• Proceedings of the Korean Geotechical Society Conference
• /
• 2000.11a
• /
• pp.705-712
• /
• 2000

The fluid generally flows through fractures in crystalline rocks where most of underground storage facilities are constructed because of their low hydraulic conductivities. The fractured rock is better to be conceptualized with a discrete fracture concept, rather continuum approach. In the aspect of fluid flow in underground, the simultaneous flow of groundwater and gas should be considered in the cases of generation and leakage of gas in nuclear waste disposal facilities, air sparging process and soil vapor extraction for eliminating contaminants in soil or rock pore, and pneumatic fracturing for the improvement of permeability of rock mass. For the purpose of appropriate analysis of groundwater-gas flow, this study presents an unsteady-state multi-phase FEM fracture network simulator. Numerical simulation has been also conducted to investigate the hydraulic head distribution and air tightness around Ulsan LPG storage cavern. The recorded hydraulic head at the observation well Y was -5 to -10 m. From the results obtained by the developed model, it shows that the discrete fracture model yielded hydraulic head of -10 m, whereas great discrepancy with the field data was observed in the case of equivalent continuum modeling. The air tightness of individual fractures around cavern was examined according to two different operating pressures and as a result, only several numbers of fractures neighboring the cavern did not satisfy the criteria of air tightness at 882 kPa of cavern pressure. In the meantime, when operating pressure is 710.5 kPa, the most areas did not satisfy air tightness criteria. Finally, in the case of gas leaking from cavern to the surrounding rocks, the resulted hydraulic head and flowing pattern was changed and, therefore, gas was leaked out from the cavern ceiling and groundwater was flowed into the cavern through the walls.

• Journal of Biomedical Engineering Research
• /
• v.16 no.4
• /
• pp.415-424
• /
• 1995

Combined models, specified by two or more modeling formalisms, can represent a wide variety of complex systems. This paper describes a methodology for the development of combined models in two model types of discrete event and continuous process. The methodology is based on transformation of continuous state space into discrete one to homomorphically represent dynamics of continuous processes in discrete events. This paper proposes a formal structure which can combine model of the DES and the CS within a framework. The structure employs the DEVS formalism for the DES models and differential or polynomial equations for the CS models. To employ the proposed structure to specify a DEVS/CS combined model, a modeler needs to take the following steps. First, a modeler should identify events in the CS and transform the states of the CS into the DES. Second, a modular employs the formalism to specify the system as the DES. Finally, a moduler developes sub-models for the CS and continguos states of the DES and establishs one-to-one correspondence between the sub-models and such states. The proposed formal structre has been applied to develop a DEVS/CS combined model for the human cardiovascular system. For this, the cardiac cycle is partitioned into a set of phases based on events identified through observation. For each phase, a CS model has been developed and associated with the phase. To validate the DEVS/CS combined model developed, then simulate the model in the DEVSIM + + environment, which is a model simulation results with the results obtained from the CS model simulation using SPICE. The comparison shows that the DEVS/CS combined model adequately represents dynamics of the human heart system at each phase of cardiac cycle.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 1997.04a
• /
• pp.413-417
• /
• 1997

Distributed piezoelectric sensor and actuator have been designed for efficient vibration control of a cantilevered beam. Both PZT and PVDF are used in this study, the former as an actuator and the latter as a sensor for our integrated structure. For the PZT actuator, the position and size have been optimized. Optimal electrode shape of the PVDF sensor has been determined. For multi-mode vibration control, we have used two PZT actuators and a PVDF sensor. Electrode shading of PVDF is more powerful for modal force adjustment than the sizing and positioning of PZT. Finite element method is used to model the structure that includes the PZT actuator and the PVDF sensor. By deciding on or off of each PZT segment, the length and the location of the PZT actuator are optimize. Considering both of the host structure and the optimized actuators, it is designed that the active electrode width of PVDF sensor along the span of the beam. Actuator design is based on the criterion of minimizing the system energy in the control modes under a given initial condition. Sensor is designed to minimize the observation spill-over. Modal control forces for the residual(uncontrolled) modes have been minimized during the sensor design. Genetic algorithm, which is suitable for this kind of discrete problems, has been utilized for optimization. Discrete LQG control law has been applied to the integrated structure for real time vibration control. Performance of the sensor, the actuator, and the integrated smart structure has been demonstrated by experiments.

• International Journal of Aeronautical and Space Sciences
• /
• v.7 no.1
• /
• pp.106-117
• /
• 2006

The active vibration control of composite shell structure has been performed with the optimized sensor/actuator system. For the design of sensor/actuator system, a method based on finite element technique is developed. The nine-node Mindlin shell element has been used for modeling the integrated system of laminated composite shell with PVDF sensor/actuator. The distributed selective modal sensor/actuator system is established to prevent the effect of spillover. Electrode patterns and lamination angles of sensor/actuator are optimized using genetic algorithm. Continuous electrode patterns are discretized according to finite element mesh, and orientation angle is encoded into discrete values using binary string. Sensor is designed to minimize the observation spillover, and actuator is designed to minimize the system energy of the control modes under a given initial condition. Modal sensor/actuator for the first and the second mode vibration control of singly curved cantilevered composite shell structure are designed with the method developed on the finite element method and optimization. For verification, the experimental test of the active vibration control is performed for the composite shell structure. Discrete LQG method is used as a control law.

