The Extreme Learning Machine（ELM） and its variants are effective in many machine learning applications such as Imbalanced Learning（IL） or Big Data（BD） learning. However, they are unable to solve both imbalanced ...The Extreme Learning Machine（ELM） and its variants are effective in many machine learning applications such as Imbalanced Learning（IL） or Big Data（BD） learning. However, they are unable to solve both imbalanced and large-volume data learning problems. This study addresses the IL problem in BD applications. The Distributed and Weighted ELM（DW-ELM） algorithm is proposed, which is based on the Map Reduce framework. To confirm the feasibility of parallel computation, first, the fact that matrix multiplication operators are decomposable is illustrated.Then, to further improve the computational efficiency, an Improved DW-ELM algorithm（IDW-ELM） is developed using only one Map Reduce job. The successful operations of the proposed DW-ELM and IDW-ELM algorithms are finally validated through experiments.展开更多
A new type of digital video monitoring system (DVMS) named user defined target tracking system (UDTTS), was developed based on the digital image processing (DIP) technology and the practice demands of construction sit...A new type of digital video monitoring system (DVMS) named user defined target tracking system (UDTTS), was developed based on the digital image processing (DIP) technology and the practice demands of construction site management in hydraulic engineering. The position, speed, and track of moving targets such as humans and vehicles, which could be calculated by their locations at anytime in images basically, were required for management. The proposed algorithm, dependent on the context-sensitive moving information of image sequences which was much more than one or two images provided, compared the blobs’ properties in current frame to the trajectories of targets in the previous frames and then corresponded them. The processing frame rate is about 10fps with the image 240-by-120 pixels. Experimental results show that position, direction, and speed measurements have an accuracy level compatible with the manual work. The user-define process makes the UDTTS available to the public whenever appropriate.展开更多
In this study,a multi-physics and multi-scale coupling program,Fluent/KMC-sub/NDK,was developed based on the user-defined functions(UDF)of Fluent,in which the KMC-sub-code is a sub-channel thermal-hydraulic code and t...In this study,a multi-physics and multi-scale coupling program,Fluent/KMC-sub/NDK,was developed based on the user-defined functions(UDF)of Fluent,in which the KMC-sub-code is a sub-channel thermal-hydraulic code and the NDK code is a neutron diffusion code.The coupling program framework adopts the"master-slave"mode,in which Fluent is the master program while NDK and KMC-sub are coupled internally and compiled into the dynamic link library(DLL)as slave codes.The domain decomposition method was adopted,in which the reactor core was simulated by NDK and KMC-sub,while the rest of the primary loop was simulated using Fluent.A simulation of the reactor shutdown process of M2LFR-1000 was carried out using the coupling program,and the code-to-code verification was performed with ATHLET,demonstrating a good agreement,with absolute deviation was smaller than 0.2%.The results show an obvious thermal stratification phenomenon during the shutdown process,which occurs 10 s after shutdown,and the change in thermal stratification phenomena is also captured by the coupling program.At the same time,the change in the neutron flux density distribution of the reactor was also obtained.展开更多
The seismic response characteristics of underground structures in saturated soils are investigated.A fully fluid-solid coupling dynamic model is developed and implemented into ABAQUS with a user-defined element to sim...The seismic response characteristics of underground structures in saturated soils are investigated.A fully fluid-solid coupling dynamic model is developed and implemented into ABAQUS with a user-defined element to simulate the dynamic behavior of saturated soils.The accuracy of the model is validated using a classic example in literature.The performance of the model is verified by its application on simulating the seismic response characteristics of a subway station built in saturated soils.The merits of the model are demonstrated by comparing the difference of the seismic response of an underground structure in saturated soils between using the fully coupling model and a single-phase medium model.The study finds that the fully coupling model developed herein can simulate the dynamic response characteristics of the underground structures in saturated soils with high accuracy.The seismic response of the underground structure tends to be underestimated by using the single-phase medium model compared with using the fully coupling model,which provides a weaker confining action to the underground structure.展开更多
