Federation of Automatic Control and industrial data,MPC has been deployed in some form in more than 50% of large-scale industrial control systems worldwide.
基金supported in part by the National Natural Science Foundation of China(62173178,62203214 and 62333010)the National Key Research&Development Program of China(2022YFB3305300)。
摘要To address the limitation problem of existing chemical processes and equipment in micro-chemical systems,this study proposes an array-type integrated distillation unit model.The multivariable collaborative control of product concentration and top pressure in the integrated distillation system is achieved via multi-agent technology.To reduce the frequent changes in feed flow and reboiler temperature input control commands,an event-triggered mechanism is introduced to ensure smooth system operation.For obtaining the states of each production variable in the process,an array integrated simulation system for the distillation unit is developed,which successfully collects data sets of key production and control variables through process software simulation.Utilizing subspace system modeling techniques and normalized datasets,the state equations for single-cluster systems within the array-integrated model are constructed.Subsequently,the proposed coordinated control algorithm is applied to the production process of o-chlorotoluene within a limited range of operating variables.Finally,the proposed control methodology is verified via simulations using Aspen Plus process data to ensure the continuity of chemical production process.
摘要BACKGROUND Patient-specific quality assurance(QA)is an essential component in the safe and precise delivery of radiotherapy,particularly in head and neck cancer cases where anatomical complexity and proximity to critical structures increase the risk of delivery deviations.As treatment techniques become more conformal and modulated,conventional gamma analysis often lacks sensitivity to subtle variations,necessitating advanced statistical tools to verify the accuracy of dose delivery.AIM To assess patient-specific QA performance,dosimetric accuracy,and long-term consistency using gamma analysis with statistical process control methods.METHODS A retrospective analysis of intensity-modulated radiation therapy and volumetric modulated arc therapy head and neck treatment plans was performed using electronic portal imaging device-based gamma pass rates under global criteria.Individual and Moving Range charts were generated to assess QA stability and detect trends.Initial control limits were derived from 20 QA plans and validated on 350 QA plans.Tolerance limits(TL)and action limits(AL)were calculated.The resulting charts offered a robust framework for monitoring QA performance and identifying deviations.RESULTS The central line values decreased progressively with stricter gamma criteria,from 98.835%(3%/3 mm)to 94.4%(2%/2 mm).TL and AL narrowed correspondingly,with TL dropping from 97.869%to 93.672%.Exponentially Weighted Moving Average charts provided smoother detection of persistent small deviations compared to Individual and Moving Range charts,enhancing sensitivity.Across all criteria,QA processes remained statistically stable,although tighter thresholds reduced pass rates.CONCLUSION Gamma analysis is a reliable method for patient-specific QA in head and neck radiotherapy.The incorporation of statistical process control adds a valuable layer of continuous monitoring,supports site-specific tolerance/AL,and strengthens treatment accuracy.
摘要As a concealed project,the bearing characteristics and construction quality of highway bridge pile foundations directly determine the safety and durability of bridge structures.However,there are still problems in the current pile foundation construction,such as poor geological adaptability,lax control of process parameters,and outdated detection methods,resulting in insufficient bearing capacity and frequent quality hazards.This article refers to the standardized academic paper architecture,systematically analyzes the load transfer mechanism and bearing characteristics evolution law of pile foundations,constructs a quality control system guided by bearing mechanisms,with process control as the core and dynamic detection as the guarantee,proposes the implementation path from"experience construction"to"data-driven construction",studies the control methods of key links such as geological exploration,pore forming technology,concrete pouring,integrity testing,etc.,aiming to provide theoretical and practical support for improving the construction quality and optimizing the bearing performance of highway bridge pile foundations.
摘要Municipal projects are core components of urban infrastructure construction, characterized by wide‑ranging construction contents, complex on‑site conditions, numerous influencing factors and high risk of cost overrun. Traditional cost control mostly focuses on post‑audit and static cost accounting, separating technical management from cost management, which easily leads to design defects, frequent engineering changes and out‑of‑control investment. Starting from the perspective of engineering technology management, this paper takes the whole‑life cycle of municipal projects as the main line, sorts out prominent cost‑control problems existing in decision‑making, design, bidding‑tendering, construction and completion‑acceptance stages. Combined with technical management means such as limit design, BIM digital technology, construction scheme optimization and technical‑economic demonstration, systematic whole‑process cost‑control strategies are put forward. The study indicates that deep integration of engineering technology management and cost control can realize pre‑risk prevention, in‑process dynamic monitoring and post‑event standardized settlement, effectively reduce invalid investment, restrain cost overrun risk and improve comprehensive benefit of government‑invested municipal projects.
