The paper combines a self-adaptive precise algorithm in the time domain with Meshless Element Free Galerkin Method (EFGM) for solving viscoelastic problems with rotationally periodic symmetry. By expanding variables...The paper combines a self-adaptive precise algorithm in the time domain with Meshless Element Free Galerkin Method (EFGM) for solving viscoelastic problems with rotationally periodic symmetry. By expanding variables at a discretized time interval, the variations of variables can be described more precisely, and iteration is not required for non-linear cases. A space-time domain coupled problem with initial and boundary values can be converted into a series of linear recursive boundary value problems, which are solved by a group theory based on EFGM. It has been proved that the coefficient matrix of the global EFG equation for a rotationally periodic system is block-circulant so long as a kind of symmetry-adapted reference coordinate system is adopted, and then a partitioning algorithm for facilitating parallel processing was proposed via a completely orthogonal group transformation. Therefore instead of solving the original system, only a series of independent small sub-problems need to be solved, leading to computational convenience and a higher computing efficiency. Numerical examples are given to illustrate the full advantages of the proposed algorithm.展开更多
By modeling direct transient heat conduction problems via finite element method (FEM) and precise integral algorithm, a new approach is presented to solve transient inverse heat conduction problems with multi-variable...By modeling direct transient heat conduction problems via finite element method (FEM) and precise integral algorithm, a new approach is presented to solve transient inverse heat conduction problems with multi-variables. Firstly, the spatial space and temporal domain are discretized by FEM and precise integral algorithm respectively. Then, the high accuracy semi-analytical solution of direct problem can be got. Finally, based on the solution, the computing model of inverse problem and expression of sensitivity analysis are established. Single variable and variables combined identifications including thermal parameters, boundary conditions and source-related terms etc. are given to validate the approach proposed in 1-D and 2-D cases. The effects of noise data and initial guess on the results are investigated. The numerical examples show the effectiveness of this approach.展开更多
In this paper,we present a phase multiplication algorithm(PMA)to obtain scalable fringe precision in laser self-mixing interferometer under a weak feedback regime.Merely by applying the double angle formula on the sel...In this paper,we present a phase multiplication algorithm(PMA)to obtain scalable fringe precision in laser self-mixing interferometer under a weak feedback regime.Merely by applying the double angle formula on the self-mixing signal multiple times,the continuously improved fringe precision will be obtained.Theoretical analysis shows that the precision of the fringe could be improved toλ/2n+1.The validity of the proposed method is demonstrated by means of simulated SMI signals and confirmed by experiments under different amplitudes.A fringe precision ofλ/128 at a sampling rate of 500 k S/s has been achieved after doing 6 th the PMA.Finally,an amplitude of 50 nm has been proved to be measurable and the absolute error is 3.07 nm,which is within the theoretical error range.The proposed method for vibration measurement has the advantage of high accuracy and reliable without adding any additional optical elements in the optical path,thus it will play an important role in nanoscale measurement field.展开更多
In view of the TIN_DDM buffer surface existing in the construction and application of special data type,algorithm efficiency and precision are not matching;the paper applied the rolling ball model in the process of TI...In view of the TIN_DDM buffer surface existing in the construction and application of special data type,algorithm efficiency and precision are not matching;the paper applied the rolling ball model in the process of TIN_DDM buffer surface construction.Based on the precision limitation analysis of rolling ball model,the overall precision control method of rolling ball model has been established.Considering the efficiency requirement of TIN_DDM buffer surface construction,the influence principle of key sampling points and rolling ball radius to TIN_DDM buffer surface construction efficiency has been elaborated,and the rule of identifying key sampling points has also been designed.Afterwards,by erecting the numerical relationship between key sampling points and rolling ball radius,a TIN_DDM buffer surface construction algorithm based on rolling ball acceleration optimization model has been brought forward.The time complexity of the algorithm is O(n).The experiments show that the algorithm could realize the TIN_DDM buffer surface construction with high efficiency,and the algorithm precision is controlled with in 2σ.展开更多
