• Journal of applied mathematics & informatics
• /
• v.3 no.2
• /
• pp.217-234
• /
• 1996

This report introduces a faster parallel LU decomposi-tion algorithm that gives a speedup almost equal to the number of nodes used. The new algorithm takes an advantage of an important C feature that lays out a matrix using a row major scheme and is based on the currently widely used LU decomposition algorithm with one major modification to eliminate most of the communication overhead. Empirical results are included in this report. For example solving a dense matrix that contains 100,000,000 elements gives a speedup of 50 when executed on 50 nodes of an intel Paragon in parallel.

• Proceedings of the KIEE Conference
• /
• 2000.07a
• /
• pp.274-275
• /
• 2000

This paper introduces new ordering algorithms using the graph of data structure and forward/backward substitution of LU decomposition using recursive function. The performance of the algorithm is compared with Tinney's algorithm using 14 bus systems. Test results show that the new fill-in element of Jacobian matrix using the proposed ordering algorithm is same as that of Tinner scheme 3 and the forward/backward substitution can reduce the computation time

• Journal of computational fluids engineering
• /
• v.17 no.4
• /
• pp.9-15
• /
• 2012

Thermal model reduction algorithms and techniques are introduced to condense a huge satellite panel thermal model into the simplified model on the purpose of calculating the thermal responses of a satellite on orbit. Guyan condensation algorithm with the substitution matrix manipulation is developed and the mathematical procedure is depicted step by step. A block-form LU decomposition method is also invited to compare the developed algorithm. The constructed reduced thermal model induced from the detailed model based on a real satellite panel is satisfying the correlation criterion of ${\pm}2^{\circ}C$ for the validity accuracy. Guyan condensation algorithm is superior to the block-form LU decomposition method on computation time.

• International Journal of Precision Engineering and Manufacturing
• /
• v.8 no.1
• /
• pp.3-10
• /
• 2007

A new approach based on implicit surface interpolation combined with domain decomposition is proposed for filling complex-shaped holes in a large polygon model, A surface was constructed by creating a smooth implicit surface from an incomplete polygon model through which the actual surface would pass. The implicit surface was defined by a radial basis function, which is a continuous scalar-value function over the domain $R^{3}$. The generated surface consisted of the set of all points at which this scalar function is zero. It was created by placing zero-valued constraints at the vertices of the polygon model. The well-known domain decomposition method was used to treat the large polygon model. The global domain of interest was divided into smaller domains in which the problem could be solved locally. The LU decomposition method was used to solve the set of small local problems; the local solutions were then combined using weighting coefficients to obtain a global solution. The validity of this new approach was demonstrated by using it to fill various holes in large and complex polygon models with arbitrary topologies.

• Journal of applied mathematics & informatics
• /
• v.30 no.5_6
• /
• pp.759-772
• /
• 2012

In this paper, semi-infinite constrained programming, a class of constrained nonsmooth optimization problems, are transformed into unconstrained nonsmooth convex programs under the help of exact penalty function. The unconstrained objective function which owns the primal-dual gradient structure has connection with $\mathcal{VU}$-space decomposition. Then a $\mathcal{VU}$-space decomposition method can be applied for solving this unconstrained programs. Finally, the superlinear convergence algorithm is proved under certain assumption.

• Wind and Structures
• /
• v.39 no.2
• /
• pp.125-140
• /
• 2024

The simulation of non-stationary wind velocity is particularly crucial for the wind resistant design of slender structures. Recently, some data-driven simulation methods have received much attention due to their straightforwardness. However, as the number of simulation points increases, it will face efficiency issues. Under such a background, in this paper, a time-varying coherence matrix decomposition method based on Diagonal Proper Orthogonal Decomposition (DPOD) interpolation is proposed for the data-driven simulation of non-stationary wind velocity based on S-transform (ST). Its core idea is to use coherence matrix decomposition instead of the decomposition of the measured time-frequency power spectrum matrix based on ST. The decomposition result of the time-varying coherence matrix is relatively smooth, so DPOD interpolation can be introduced to accelerate its decomposition, and the DPOD interpolation technology is extended to the simulation based on measured wind velocity. The numerical experiment has shown that the reconstruction results of coherence matrix interpolation are consistent with the target values, and the interpolation calculation efficiency is higher than that of the coherence matrix time-frequency interpolation method and the coherence matrix POD interpolation method. Compared to existing data-driven simulation methods, it addresses the efficiency issue in simulations where the number of Cholesky decompositions increases with the increase of simulation points, significantly enhancing the efficiency of simulating multivariate non-stationary wind velocities. Meanwhile, the simulation data preserved the time-frequency characteristics of the measured wind velocity well.