基金supported by the National Natural Science Foundation of China under Grant No.51671030.
摘要Experiments were conducted to evaluate the microstructure and tensile properties of a medium carbon Cr-Ni-W-Mo steel processedthermo-mechanical controlled processing(TMCP)with cooling at different conditions in water,oil,air or lime followedlow tempering.Compared to normal heat-treatment processing,TMCP with water-cooling after deformation enhances the yield strength and tensile strength of the steelabout 323 MPa and about 251 MPa,respectively,due to higher dislocation strengthening and grain boundary strengthening.Meanwhile,it increases the elongation by ;about 1.76%attributed to the increase in volume percentage of the retained austenite and the refined laths of tempered martensite.Slowing the cooling rate after deformation during TMCP leads to a decrease in the strength.This results the coupling effectsthe reduction in dislocation density and volume fraction of tempered martensite together with the coarseness in martensite sizes.However,cooling rate decreasing has less influences on ductility becathe improved elongation the increase in the volume fractions of both retained austenite and lower bainite together with dislocation density decreasing is compensatedthe reduced elongation coarsened grains.
摘要Abstract Most papers in scheduling research have treated individual job processing times as fixed parameters. However, in many practical situations, a manager may control processing time by reallocating resources. In this paper, authors consider a machine scheduling problem with controllable processing times. In the first part of this paper, a special case where the processing times and compression costs are uniform among jobs is discussed. Theoretical results are derived that aid in developing an O(n 2) algorithm to slove the problem optimally. In the second part of this paper, authors generalize the discussion to general case. An effective heuristic to the general problem will be presented.
基金Sponsored by National Natural Science Foundation of China (50334010)Shenyang City Application Basic Research Project (1071198-1-00)
摘要Thermomechanical controlled processing (TMCP) of low carbon cold heading steel in different austenite conditions were conducted by a laboratory hot rolling mill. Effect of various processing parameters on the mechanical properties of the steel was investigated. The results showed that the mechanical properties of the low carbon cold heading steel could be significantly improved by TMCP without heat treatment. The improvement of mechanical properties can be attributed mainly to the ferrite grain refinement due to low temperature rolling. In the experiments the better ultimate tensile strength and ductility are obtained by lowering finishing cooling temperature within the temperature range from 650 ℃ to 550 ℃ since the interlamellar space in pearlite colonies become smaller. Good mechanical properties can be obtained in a proper austenite condition and thermomechanical processing parameter. The ferrite morphology has a more pronounced effect on the mechanical behavior than refinement of the microstructure. It is possible to realize the replacement of medium-carbon by low-carbon for 490 MPa grade cold heading steel with TMCP.
基金supported by the National Natural Science Foundation of China(No.51974131)Hebei Outstanding Youth Fund Project(No.E2020209082),Tangshan Key R&D Program project(No.22150232J)Sixth Division Wujiaqu City Science and Technology Plan Project(2410).
摘要During the sintering process of iron ore,a large amount of nitrogen oxides is generated,for which there is currently no efficient and economical treatment process.Therefore,it is necessary to implement process control in sintering production to keep the mass concentration of NOxin sintering flue gas at a low level.Through industrial trials at sintering sites,methods such as correlation analysis,path analysis,and multiple linear regression were applied to analyze the influence of various factors on NO emissions during the sintering process.The results indicate that negative correlations exist between nitrogen monoxide(NO)emissions and negative pressure,permeability index,O2 concentration,CO concentration,and flue gas temperature.Conversely,positive correlations exist between NO emissions and dust concentration,water vapor volume fraction,and sintering bed speed.Among these factors,O2 concentration and dust concentration are identified as the most significant influencing factors on NO emissions.By analyzing the masses and modes of influence of different factors,the mechanisms of action of each factor were obtained.Specifically,O2 concentration,dust concentration,permeability index,CO concentration,and flue gas temperature play a direct dominant role in NO emissions during the sintering process,while water vapor volume fraction,sintering trolley speed,and negative pressure have an indirect effect.A predictive model for NO mass concentration in flue gas was established with an accuracy rate of 91.6%,showing consistent overall trends with actual values.Finally,denitrification strategies for sintering industrial production were proposed,along with prospects for preliminary denitrification of sintering flue gas using fluidized bed conditions in the duct.
