• Journal of Institute of Control, Robotics and Systems
• /
• v.14 no.3
• /
• pp.226-231
• /
• 2008

In this paper, we propose novel eigenstructure assignment method in full-state feedback for linear time-invariant MIMO system such that directions of some left eigenvectors are exactly assigned to the desired directions. It is required to consider the direction of left eigenvector in designing eigenstructure of closed-loop system, because the direction of left eigenvector has influence over excitation by associated input variables in time-domain response. Exact direction of a left eigenvector can be achieved by assigning proper right eigenvector set satisfying the conditions of the presented theorem based on Moore's theorem and the orthogonality of left and right eigenvector. The right eigenvector should reside in the subspace given by the desired eigenvalue, which restrict a number of designable left eigenvector. For the two cases in which desired eigenvalues are all real and contain complex number, design freedom of designable left eigenvector are given.

• The Journal of Information Technology
• /
• v.7 no.2
• /
• pp.43-54
• /
• 2004

Since Dr. John J. Hopfield has proposed the HOpfield network, it has been widely applied to the pattern recognition and the routing optimization. The method of Jian-Hua Li improved efficiency of Hopfield network which input pattern's weights are regenerated by SVD(singluar value decomposition). This paper deals with Li's Hopfield Network by linear pre-processing. Linear pre-processing is used for increasing orthogonality of input pattern set. Two methods of pre-processing are used, Hadamard method and random method. In manner of success rate, radom method improves maximum 30 percent than the original and hadamard method improves maximum 15 percent. In manner of success time, random method decreases maximum 5 iterations and hadamard method decreases maximum 2.5 iterations.

• Journal of computational fluids engineering
• /
• v.6 no.1
• /
• pp.1-13
• /
• 2001

The natural convection heat transfer and solidification in a stratified pool are studied. The flow and heat transfer characteristics in a heat generating pool are compared between single-layered and double-layered pools. And local Nusselt number distributions on outer walls are obtained to consider thermal loads on a vessel wall. The cooling and solidification of Al₂O₃/Fe melt in a hemispherical vessel are simulated to study the mechanism of heat transfer and temperature distribution. A unstructured mesh is chosen for this study because of the non-orthogonality originated from the boundaries of double-layered pool. Interface between the layers is modeled to be fixed. With this assumption mass flux across the interface is neglected, but shear force and heat flux are considered by boundary conditions. The colocated cell-centered finite volume method is used with the Rhie-Chow interpolation to compute cell face velocity. To prevent non-physical solutions near walls in case body force is large the wall pressure is extrapolated by the way to include body force. The numerical solutions calculated by current method show that averaged downward heat flux of the double-layered pool increases compared to single-layered pool and maximum temperature occurs right below the interface of the layers.

• Kyungpook Mathematical Journal
• /
• v.55 no.1
• /
• pp.63-71
• /
• 2015

Let H be a separable infinite dimensional complex Hilbert space, and let $\mathcal{L}(H)$ denote the algebra of all bounded linear operators on H into itself. Let $A,B{\in}\mathcal{L}(H)$ we define the generalized derivation ${\delta}_{A,B}:\mathcal{L}(H){\mapsto}\mathcal{L}(H)$ by ${\delta}_{A,B}(X)=AX-XB$, we note ${\delta}_{A,A}={\delta}_A$. If the inequality ${\parallel}T-(AX-XA){\parallel}{\geq}{\parallel}T{\parallel}$ holds for all $X{\in}\mathcal{L}(H)$ and for all $T{\in}ker{\delta}_A$, then we say that the range of ${\delta}_A$ is orthogonal to the kernel of ${\delta}_A$ in the sense of Birkhoff. The operator $A{\in}\mathcal{L}(H)$ is said to be finite [22] if ${\parallel}I-(AX-XA){\parallel}{\geq}1(*)$ for all $X{\in}\mathcal{L}(H)$, where I is the identity operator. The well-known inequality (*), due to J. P. Williams [22] is the starting point of the topic of commutator approximation (a topic which has its roots in quantum theory [23]). In [16], the author showed that a paranormal operator is finite. In this paper we present some new classes of finite operators containing the class of paranormal operators and we prove that the range of a generalized derivation is orthogonal to its kernel for a large class of operators containing the class of normal operators.

• Journal of the Society of Naval Architects of Korea
• /
• v.28 no.1
• /
• pp.83-91
• /
• 1991

In this paper the dynamic analysis of the multi-span beam on elastic foundation, which include discrete transnational and rotational springs, was performed. Furthermore, the effects of the intermediate supports were investigated. As a solution method, first the orthogonal polynomial functions which satisfy both the geometric and dynamic boundary conditions are obtained by imposing the orthogonality conditions. Then, the Galerkin's method is used to obtain the natural frequencies of the system. From numerical tests for various constraint and boundary conditions, it was found that the higher order orthogonal polynomial functions obtained by the present method can be used to get the accurate solution.

• Nuclear Engineering and Technology
• /
• v.49 no.5
• /
• pp.1006-1009
• /
• 2017

We discuss the problems of scattering in this framework, and show that the applied method is very useful in the investigation of the effect of the resonance in the observed scattering cross sections. In this study, not only the scattering cross sections but also the decomposition of the scattering cross sections was computed for the ${\alpha}-{\alpha}$ system. To obtain the decomposition of scattering cross sections into resonance and residual continuum terms, the complex scaled orthogonality condition model and the extended completeness relation are used. Applying the present method to the ${\alpha}-{\alpha}$ and ${\alpha}-n$ systems, we obtained good reproduction of the observed phase shifts and cross sections. The decomposition into resonance and continuum terms makes clear that resonance contributions are dominant but continuum terms and their interference are not negligible. To understand the behavior of observed phase shifts and the shape of the cross sections, both resonance and continuum terms are calculated.