Overload and oversize transport not only causes serious damage to highway,but also brings about irreparable traffic accidents.WIM technology can control the overload and oversize of vehicles and provide a reliable bas...Overload and oversize transport not only causes serious damage to highway,but also brings about irreparable traffic accidents.WIM technology can control the overload and oversize of vehicles and provide a reliable basis for the heavy truck tol collection.The main existing WIM equipment includes bending-plate(weighing part wheel load),single-scale-platform(weighing load of single-axle),double-scale-platform(weighing load of single-axle),axis-group-scale(weighing weight of axisgroup),and vehicle-scale(weighing total weight of vehicle).They have many advantages and disadvantages in cost,installation,maintenance,traffic efficiency and accuracy of WIM.The WIM accuracy is one of the key indicators considered in application of technology governance,toll by weight,and traffic investigation.The carrier length of WIM system increases gradually with the increase of WIM accuracy,which leads to the construction cost and time increase also.In order to ensure the WIM accuracy,considering the economy of the carrier length,based on the vibration characteristics of the vehicle and the scale,the mathematical model of the WIM signal is established by means of moving average filtering and B spline least square method,and the vibration disturbance of low frequency dynamic load is effectively eliminated.The test result shows that(1)the weight erron is less than 0.5%if test vehicles pass through the 0.8 m and 1.6 m long scales in the speed range of 0-5 km/h and 0-12km/h,which meets the national standard about Class 1 scale;(2)if test vehicles pass through the 2.6 m and 3 m long scales with the theoretical limit speed of 20.7 km/h and 24.3 km/h,the statistical weight error is less than 0.5%.The research not only provided a theoretical basis for design the length of the scale,but also has important reference value for improvement of the WIM accuracy of the existing installed single scale and double scales.展开更多
To solve the different time delays that exist in the control device installed on spatial structures, in this study, discrete analysis using a 2N precise algorithm was selected to solve the multi-time-delay issue for l...To solve the different time delays that exist in the control device installed on spatial structures, in this study, discrete analysis using a 2N precise algorithm was selected to solve the multi-time-delay issue for long-span structures based on the market-based control (MBC) method. The concept of interval mixed energy was introduced from computational structural mechanics and optimal control research areas, and it translates the design of the MBC multi-time-delay controller into a solution for the segment matrix. This approach transforms the serial algorithm in time to parallel computing in space, greatly improving the solving efficiency and numerical stability. The designed controller is able to consider the issue of time delay with a linear controlling force combination and is especially effective for large time-delay conditions. A numerical example of a long-span structure was selected to demonstrate the effectiveness of the presented controller, and the time delay was found to have a significant impact on the results.展开更多
基金The project supported by the National Natural Science Foundation of China (10421002. 10472019 and 10172024) NKBRSF (2005CB321704) and the Fund of Disciplines Leaders of Young and Middle Age Faculty in Colleges of Liaoning Province. The English text was polished by Yunming Chen.
摘要The paper combines a self-adaptive precise algorithm in the time domain with Meshless Element Free Galerkin Method (EFGM) for solving viscoelastic problems with rotationally periodic symmetry. By expanding variables at a discretized time interval, the variations of variables can be described more precisely, and iteration is not required for non-linear cases. A space-time domain coupled problem with initial and boundary values can be converted into a series of linear recursive boundary value problems, which are solved by a group theory based on EFGM. It has been proved that the coefficient matrix of the global EFG equation for a rotationally periodic system is block-circulant so long as a kind of symmetry-adapted reference coordinate system is adopted, and then a partitioning algorithm for facilitating parallel processing was proposed via a completely orthogonal group transformation. Therefore instead of solving the original system, only a series of independent small sub-problems need to be solved, leading to computational convenience and a higher computing efficiency. Numerical examples are given to illustrate the full advantages of the proposed algorithm.
摘要By modeling direct transient heat conduction problems via finite element method (FEM) and precise integral algorithm, a new approach is presented to solve transient inverse heat conduction problems with multi-variables. Firstly, the spatial space and temporal domain are discretized by FEM and precise integral algorithm respectively. Then, the high accuracy semi-analytical solution of direct problem can be got. Finally, based on the solution, the computing model of inverse problem and expression of sensitivity analysis are established. Single variable and variables combined identifications including thermal parameters, boundary conditions and source-related terms etc. are given to validate the approach proposed in 1-D and 2-D cases. The effects of noise data and initial guess on the results are investigated. The numerical examples show the effectiveness of this approach.