基金supported by the Central Government Guides Local Science and Technology Development Foundation(No.2023JH6/100100046)the Project funded by China Postdoctoral Science Foundation(No.2023M730230).
摘要A mathematical model of the decarburization reaction zone was established for the Ruhrstahl–Heraeus (RH) forced oxygen blowing decarburization process by Matlab R2022b software. For the problem of inaccurate prediction due to the large variation range of oxygen absorption rate under different process conditions, we statistically analyzed the main factors affecting the oxygen absorption rate. The backpropagation neural network was used to train and predict the oxygen absorption rate and was used to calculate the RH decarburization reaction zone model. We designed and developed a mathematical modeling software with process control of decarburization in RH degasser, which can realize the change of operating process parameters in the dynamic prediction process. The optimized mathematical model has more than 95% of the furnaces whose absolute error in calculation of carbon content is within ± 5 × 10−6, more than 90% of the heats whose relative error in calculation of oxygen content is within ± 15%, and the average absolute error of calculation of oxygen content is 26.4 × 10−6. Finally, we studied the influence of oxygen blowing timing, oxygen blowing volume and initial oxygen content on the forced decarburization process.
基金financially by the National Research Council of Thailand(NRCT)under Contract No.N42A670894.
摘要This study presents an innovative development of the exponentially weighted moving average(EWMA)control chart,explicitly adapted for the examination of time series data distinguished by seasonal autoregressive moving average behavior—SARMA(1,1)L under exponential white noise.Unlike previous works that rely on simplified models such as AR(1)or assume independence,this research derives for the first time an exact two-sided Average Run Length(ARL)formula for theModified EWMAchart under SARMA(1,1)L conditions,using a mathematically rigorous Fredholm integral approach.The derived formulas are validated against numerical integral equation(NIE)solutions,showing strong agreement and significantly reduced computational burden.Additionally,a performance comparison index(PCI)is introduced to assess the chart’s detection capability.Results demonstrate that the proposed method exhibits superior sensitivity to mean shifts in autocorrelated environments,outperforming existing approaches.The findings offer a new,efficient framework for real-time quality control in complex seasonal processes,with potential applications in environmental monitoring and intelligent manufacturing systems.
摘要L2 reading is not only an important channel for people to obtain information and knowledge,but also the main way for people to learn a foreign language.Reading information processing can be divided into controlled processing and automatic processing.Controlled information processing is a conscious and resource-intensive processing model,while automatic information processing is an unconscious and automatic processing model.This study investigates the characteristics and interactivity of controlled and automatic information processing in L2 reading,and explores the roles of controlled and automatic information processing strategies in improving L2 reading ability.The findings are as follows:(a)controlled and automatic information processing is interactive in L2 reading;and(b)the uses of controlled and automatic information processing strategies are beneficial to the improvement of the reading ability of L2 learners.This study has important theoretical and practical value in improving the efficiency of L2 reading teaching and learning.
基金supported by the Frontier Technologies R&D Program of Jiangsu Province(Grant No.BF2024077)the Key Research and Development Program of Jiangsu Province(Grant No.BE2022069-3).
摘要Directed energy deposition arc welding(DED-arc)encounters challenges such as low forming precision and compromised process stability during the deposition process owing to its intricate molten pool dynamics.The measurement and control of the molten-pool size are essential to addressing these problems.However,most contemporary research focuses on controlling a single variable(e.g.,molten-pool width or height)in either open-or closed-loop configurations.This approach is inherently limited in addressing the dynamic nonlinear characteristics and high-precision requirements of the DED-arc process.To address this challenge,we designed a binocular passive vision sensing system that utilizes an image processing algorithm to facilitate the real-time measurement and size extraction of the molten-pool width and height.A dual fuzzy proportional–integral–derivative controller was also developed to control the width and height of the molten pool by adjusting the arc current and voltage separately.This controller was then compared with open-loop control experiments under constant parameters.The results demonstrate that the proposed bivariate control method can effectively regulate the error range of the actual and detected values of the molten-pool width and height within 0.5 and 0.3 mm,respectively,thereby significantly enhancing the control accuracy and optimizing the forming appearance of thin-walled metal parts.