The smelting reduction process of the ilmenite in an electric arc furnace(EAF)is a commonly used technology for producing titanium slag in the world.It has particular significance to analyze the velocity-temperature-e...The smelting reduction process of the ilmenite in an electric arc furnace(EAF)is a commonly used technology for producing titanium slag in the world.It has particular significance to analyze the velocity-temperature-electromagnetics multi-physical field in an EAF for improving its productivity and reducing energy consumption.A transient three-dimensional mathematical model was developed to characterize the flow,heat transfer,and electromagnetic behavior in a titanium slag EAF.For describing the electromagnetic field and its effects on velocity and temperature distribution in the furnace,magnetohydrodynamic equations and conservation equations for mass,momentum,and energy were solved simultaneously by compiling the user-defined function program.The numerical model was verified by comparing with the literature data.The results indicate that the Lorentz force is the main driving force of the velocity and temperature distribution.Moreover,the influence of input current and location of electrodes on the multi-physical field distribution was also investigated.It is found that the appropriate range of input current and diameter of pitch circle are about 30,000 A and 3000-3500 mm,respectively.The mathematical model established can characterize the multi-physical field more accu-rately than before,which can provide valuable guidance for the operation improvement and design optimization of the EAF for producing titanium slag.展开更多
The influence of the structural features of dragonfly wings, including the sandwich-type configuration of longitudinal veins and the longitudinal corrugations, on the impact response of a bio-inspired structure is inv...The influence of the structural features of dragonfly wings, including the sandwich-type configuration of longitudinal veins and the longitudinal corrugations, on the impact response of a bio-inspired structure is investigated. According to experimental observations of the wing morphology, a novel foam-based composite structure is introduced consisting of E-glass/epoxy face-sheets bonded to a polyurethane foam core. A finite element model is employed to simulate the structural responses of the biomimetic structure under low velocity impact. The initiation and evolution of the impact-induced damage in composite skins are simulated by applying a user-defined progressive damage model together with the interracial cohesive law for intra- and inter-laminar damages, respectively. To simulate the nonlinear behavior of the foam core, a crushable plasticity model is implemented. The numerically obtained results are found to correlate with the experimentally measured ones, acquired by drop-weight testing on a bio-inspired structure. It is numerically predicted that reinforcing the structure with the veins gives the more impact load-bearing capacity and the longitudinal corrugation can increase the stiffness and damage resistance of the structure. Effects of the change in impact location, the configuration of the veins and the corrugated angle on damage resistance of the structures are fully discussed.展开更多
This work aims to investigate local stress distribution, damage evolution and failure of notched composite laminates under in-plane loads. An analytic method containing uniformed boundary equations using a complex var...This work aims to investigate local stress distribution, damage evolution and failure of notched composite laminates under in-plane loads. An analytic method containing uniformed boundary equations using a complex variable approach is developed to present layer-by-layer stresses around the notch. The uniformed boundary equations established in series together with conformal mapping functions are capable of dealing with irregular boundary issues around the notch and at infinity. Stress results are employed to evaluate the damage initiation and propagation of notched composites by progressive damage analysis（PDA）. A user-defined subroutine is developed in the finite element（FE） model based on coupling theories for mixed failure criteria and damage mechanics to efficiently investigate damage evolution as well as failure modes. Carbon/epoxy laminates with a stacking sequence of [45°/0°/ 60°/90°]sare used to investigate surface strains, in-plane load capacity and microstructure of failure zones to provide analytic and FE methods with strong validation. Good agreement is observed between the analytic method, the FE model and experiments in terms of the stress（strain） distributions, damage evaluation and ultimate strength, and the layerby-layer stress components vary according to a combination effect of fiber orientation and loading type, causing diverse failure modes in individuals.展开更多
To accelerate the training of support vector domain description （SVDD）, confidence support vector domain description （CSVDD） is proposed based on the observation that the description boundary is determined by a sm...To accelerate the training of support vector domain description （SVDD）, confidence support vector domain description （CSVDD） is proposed based on the observation that the description boundary is determined by a small subset of training data called support vectors. Namely, the number of training samples in the userdefined sphere is calculated and taken as the confidence measure, according to which the training samples are ranked in ascending order. Those former ranked ones are selected as the boundary targets for the SVDD training. Simulations on UCI data demonstrate the effectiveness and superiority of CSVDD： the number of training targets and the training time are reduced without any loss of accuracy.展开更多
