• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2002년도 ITC-CSCC -1
• /
• pp.22-25
• /
• 2002

The Mass production of ceramic heater has encountered with the estimation for the proper parameters of the printing conditions. This paper presents a method to estimate the squeegee pressure. It uses resistance distribution from the trial run with approximate squeegee pressure which comes from statistical process control (SPC). Then, the resistance distribution and its total resistance are input to the backpropagation neural networks that can recognize resistance's distribution patterns. The value of output network derived from the input value can identify to the appropriate squeegee pressure. The experimental results are demonstrated In ensure the efficiency and the reliability of this method with the accuracy 96.75 percent. Indeed, embedded on this method will aid us to reduce the loss from the normal mass production.

• 오토저널
• /
• 제14권6호
• /
• pp.74-80
• /
• 1992

The purpose of this study is to improve the performance of gasoline engine. Combustion-characteristics orignated from supplying cylinder with fuel-air mixture which was formed by the rise of oxygen-concentration in air with oxygen-enriching membrane have been investigated. The results showed that the poor-limit of oxygen-concentration was increased by shortening combustion-duration because the rise of oxygen-concentration in fuel-air mixture resulted in the promotion of combustion-velocity. Also, the generation of large output of power was expected from combustion in proportion as the amount of oxygen was increased.

• 제어로봇시스템학회논문지
• /
• 제14권9호
• /
• pp.886-892
• /
• 2008

In this paper, a closed-loop observer to extract the center of mass (CoM) of a bipedal robot is suggested. Comparing with the simple conversion method of just using joint angle measurements, it enables to get more reliable estimates by fusing both joint angle measurements and F/T sensor outputs at ankle joints. First, a nonlinear-type observer is constructed to estimate the flexible rotational motion of the biped in the extended Kalman filter framework. It adopts the flexible inverted pendulum model which is appropriate to address the flexible motion of bipeds, specifically in the single support phase. The predicted estimates of CoM in terms of the flexible motion observer are combined with measurements (that is, output of the CoM conversion equation with joint angles). Then, we have final CoM estimates depending on the weighting values which penalize the flexible motion model and the CoM conversion equation. Simulation results show the effectiveness of the proposed algorithm.

• 한국자동차공학회논문집
• /
• 제6권4호
• /
• pp.1-8
• /
• 1998

In order to solve the conflict problem between the ride comfort and the road holding, the optimal design of shock absorber that minimizes the r.m.s. of sprung mass vertical acceleration and pitch rate with the understeer characteristics constraints in the high speed stability is proposed. The design of experiments and the nonlinear optimization algorithm are used together to obtain the optimal design of shock absorber. The second order regression models of the input variables(front and rear damping coefficients) and the output variables (ride comfort index and road holding one) are obtained by the central composite design in the design of experiments. Then the optimal design of shock absorber can be systematically adjusted with applying the nonlinear optimization algorithm to the obtained second order regression model. The frequency response analysis of sprung mass acceleration and pitch rate shows the effectiveness of the proposed optimal design of shock absorber in the sprung mass resonance range with the understeer characteristics constraints.

• 설비공학논문집
• /
• 제23권3호
• /
• pp.223-230
• /
• 2011

The power cycle using ammonia-water mixture as a working fluid is a possible way to improve efficiency of the system of low-temperature source. In this work thermodynamic performance of the ammonia-water regenerative Rankine cycle with partial-boiling flow is analyzed for purpose of extracting maximum power from the source. Effects of the system parameters such as mass fraction of ammonia, turbine inlet pressure or ratio of partial-boiling flow on the system are parametrically investigated. Results show that the power output increases with the mass fraction of ammonia but has a maximum value with respect to the turbine inlet pressure, and is able to reach 22 kW per unit mass flow rate of source air at $180^{\circ}C$.

