• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 2000년도 추계학술대회 논문집
• /
• pp.647-652
• /
• 2000

This paper presents a study on the position tracking control of robot system on the uncertain elastic base. The elastic base is modeled as a virtual robot which has passive joints and the control strategy is using approximate Jacobian operators. Jacobian operators represent the overall robot system including base movement. However, because we don't know the base movement we can't estimate the jacobian operators directly. The control algorithm is proposed which uses only Jacobian operators of a real robot as approximate Jacobian operators. The measured errors from external sensor are compensated by approximate Jacobian operators. The simulation results of a single-axis robot system show that the control strategy can be used for position tracking.

• 한국정밀공학회지
• /
• 제18권10호
• /
• pp.45-52
• /
• 2001

This paper presents a study on the position tracking control of a robot system on the uncertain elastic base. The elastic bathe is a nonholonomic system but it can be changed into holonomic system, which is much easier to analyze, by modeling an elastic base as a virtual robot that has passive joints. Also, Jacobian operators, which represent the overall robot system including base movement, are defined and applied to the changed model. However, because base movements are not known, the exact Jacobian operators can＇t be estimated. The control algorithm proposed is that uses only Jacobians of a real robot as approximate Jacobian operators. Therefore the approximate Jacobian operators compensate the measured errors from external sensors. The proposed control strategy is evaluated by the simulation and experiment of a single-axis robot system on the elastic base.

• 제어로봇시스템학회논문지
• /
• 제6권12호
• /
• pp.1086-1092
• /
• 2000

To realize a visual servo control of slender manipulators, two problems to be solved are analysed. The stability problem on so-called noncolocation control and the infinite order problem of the real Jacobian matrix caused by the elastic deformation are discussed. By considering the dynamic relations between rigid and elastic modes, a Jacobian operator is derived and the physical meaning is also explained. Then, for practical control, a simple control scheme using an approximate Jacobian is proposed and its stable conditions are proven by means of the $L_$2$ stability theory. The scheme is structurally similar to the conventional PD control laws, but external sensors(e. g. visual sensor) are used for positioning and internal sensors for damping. A good performance is obtained via control experiments of a slender two link manipulator.

• 한국전산유체공학회:학술대회논문집
• /
• 한국전산유체공학회 1999년도 추계 학술대회논문집
• /
• pp.59-64
• /
• 1999

The Equations of Chemical kinetics ate very stiff, which forces the use of an implicit scheme. The problem of implicit scheme, however, is that the jacobian must be solved at each time step. In this paper, we examined the approximate chemical jacobian methods such as Gauss-Seidel, Jacobi partial jacobian and diagonalized jacobian that can be stable without full jacobian, We show that Gauss-Seidel jacobian method is stable and accurate as well as full jacobian and that this is more efficient in supersonic combustion problem about $20\%$ than the full jacobian method with same accuracy,

• 한국전산유체공학회:학술대회논문집
• /
• 한국전산유체공학회 2005년도 춘계 학술대회논문집
• /
• pp.219-222
• /
• 2005

A compressible two-fluid two-phase flow computation model using the stiffened-gas equation of state is formulated. Since the conservation equation system is of mixed type, it gives complex eigenvalues. The sonic speeds obtained from the individual single phase have been simply used in the literature for the fastest wave speeds necessary in the HLL scheme. This method has worked fine but proved to be quite diffusive according to our test. To improve the accuracy, we here propose to utilize the analytic eigenvalues evaluated from an approximate Jacobian matrix lot the fastest wave speeds. The interfacial transfer terms were dropped in constituting the Jacobian matrix for this purpose. The present scheme proved efficient, robust and accurate in comparison with other existing methods. We solved the cavitating flow problem using the present scheme. The result shows more detailed wave structure in the cavitating process caused by the strong expansion waves.

• 한국전산유체공학회:학술대회논문집
• /
• 한국전산유체공학회 2005년도 추계 학술대회논문집
• /
• pp.183-186
• /
• 2005

We developed an upwind numerical formulation based on the eigenvalues of the approximate Jacobian matrix in order to solve the hyperbolic conservation laws governing the two-fluid two-phase flow models. We obtained eight analytic eigenvalues in the two dimensions that can be used for estimate of the wave speeds essential in constructing an upwind numerical method. Two-dimensional underwater cavitation in a flow past structural shapes or by underwater explosion can be solved using this method. We present quantitative prediction of cavitation for the water tunnel wall and airfoils that has both experimental data as well as numerical results by other numerical methods and models.

• 한국정밀공학회지
• /
• 제22권8호
• /
• pp.34-41
• /
• 2005

Generally, it is not easy to get a suitable controller for multi variable systems. If the modeling equation of the system can be found, it is possible to get LQR control as an optimal solution. This paper suggests an LQR learning method to design LQR controller without the modeling equation. The proposed algorithm uses the same cost function with error and input energy as LQR is used, and the LQR controller is trained to reduce the function. In this training process, the Jacobian matrix that informs the converging direction of the controller Is used. Jacobian means the relationship of output variations for input variations and can be approximately found by the simple experiments. In the simulations of a hydrofoil catamaran with multi variables, it can be confirmed that the training of LQR controller is possible by using the approximate Jacobian matrix instead of the modeling equation and this controller is not worse than the traditional LQR controller.

• Journal of Mechanical Science and Technology
• /
• 제15권1호
• /
• pp.125-138
• /
• 2001

An efficient correction storage scheme on a structured grid is applied to a sequence of approximate Jacobian systems arising at each time step from a linearization of the discrete nonlenear system of equations, obtained by the implicit time discretization of the conservation laws for unsteady fluid flows. The contribution of freezing the Jacobian matrix to computing costs is investigated within the correction storage scheme. The performance of the procedure is exhibited by measuring CPU time required to obtain a fully developed laminar vortex shedding flow past a circular cylinder, and is compared with that of a collective iterative method on a single grid. In addition, some computed results of the flow are presented in terms of some functionals along with measured data. The computational test shows that the computing costs may be saved in favor of the correction storage scheme with the frozen Jacobian matrix, to a great extent.

• Nuclear Engineering and Technology
• /
• 제49권6호
• /
• pp.1181-1188
• /
• 2017

This paper proposes the adoption of Monte Carlo perturbation theory to approximate the Jacobian matrix of coupled neutronics/thermal-hydraulics problems. The projected Jacobian is obtained from the eigenvalue decomposition of the fission matrix, and it is adopted to solve the coupled problem via the Newton method. This avoids numerical differentiations commonly adopted in Jacobian-free Newton-Krylov methods that tend to become expensive and inaccurate in the presence of Monte Carlo statistical errors in the residual. The proposed approach is presented and preliminarily demonstrated for a simple two-dimensional pressurized water reactor case study.

• 전기학회논문지
• /
• 제66권5호
• /
• pp.819-832
• /
• 2017

We propose a new switching control scheme for a ball and beam system by utilizing two linearization methods. First, the Jacobian linearization is applied and state observer is developed afterward. Then, motivated [6], the approximate input-output linearization is carried out, and after that, the Jacobian linearization is applied along with the design of state observer. Since the second approach requires two linearizations, it is called a two-step linearization method. The state observer is needed for the estimation of the velocities of ball and motor movement. Since the Jacobian linearization based controller tends to provide faster response at the initial time, and after that, the two-step linearization based controller tends to provide better response in terms of output overshoot and convergence to the origin, it is natural to give a switching control scheme to provide the best overall control response. The validity of our control scheme is shown in both simulation and experimental results.