Various code development platforms, such as the ATHENA Framework [1] of the ATLAS [2] experiment encounter lengthy compilation/linking times. To augment this situation, the IRIS Development Platform was built as a sof...Various code development platforms, such as the ATHENA Framework [1] of the ATLAS [2] experiment encounter lengthy compilation/linking times. To augment this situation, the IRIS Development Platform was built as a software development framework acting as compiler, cross-project linker and data fetcher, which allow hot-swaps in order to compare various versions of software under test. The flexibility fostered by IRIS allowed modular exchange of software libraries among developers, making it a powerful development tool. The IRIS platform used input data ROOT-ntuples [3];however a new data model is sought, in line with the facilities offered by IRIS. The schematic of a possible new data structuring—as a user implemented object oriented data base, is presented.展开更多
基金partially supported by the National Natural Science Foundation of China(Nos.61402089,61472069,and 61501101)the Fundamental Research Funds for the Central Universities(Nos.N161904001,N161602003,and N150408001)+2 种基金the Natural Science Foundation of Liaoning Province(No.2015020553)the China Postdoctoral Science Foundation(No.2016M591447)the Postdoctoral Science Foundation of Northeastern University(No.20160203)
文摘The Extreme Learning Machine（ELM） and its variants are effective in many machine learning applications such as Imbalanced Learning（IL） or Big Data（BD） learning. However, they are unable to solve both imbalanced and large-volume data learning problems. This study addresses the IL problem in BD applications. The Distributed and Weighted ELM（DW-ELM） algorithm is proposed, which is based on the Map Reduce framework. To confirm the feasibility of parallel computation, first, the fact that matrix multiplication operators are decomposable is illustrated.Then, to further improve the computational efficiency, an Improved DW-ELM algorithm（IDW-ELM） is developed using only one Map Reduce job. The successful operations of the proposed DW-ELM and IDW-ELM algorithms are finally validated through experiments.
文摘A new type of digital video monitoring system (DVMS) named user defined target tracking system (UDTTS), was developed based on the digital image processing (DIP) technology and the practice demands of construction site management in hydraulic engineering. The position, speed, and track of moving targets such as humans and vehicles, which could be calculated by their locations at anytime in images basically, were required for management. The proposed algorithm, dependent on the context-sensitive moving information of image sequences which was much more than one or two images provided, compared the blobs’ properties in current frame to the trajectories of targets in the previous frames and then corresponded them. The processing frame rate is about 10fps with the image 240-by-120 pixels. Experimental results show that position, direction, and speed measurements have an accuracy level compatible with the manual work. The user-define process makes the UDTTS available to the public whenever appropriate.
基金supported by Science and Technology on Reactor System Design Technology Laboratory,Chengdu,China(LRSDT2020106)
文摘In this study,a multi-physics and multi-scale coupling program,Fluent/KMC-sub/NDK,was developed based on the user-defined functions(UDF)of Fluent,in which the KMC-sub-code is a sub-channel thermal-hydraulic code and the NDK code is a neutron diffusion code.The coupling program framework adopts the"master-slave"mode,in which Fluent is the master program while NDK and KMC-sub are coupled internally and compiled into the dynamic link library(DLL)as slave codes.The domain decomposition method was adopted,in which the reactor core was simulated by NDK and KMC-sub,while the rest of the primary loop was simulated using Fluent.A simulation of the reactor shutdown process of M2LFR-1000 was carried out using the coupling program,and the code-to-code verification was performed with ATHLET,demonstrating a good agreement,with absolute deviation was smaller than 0.2%.The results show an obvious thermal stratification phenomenon during the shutdown process,which occurs 10 s after shutdown,and the change in thermal stratification phenomena is also captured by the coupling program.At the same time,the change in the neutron flux density distribution of the reactor was also obtained.
基金National Natural Science Foundation of People’s Republic of China under Grant Nos.51178011 and 51778386the Key Fundamental Study Development Project of People’s Republic of China under Grant No.2011CB013602。
文摘The seismic response characteristics of underground structures in saturated soils are investigated.A fully fluid-solid coupling dynamic model is developed and implemented into ABAQUS with a user-defined element to simulate the dynamic behavior of saturated soils.The accuracy of the model is validated using a classic example in literature.The performance of the model is verified by its application on simulating the seismic response characteristics of a subway station built in saturated soils.The merits of the model are demonstrated by comparing the difference of the seismic response of an underground structure in saturated soils between using the fully coupling model and a single-phase medium model.The study finds that the fully coupling model developed herein can simulate the dynamic response characteristics of the underground structures in saturated soils with high accuracy.The seismic response of the underground structure tends to be underestimated by using the single-phase medium model compared with using the fully coupling model,which provides a weaker confining action to the underground structure.
基金supported by the National Natural Science Foundation of China(No.U2003215).