基金supported by the Natural Science Foundation of Fujian Province(No.2020J01705)the School Foundation of Jimei University(No.C150345)。
摘要In this paper,we present a phase multiplication algorithm(PMA)to obtain scalable fringe precision in laser self-mixing interferometer under a weak feedback regime.Merely by applying the double angle formula on the self-mixing signal multiple times,the continuously improved fringe precision will be obtained.Theoretical analysis shows that the precision of the fringe could be improved toλ/2n+1.The validity of the proposed method is demonstrated by means of simulated SMI signals and confirmed by experiments under different amplitudes.A fringe precision ofλ/128 at a sampling rate of 500 k S/s has been achieved after doing 6 th the PMA.Finally,an amplitude of 50 nm has been proved to be measurable and the absolute error is 3.07 nm,which is within the theoretical error range.The proposed method for vibration measurement has the advantage of high accuracy and reliable without adding any additional optical elements in the optical path,thus it will play an important role in nanoscale measurement field.
基金National Natural Science Foundation of China(Nos.41601498,41471380)National Key R&D Program of China(No.2017YFC1405505)。
摘要In view of the TIN_DDM buffer surface existing in the construction and application of special data type,algorithm efficiency and precision are not matching;the paper applied the rolling ball model in the process of TIN_DDM buffer surface construction.Based on the precision limitation analysis of rolling ball model,the overall precision control method of rolling ball model has been established.Considering the efficiency requirement of TIN_DDM buffer surface construction,the influence principle of key sampling points and rolling ball radius to TIN_DDM buffer surface construction efficiency has been elaborated,and the rule of identifying key sampling points has also been designed.Afterwards,by erecting the numerical relationship between key sampling points and rolling ball radius,a TIN_DDM buffer surface construction algorithm based on rolling ball acceleration optimization model has been brought forward.The time complexity of the algorithm is O(n).The experiments show that the algorithm could realize the TIN_DDM buffer surface construction with high efficiency,and the algorithm precision is controlled with in 2σ.
基金Supported by Special Project of Science and Technology Innovation for Social and Livelihood in Chongqing(No.cstc2016shmszx30018)Special Project of Science and Technology Innovation for Social Undertakings and People's Livelihood Guarantee in Chongqing(No.cstc2015shms-ztzx30016)。
摘要Overload and oversize transport not only causes serious damage to highway,but also brings about irreparable traffic accidents.WIM technology can control the overload and oversize of vehicles and provide a reliable basis for the heavy truck tol collection.The main existing WIM equipment includes bending-plate(weighing part wheel load),single-scale-platform(weighing load of single-axle),double-scale-platform(weighing load of single-axle),axis-group-scale(weighing weight of axisgroup),and vehicle-scale(weighing total weight of vehicle).They have many advantages and disadvantages in cost,installation,maintenance,traffic efficiency and accuracy of WIM.The WIM accuracy is one of the key indicators considered in application of technology governance,toll by weight,and traffic investigation.The carrier length of WIM system increases gradually with the increase of WIM accuracy,which leads to the construction cost and time increase also.In order to ensure the WIM accuracy,considering the economy of the carrier length,based on the vibration characteristics of the vehicle and the scale,the mathematical model of the WIM signal is established by means of moving average filtering and B spline least square method,and the vibration disturbance of low frequency dynamic load is effectively eliminated.The test result shows that(1)the weight erron is less than 0.5%if test vehicles pass through the 0.8 m and 1.6 m long scales in the speed range of 0-5 km/h and 0-12km/h,which meets the national standard about Class 1 scale;(2)if test vehicles pass through the 2.6 m and 3 m long scales with the theoretical limit speed of 20.7 km/h and 24.3 km/h,the statistical weight error is less than 0.5%.The research not only provided a theoretical basis for design the length of the scale,but also has important reference value for improvement of the WIM accuracy of the existing installed single scale and double scales.
基金provided by the Science Fund for Creative Research Groups of the National Natural Science Foundation of China under Grant Nos.51261120375 and 51421064
摘要To solve the different time delays that exist in the control device installed on spatial structures, in this study, discrete analysis using a 2N precise algorithm was selected to solve the multi-time-delay issue for long-span structures based on the market-based control (MBC) method. The concept of interval mixed energy was introduced from computational structural mechanics and optimal control research areas, and it translates the design of the MBC multi-time-delay controller into a solution for the segment matrix. This approach transforms the serial algorithm in time to parallel computing in space, greatly improving the solving efficiency and numerical stability. The designed controller is able to consider the issue of time delay with a linear controlling force combination and is especially effective for large time-delay conditions. A numerical example of a long-span structure was selected to demonstrate the effectiveness of the presented controller, and the time delay was found to have a significant impact on the results.