Kiwoong Kim

Bi-level optimizing control of a simulated moving bed process with nonlinear adsorption isotherms.

A bi-level optimizing control scheme originally proposed for a simulated moving bed (SMB) with linear isotherms has been extended to an SMB with nonlinear isotherms. Cyclic steady state optimization is performed in the upper level to determine the optimum switching period and time-varying feed/desorbent flow rates, and repetitive model predictive control is run in the lower level for purity regulation, taking the decision variables from the upper level as feed-forward information. Experimental as well as numerical study for an SMB process separating a high-concentration mixture of aqueous L-ribose and L-arabinose solutions showed that the proposed scheme performs satisfactorily against various disturbances. In contrast, an alternative scheme based on an SMB model with linear isotherms showed a limitation in the control performance; this scheme was apt to fail in purity regulation.

Optimal Reactive Power Planning, Part I—Load Level Decomposition

Abstract This paper presents an efficient algorithm for the reactive power planning of transmission networks in normal operations. In order to obtain optimal investment planning an optimal operation is essential. The objective function of an optimal operation is the fuel cost of the generators. The investment variables are the shunt reactors and the capacitors at selected buses. The operation problem contains every physical constraint such as voltage limits, generators capacities and Var compensators capacities. The number of operation modes is so large because of the various load levels in a year. So the variables are increased tremendously, in proportion to the load levels and the investment variables. This leads to difficult computation and to a storage limit. To overcome these difficulties a new variable decomposition technique is adopted which can seperate the operation and the investment variables. This technique makes it possible to decompose load levels and to solve optimal operation problems independently in each level. Therfore the only problem remaining to be solved is the investment problem. The modified investment problem includes implicitly the operation and the operation problem needs no longer to be computed. The quasi linearization method is applied to each problem and so computing time is reduced in a great deal.

Molecular Dynamics Study of Diffusion Behaviors of CO2 and N2 Confined to a Uni-directional Zeolite Structure

Abstract A study of a molecular dynamic simulation was conducted to investigate the diffusion characteristics of the guest molecules, which diffuse into uni-directional zeolitic structures. Because of the complex and narrow micro-pore structures the zeolites possess, an observation of the anomalous diffusion behavior was reported. In particular, single-file diffusion occurs when the zeolites have an unconnected, narrow, single channel, which hinders the mutual passage of guest molecules. In this work, the diffusion behavior of the mixture consisting of CO 2 and N 2 in the zeolite of TON, which has an unconnected single channel, was investigated. Afterward, the change in diffusion behavior for TON anchored with the functional group selected as the amino group was also investigated.

Repetitive Control and Cyclic Steady State Optimization of a Simulated Moving Bed Process

Abstract A two-tier optimizing control strategy for the simulated moving bed (SMB) process proposed in Kim et al.Kim (2010) has been studied experimentally as well as numerically. The strategy was designed to carry out repetitive model predictive control (RMPC) of average product purities in the lower tier and off-line cyclic steady state optimization of feed and desorbent flow rates and the length of switching period in the upper tier. Experimental studies have been conducted in a four zone SMB system to separate L-ribose and L-arabinose from their mixture dissolved in water for the linear isotherm case and numerical studies have been carried out for the nonlinear isotherm case. As a consequence, the optimizing control strategy is observed to satisfactorily attain the intended optimum while robustly regulating the product purities against various uncertainties.