文摘The smelting reduction process of the ilmenite in an electric arc furnace(EAF)is a commonly used technology for producing titanium slag in the world.It has particular significance to analyze the velocity-temperature-electromagnetics multi-physical field in an EAF for improving its productivity and reducing energy consumption.A transient three-dimensional mathematical model was developed to characterize the flow,heat transfer,and electromagnetic behavior in a titanium slag EAF.For describing the electromagnetic field and its effects on velocity and temperature distribution in the furnace,magnetohydrodynamic equations and conservation equations for mass,momentum,and energy were solved simultaneously by compiling the user-defined function program.The numerical model was verified by comparing with the literature data.The results indicate that the Lorentz force is the main driving force of the velocity and temperature distribution.Moreover,the influence of input current and location of electrodes on the multi-physical field distribution was also investigated.It is found that the appropriate range of input current and diameter of pitch circle are about 30,000 A and 3000-3500 mm,respectively.The mathematical model established can characterize the multi-physical field more accu-rately than before,which can provide valuable guidance for the operation improvement and design optimization of the EAF for producing titanium slag.
文摘The influence of the structural features of dragonfly wings, including the sandwich-type configuration of longitudinal veins and the longitudinal corrugations, on the impact response of a bio-inspired structure is investigated. According to experimental observations of the wing morphology, a novel foam-based composite structure is introduced consisting of E-glass/epoxy face-sheets bonded to a polyurethane foam core. A finite element model is employed to simulate the structural responses of the biomimetic structure under low velocity impact. The initiation and evolution of the impact-induced damage in composite skins are simulated by applying a user-defined progressive damage model together with the interracial cohesive law for intra- and inter-laminar damages, respectively. To simulate the nonlinear behavior of the foam core, a crushable plasticity model is implemented. The numerically obtained results are found to correlate with the experimentally measured ones, acquired by drop-weight testing on a bio-inspired structure. It is numerically predicted that reinforcing the structure with the veins gives the more impact load-bearing capacity and the longitudinal corrugation can increase the stiffness and damage resistance of the structure. Effects of the change in impact location, the configuration of the veins and the corrugated angle on damage resistance of the structures are fully discussed.
基金sponsored by the National Natural Science Foundation of China, with three different programs (No. 51275410, No. 51305349 and No. 51305352) that supports the present work financially
文摘This work aims to investigate local stress distribution, damage evolution and failure of notched composite laminates under in-plane loads. An analytic method containing uniformed boundary equations using a complex variable approach is developed to present layer-by-layer stresses around the notch. The uniformed boundary equations established in series together with conformal mapping functions are capable of dealing with irregular boundary issues around the notch and at infinity. Stress results are employed to evaluate the damage initiation and propagation of notched composites by progressive damage analysis（PDA）. A user-defined subroutine is developed in the finite element（FE） model based on coupling theories for mixed failure criteria and damage mechanics to efficiently investigate damage evolution as well as failure modes. Carbon/epoxy laminates with a stacking sequence of [45°/0°/ 60°/90°]sare used to investigate surface strains, in-plane load capacity and microstructure of failure zones to provide analytic and FE methods with strong validation. Good agreement is observed between the analytic method, the FE model and experiments in terms of the stress（strain） distributions, damage evaluation and ultimate strength, and the layerby-layer stress components vary according to a combination effect of fiber orientation and loading type, causing diverse failure modes in individuals.
基金supported by the National Natural Science Foundation of China(6057407560674108).
文摘To accelerate the training of support vector domain description （SVDD）, confidence support vector domain description （CSVDD） is proposed based on the observation that the description boundary is determined by a small subset of training data called support vectors. Namely, the number of training samples in the userdefined sphere is calculated and taken as the confidence measure, according to which the training samples are ranked in ascending order. Those former ranked ones are selected as the boundary targets for the SVDD training. Simulations on UCI data demonstrate the effectiveness and superiority of CSVDD： the number of training targets and the training time are reduced without any loss of accuracy.
文摘Various code development platforms, such as the ATHENA Framework [1] of the ATLAS [2] experiment encounter lengthy compilation/linking times. To augment this situation, the IRIS Development Platform was built as a software development framework acting as compiler, cross-project linker and data fetcher, which allow hot-swaps in order to compare various versions of software under test. The flexibility fostered by IRIS allowed modular exchange of software libraries among developers, making it a powerful development tool. The IRIS platform used input data ROOT-ntuples [3];however a new data model is sought, in line with the facilities offered by IRIS. The schematic of a possible new data structuring—as a user implemented object oriented data base, is presented.