基金co-supported by the National Natural Science Foundation of China(No.51922098)the National Science and Technology Major Project of China(No.2017-Ⅱ-0004-0017)the Special Fund for the Member of Youth Innovation Promotion Association of CAS(No.2018173).
摘要The adaptive feedback control of stability with circumferential inlet distortion has been experimentally investigated in a low-speed,axial compressor.The flat-baffles with different span heights are used to simulate different distorted inflow cases.Compared with auto-correlation and root-mean-square analysis,cross-correlation analysis used to predict early stall warning does not depend on the distortion position.Hence,the cross-correlation coefficient was used to monitor the stable status of the compressor and provide the feedback signal in the active control strategy when suffering from different distortions.Based on the stall margin improvement of tip air injection obtained under different distorted inflow cases and the sensitivity analysis of cross-correlation coefficients to injected momentum ratios,tip air injection was adopted as the actuator for adaptive feedback control.The digital signal processing controller was designed and applied to achieve adaptive feedback control in distorted inflow conditions.The results show that the adaptive feedback control of air injection nearly achieves the same stall margin improvement as steady air injection under different distortion intensities with a reduced injection mass flow.Thus,the proposed adaptive feedback control method is ideal for the engine operation with circumferential distorted inflow,which frequently occurs in flight.
摘要Fiber quality measurement in spinning preparation is crucial for optimizing waste and meeting yarn quality specifications.The brand-new Uster AFIS 6–the next-generation laboratory instrument from Uster Technologies–uniquely tests man-made fiber properties in addition to cotton.It provides critical data to optimize fiber process control for cotton,man-made fibers,and blended yarns.
基金financially supported by the National Natural Science Foundation of China(Nos.52575504,52175415,52205475,and 92160301).
摘要Traditional tools have limited adaptability in complex machining environments due to their lack of working-condition perception and autonomous regulation.With advances in sensors,materials,and data-processing technologies,tool design is shifting from a single-function‘mechanical arm’for cutting towards integrated intelligent terminals.This paper systematically reviews progress in intelligent tool technology from two perspectives:design and regulation.For intelligent design,the fundamental principles of condition-perception tools equipped with built-in multi-type sensors are discussed,enabling in situ,real-time monitoring of multidimensional parameters such as cutting force,temperature,and vibration.Force monitoring is achieved through elastic deformation or dynamic charge response,temperature monitoring through the thermoelectric effect,and vibration monitoring through micro-displacement and intensity detection.The design focus emphasises sensor miniaturisation and integration,balancing measurement accuracy with tool stiffness while minimising machining interference.In regulation,key technologies for constructing closed-loop control systems(CLCS)are summarised,which dynamically adjust cutting speed,feed rate,and other parameters based on sensed data,achieving precise control of force,temperature,and vibration via feedback mechanisms and driving units.Breakthroughs in tool wear compensation(TWC)mechanisms are introduced.Multi-source signal fusion combined with deep learning algorithms is further examined for improving monitoring accuracy and remaining useful life(RUL)prediction.Through model predictive control,intelligent regulation of cutting parameters within process flows is realised.Finally,challenges such as sensor reliability,multi-source coupling,and balancing cost with industrial applicability are analysed.Future directions highlight novel structural designs,high-performance material development,and multi-technology integration,aiming to establish a fully intelligent machining system through the integrated design of‘perception-decision-execution’.