• The Transactions of the Korean Institute of Power Electronics
• /
• v.17 no.5
• /
• pp.416-422
• /
• 2012

This paper proposes a low power-loss current measurement using a resistor and bypass switch. Conventional current sensing method using a resistor has a disadvantage of power loss which degrades the efficiency of the entire systems. On the other hand, proposed measurement technique operating with bypass-switch connected in parallel with sensing resistor can reduce power loss significantly the current sensor. The propose measurement works for discrete-time sampling of current sensing. Even while the analog-digital conversion does not occur at the controller, the sensing voltage across the sensor still causes ohmic conduction loss without information delivery. Hence, the bypass switch bypasses the sensing current with a small amount of power loss. In this paper, a 90[W] prototype hardware has been implemented for photovoltaic MPPT experimental verification of the proposed low power-loss current measurement technique. From the results, it can be seen that PV power observation is successfully done with the proposed method.

• Geomechanics and Engineering
• /
• v.18 no.3
• /
• pp.259-266
• /
• 2019

This paper presents a comparison between experimental measurements and numerical estimations of penetration length of a cement grout injected in discrete joints. In the experiment, a joint was generated by planar acryl plates with a certain separation distance (; aperture) and was designed in such a way to vary the separation distances. Since a cement grout was used, the grout viscosity can be varied by controlling water-cement (W/C) ratios. Throughout these experiments, the influence of joint aperture, cement grout viscosity, and injection rate on a penetration length in a discrete joint was investigated. During the experiments, we also measured the time-dependent variation of grout viscosity due to a hardening process. The time-dependent viscosity was included in our numerical simulations as a function of elapsed time to demonstrate its impact on the estimation of penetration length. In the numerical simulations, Bingham fluid model that has been known to be applicable to a viscous cement material, was employed. We showed that the estimations by the current numerical approach were well comparable to the experimental measurements only in limited conditions of lower injection rates and smaller joint apertures. The difference between two approaches resulted from the facts that material separation (; bleeding) of cement grout, which was noticeable in higher injection rate and there could be a significant surface friction between the grout and joint planes, which are not included in the numerical simulations. Our numerical simulation, meanwhile, could well demonstrate that penetration length can be significantly over-estimated without considering a time-dependency of viscosity in a cement grout.

• Journal of the Chungcheong Mathematical Society
• /
• v.26 no.4
• /
• pp.853-867
• /
• 2013

In this paper we study the cardinality of the dot product set generated by two subsets of vector spaces over finite fields. We notice that the results on the dot product problems for one set can be simply extended to two sets. Let E and F be subsets of the d-dimensional vector space $\mathbb{F}^d_q$ over a finite field $\mathbb{F}_q$ with q elements. As a new result, we prove that if E and F are subsets of the paraboloid and ${\mid}E{\parallel}F{\mid}{\geq}Cq^d$ for some large C > 1, then ${\mid}{\Pi}(E,F){\mid}{\geq}cq$ for some 0 < c < 1. In particular, we find a connection between the size of the dot product set and the number of lines through both the origin and a nonzero point in the given set E. As an application of this observation, we obtain more sharpened results on the generalized dot product set problems. The discrete Fourier analysis and geometrical observation play a crucial role in proving our results.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.37 no.5
• /
• pp.34-45
• /
• 2000

In this paper, a multiagent fault-tolerant supervisory control methodology is proposed for serial production systems under partial observation. To this end, the idea of multiagent control is incorporated with the fault-tolerant supervisory control of discrete event systems. Especially, the concept of mutual fault-tolerance between cascaded processes is established and the unobservable fault identification utilizing the difference information of the controlled results is investigated. Then the synthesis of agent supervisors is formulated based on the proposed concept. A case study of the fault-propagation control of polypropylene prepolymerization and polymerization processes is provided to illustrate the proposed control policy.

• Journal of KSNVE
• /
• v.7 no.6
• /
• pp.967-974
• /
• 1997

Distributed piezoeletric sensor and actuator have been designed for efficient vibration control of a cantilevered beam. Both PZT and PVDF have been used in this study, the former as an actuator and the latter as a sensor for the integrated structure. We have optimized the position and the size of the PZT actuator and the electrode shape of the PVDF sensor. Finite element method is used to model the structure and the optimized actuators, we have designed the active electrode width of the PVDF sensor along the span of the beam. Actuator design is based on the criterion of minimizing the system energy in the control modes under a given initial condition. Model control forces for the residual (uncontrolled) modes have been minimized during the sensor design to minimize the observation spill-over. Genetic algorithm and sequential quadratic programming technique have been utilized as an optimization scheme. Discrete LQG control law has been applied to the integrated structure for real time vibration control. Performance of the sensor, the actuator, and the integrated smart structure has been demonstrated by experiments.

• International Journal of Reliability and Applications
• /
• v.1 no.1
• /
• pp.65-79
• /
• 2000

This research considers a system which has an ultimate terminal event such as death, critical failure, bankruptcy together with a certain indicative events (temporary malfunction, special treatment, kind of defaults) that frequently occurs before the terminal event comes to the system. Some investigation of a model for the corresponding bivariate data of the system have been done with an explanation of the situation in terms of two continuous variables instead of continuous-discrete variables and some other properties. Also an analysis has been carried out to evaluate the effect of intermediate observation of occurrence of indicative event so that the result can be used for a possible suggestion of an intermediate observing schedule.