• Transactions of the Korean Society of Mechanical Engineers
• /
• v.16 no.10
• /
• pp.1915-1927
• /
• 1992

A three-dimensional Navier-Stokes code has been developed for analysis of viscous flows through turbomachinery blade rows or other internal passages. The Navier-Stokes equations are written in a cartesian coordinate system, then mapped to a general body-fitted coordinate system. Streamwise viscous terms are neglected and turbulent effects are modeled using the baldwin-Lomax model. Equations are discretized using finite difference method on the stacked C-type grids and solved using LU-ADI decomposition scheme. calculations are made for a two-dimensional cascade in a transonic wind-tunnel to see the infuence of the endwalls. The flow pattern of the three-dimensional flow near the endwall is found to be different from that of the two-dimensional flow due to the existence of the endwalls.

• Journal of the Korean Society for Precision Engineering
• /
• v.23 no.1 s.178
• /
• pp.174-184
• /
• 2006

In order to fill the holes with complex shapes in the large polygon model, a new approach which is based on the implicit surface interpolation method combined with domain decomposition method is presented. In the present study, a surface is constructed by creating smooth implicit surface from the incomplete polygon model through which the surface should pass. In the method an implicit surface is defined by a radial basis function, a continuous scalar-valued function over the domain $R^3$ The generated surface is the set of all points at which this scalar function takes on the value zero and is created by placing zero-valued constraints at the vertices of the polygon model. In this paper the well-known domain decomposition method is used in order to treat the large polygon model. The global domain of interest is divided into smaller domains where the problem can be solved locally. LU decomposition method is used to solve a set of small local problems and their local solutions are combined together using the weighting coefficients to obtain a global solution. In order to show the validity of the present study, various hole fillings are carried out fur the large and complex polygon model of arbitrary topology.

• Wind and Structures
• /
• v.38 no.1
• /
• pp.75-91
• /
• 2024

Reliable wind signal reconstruction can be beneficial to the operational safety of long-span bridges. Non-Gaussian characteristics of wind signals make the reconstruction process challenging. In this paper, non-Gaussian wind signals are converted into a combined prediction of two kinds of features, actual wind speeds and wind angles of attack. First, two decomposition techniques, empirical mode decomposition (EMD) and variational mode decomposition (VMD), are introduced to decompose wind signals into intrinsic mode functions (IMFs) to reduce the randomness of wind signals. Their principles and applicability are also discussed. Then, four artificial intelligence (AI) algorithms are utilized for wind signal reconstruction by combining the particle swarm optimization (PSO) algorithm with back propagation neural network (BPNN), support vector regression (SVR), long short-term memory (LSTM) and bidirectional long short-term memory (Bi-LSTM), respectively. Measured wind signals from a bridge site in a deep-cutting gorge are taken as experimental subjects. The results showed that the reconstruction error of high-frequency components of EMD is too large. On the contrary, VMD fully extracts the multiscale rules of the signal, reduces the component complexity. The combination of VMD-PSO-Bi-LSTM is demonstrated to be the most effective among all hybrid models.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.2
• /
• pp.810-831
• /
• 2019

Recently, continuous dimensional emotion recognition from audiovisual clues has attracted increasing attention in both theory and in practice. The large amount of data involved in the recognition processing decreases the efficiency of most bimodal information fusion algorithms. A novel algorithm, namely the incomplete Cholesky decomposition based kernel cross factor analysis (ICDKCFA), is presented and employed for continuous dimensional audiovisual emotion recognition, in this paper. After the ICDKCFA feature transformation, two basic fusion strategies, namely feature-level fusion and decision-level fusion, are explored to combine the transformed visual and audio features for emotion recognition. Finally, extensive experiments are conducted to evaluate the ICDKCFA approach on the AVEC 2016 Multimodal Affect Recognition Sub-Challenge dataset. The experimental results show that the ICDKCFA method has a higher speed than the original kernel cross factor analysis with the comparable performance. Moreover, the ICDKCFA method achieves a better performance than other common information fusion methods, such as the Canonical correlation analysis, kernel canonical correlation analysis and cross-modal factor analysis based fusion methods.