基金supported by the National Key Research and Development Program of China(Grant Nos.2021YFA1200600,2024YFA1208902,and 2024YFA1408000)the National Natural Science Foundation of China(Grant Nos.52231007,12327804,T2321003,22088101,and 22405050)+1 种基金the Science and Technology Commission of Shanghai Municipality(Grant No.24ZR1406400)Shanghai Municipal Education Commission(Grant No.24KXZNA06)。
摘要Multi-beam scanning electron microscope(MBSEM)reconciles the inherent contradiction between“resolution and throughput”of traditional scanning electron microscopes(SEMs)through parallel electron beam manipulation,emerging as a key technology to address the bottlenecks in large-volume,high-resolution imaging and advanced industrial inspection.This article provides a comprehensive overview of MBSEM,covering its evolutionary course of technology,core design fundamentals,and interdisciplinary applications.First,it sorts out the evolutionary process from conceptualization in the early 21 st century to commercialization in the 2010s,clarifying the core logic of breaking through the physical limitations of single-beam systems via“multi-beam parallelism”.Subsequently,focusing on the core technological chain of“beam generation–optical focusing–signal detection–data processing”,it conducts an in-depth analysis of the design concepts,technical characteristics,and applicable scenarios of three mainstream architectures:the single-source single-column,split optical system,and semiconductor-specific multi-beam inspection(MBI).Combined with representative research and product data from teams such as Delft University of Technology,Zeiss,and ASML/HMI,it reveals the differentiated advantages of each architecture in beam uniformity(the relative deviation percentage of the current density and probe size of each sub-beam in the multi-beam array from the central beam,with a smaller deviation value indicating better uniformity)and signal crosstalk(the percentage of the interference signal intensity to the target signal intensity when the detection signal of a single beam in the multi-beam array interferes with the detection channel of adjacent beams)control,and throughput(the percentage of the interference signal intensity to the target signal intensity when the detection signal of a single beam in the multi-beam array interferes with the detection channel of adjacent beams)improvement.Finally,integrating the practical demands of neuroscience connectomics,advanced semiconductor manufacturing processes,and biomedicine,it elaborates on the application breakthroughs of MBSEM in the three-dimensional(3D)reconstruction of large-volume brain tissue,wafer defect screening for 7 nm and smaller nodes,and low-damage imaging of thin biological tissues,and compares its disruptive value relative to traditional technologies(single-beam SEM,optical inspection).Unlike previous reviews,this work systematically integrates both academic prototypes and industrial systems for the first time,providing an in-depth analysis of the differentiated trade-offs among imaging speed,resolution,and sample adaptability across different architectures,offering a systematic reference for MBSEM R&D and interdisciplinary applications.
基金supported by the Xinjiang Uygur Autonomous Region Key R&D Project,China(2023B01031-3)the National Natural Science Foundation of China(62422303 and 62373035).
摘要Ethylene yield serves as a key metric in petrochemical production,where optimizing its energy efficiency remains a critical challenge for sustainable production.Meanwhile,manual hyperparameter tuning of deep learning based yield prediction models often results in suboptimal configurations,which reduces prediction accuracy and reliability due to extreme operating conditions generating outliers in ethylene production processes.Therefore,a novel neural network automatic design method(NNADM)is proposed,which incorporates the neural network parameters automatic optimization and loss function adaptive construction.An innovative adaptive loss formulation is proposed to strategically integrate the complementary strengths of the mean squared error(MSE)and the Log-Cosh functions,featuring dynamic outlier resistance through self-adjusting weight coefficients.Then,the Bayesian optimization search algorithm is utilized to discover optimal hyperparameters of the neural network,including hidden layer unit,epoch,batch size,and the loss function.Finally,the NNADM is integrated with several classical neural networks for ethylene yield prediction.Experimental results show that several classical neural networks realize an average of 16.86%decrease in MSE index after integrating the NNADM.In addition,the proposed model offers direction and development blueprints for ethylene production facilities that have low energy efficiency.By implementing this model,it is possible to cut approximately 8376.4 tons of carbon emissions and simultaneously secure an extra 499 tons of ethylene output.
基金supported by the National Natural Science Foundation of China(Grant Nos.72471142 and 72101147).
摘要Recent advances in artificial intelligence(AI)have significantly enhanced quality management,enabling more effective handling of complex,high-dimensional,and multi-modal data.AI methods,including machine learning(ML)and deep learning(DL),have been pivotal in advancing key areas such as quality optimization,monitoring,and diagnosis.These methods have increased adaptability,efficiency,and scalability,making them particularly suitable for modern industrial applications.This review provides a comprehensive examination of AI methods in quality management,covering the integration of surrogate models,Bayesian optimization(BO),intelligent control charts,change-point detection(CPD),and interpretable quality diagnosis.The review concludes with proposed directions for future research aimed at overcoming existing challenges and enhancing the deployment of AI in real-world quality management implementation.