In this study, for the purpose of improving the efficiency and accuracy of numerical simulation of massive concrete, the symmetric successive over relaxation-preconditioned conjugate gradient method （SSOR-PCGM） with...In this study, for the purpose of improving the efficiency and accuracy of numerical simulation of massive concrete, the symmetric successive over relaxation-preconditioned conjugate gradient method （SSOR-PCGM） with an improved iteration format was derived and applied to solution of large sparse symmetric positive definite linear equations in the computational process of the finite element analysis. A three-dimensional simulation program for massive concrete was developed based on SSOR-PCGM with an improved iteration format. Then, the programs based on the direct method and SSOR-PCGM with an improved iteration format were used for computation of the Guandi roller compacted concrete （RCC） gravity dam and an elastic cube under free expansion. The comparison and analysis of the computational results show that SSOR-PCGM with the improved iteration format occupies much less physical memory and can solve larger-scale problems with much less computing time and flexible control of accuracy.展开更多
This research paper deals with the boundary and initial value problems for the Bratu-type model by using the New Improved Variational Homotopy Perturbation Method. The New Method does not require discritization, linea...This research paper deals with the boundary and initial value problems for the Bratu-type model by using the New Improved Variational Homotopy Perturbation Method. The New Method does not require discritization, linearization or any restrictive assumption of any form in providing analytical or approximate solutions to linear and nonlinear equation without the integral related with nonlinear term. Theses virtues make it to be reliable and its efficiency is demonstrated with numerical examples.展开更多
Although deep learning methods have been widely applied in slam visual odometry(VO)over the past decade with impressive improvements,the accuracy remains limited in complex dynamic environments.In this paper,a composi...Although deep learning methods have been widely applied in slam visual odometry(VO)over the past decade with impressive improvements,the accuracy remains limited in complex dynamic environments.In this paper,a composite mask-based generative adversarial network(CMGAN)is introduced to predict camera motion and binocular depth maps.Specifically,a perceptual generator is constructed to obtain the corresponding parallax map and optical flow between two neighboring frames.Then,an iterative pose improvement strategy is proposed to improve the accuracy of pose estimation.Finally,a composite mask is embedded in the discriminator to sense structural deformation in the synthesized virtual image,thereby increasing the overall structural constraints of the network model,improving the accuracy of camera pose estimation,and reducing drift issues in the VO.Detailed quantitative and qualitative evaluations on the KITTI dataset show that the proposed framework outperforms existing conventional,supervised learning and unsupervised depth VO methods,providing better results in both pose estimation and depth estimation.展开更多
基金supported by the National Natural Science Foundation of China (Grant No.50808066)
文摘In this study, for the purpose of improving the efficiency and accuracy of numerical simulation of massive concrete, the symmetric successive over relaxation-preconditioned conjugate gradient method （SSOR-PCGM） with an improved iteration format was derived and applied to solution of large sparse symmetric positive definite linear equations in the computational process of the finite element analysis. A three-dimensional simulation program for massive concrete was developed based on SSOR-PCGM with an improved iteration format. Then, the programs based on the direct method and SSOR-PCGM with an improved iteration format were used for computation of the Guandi roller compacted concrete （RCC） gravity dam and an elastic cube under free expansion. The comparison and analysis of the computational results show that SSOR-PCGM with the improved iteration format occupies much less physical memory and can solve larger-scale problems with much less computing time and flexible control of accuracy.
文摘This research paper deals with the boundary and initial value problems for the Bratu-type model by using the New Improved Variational Homotopy Perturbation Method. The New Method does not require discritization, linearization or any restrictive assumption of any form in providing analytical or approximate solutions to linear and nonlinear equation without the integral related with nonlinear term. Theses virtues make it to be reliable and its efficiency is demonstrated with numerical examples.
基金supported by the Program of Graduate Education and Teaching Reform in Tianjin University of Technology(Nos.YBXM2204 and ZDXM2202)the National Natural Science Foundation of China(Nos.62203331 and 62103299)。
文摘Although deep learning methods have been widely applied in slam visual odometry(VO)over the past decade with impressive improvements,the accuracy remains limited in complex dynamic environments.In this paper,a composite mask-based generative adversarial network(CMGAN)is introduced to predict camera motion and binocular depth maps.Specifically,a perceptual generator is constructed to obtain the corresponding parallax map and optical flow between two neighboring frames.Then,an iterative pose improvement strategy is proposed to improve the accuracy of pose estimation.Finally,a composite mask is embedded in the discriminator to sense structural deformation in the synthesized virtual image,thereby increasing the overall structural constraints of the network model,improving the accuracy of camera pose estimation,and reducing drift issues in the VO.Detailed quantitative and qualitative evaluations on the KITTI dataset show that the proposed framework outperforms existing conventional,supervised learning and unsupervised depth VO methods,providing better results in both pose estimation and depth estimation.