• 한국지진공학회:학술대회논문집
• /
• 한국지진공학회 1999년도 춘계 학술발표회 논문집 Proceedings of EESK Conference-Spring
• /
• pp.202-216
• /
• 1999

To control the motion of building structures under earthquakes their response should be measured first by various sensors and transformed into the control forces using some control algorithms. Of many control algorithms linear quadratic control is widely used as it is easy to implement and analyze. However the algorithms has the disadvantage that it needs the real-time measurements of all state variables(i.e, building's displacements and velocities) which are difficult to achieve for the building structures under earthquakes. Thus the practical algorithms employing output feedback are developed. In this paper LQG algorithm is used for the control of the building model with an active mass driver. The building's acceleration is used to obtain the control gain and the Kalman filter gain. The LQG control strategy is verified with the experimental study on the one-storybuilding model equipped with the active mass driver. This paper demonstrates experimentally the efficacy of the LQG algorithm based on the active mass driver system in reducing the response of seismically excited buildings.

• 한국정밀공학회지
• /
• 제17권1호
• /
• pp.34-44
• /
• 2000

In this paper, a modified skyhook control for the semi-active Macpherson suspension system is investigated. A new model for the semi-active type suspension, which incorporates the rotational motion of the unsprung mass, is introduced and an output feedback control law using the skyhook control method is derived. The gains in the skyhook controller are adaptively adjusted by estimating the road conditions. Because two vertical acceleration sensors, one for the sprung mass and another for the unsprung mass, are used rather than using the angle sensor for the rotational motion of the control arm, the relative velocity of the rattle space is filtered using the acceleration signals. For testing the control performance, the actual damping force has been incorporated via the hardware-in-the-loop simulations. The performances of a passive damper and a semi-active damper are compared. Simulation results are provided.

• 한국구조물진단유지관리공학회 논문집
• /
• 제20권3호
• /
• pp.10-17
• /
• 2016

본 연구는 단경간 교량의 정적하중입력/변위출력관계를 이용한 새로운 교량 유한요소모델 개선 방법을 제안하였고, 실내 모형교량 실험을 통해 검증하였다. 기존의 유한요소모델개선기법은 실험으로부터 얻어진 모드계수와 유한요소모델로부터 예측된 모드계수가 유사해지도록 유한요소모델을 개선하는데, 이 과정에서 구조계의 질량행렬에 대한 가정을 필요로 한다. 제안된 기법은 질량행렬을 가정하지 않고, 오히려 질량행렬 추정을 가능하게 하는 장점을 가진다. 제안된 기법은 두 단계로 구성된다. 첫째, 정적 하중입력-변위응답으로부터 강성행렬을 개선하고, 둘째, 실측된 고유진동수를 이용하여 질량행렬을 개선한다. 실험검증을 위하여 실내 모형교량을 제작하였고, 제안된 기법을 이용하여 모형교량의 탄성계수를 추정하였으며, Universal Testing Machine으로 부터 얻어진 탄성계수와 비교하였다. 또한 기존의 유한요소모델개선기법으로 추정된 탄성계수와 비교하였다. 실험의 결과들로부터 제안된 기법이 합리적으로 탄성계수와 질량밀도를 추정하는 것이 관찰되었고, 기존의 유한요소모델개선기법은 고차모드를 사용했을 때 상대적으로 큰 오차를 주는 것이 관찰되었다. 추가적으로 유한요소모델링 오차에 대하여 토의하였다.

• Smart Structures and Systems
• /
• 제32권5호
• /
• pp.281-295
• /
• 2023

The purpose of this study is to demonstrate and verify the application of phase-control absolute-acceleration-feedback active tuned mass dampers (PCA-ATMD) to multiple-degree-of-freedom (MDOF) building structures. In addition, servo speed control technique has been developed as a replacement for force control in order to mitigate the negative effects caused by friction and inertia. The essence of the proposed PCA-ATMD is to achieve a 90° phase lag for a structure by implementing the desired control force so that the PCA-ATMD can receive the maximum power flow with which to effectively mitigate the structural vibration. An MDOF building structure with a PCA-ATMD and a real-time filter forming a complete system is modeled using a state-space representation and is presented in detail. The feedback measurement for the phase control algorithm of the MDOF structure is compact, with only the absolute acceleration of one structural floor and ATMD's velocity relative to the structure required. A discrete-time direct output-feedback optimization method is introduced to the PCA-ATMD to ensure that the control system is optimized and stable. Numerical simulation and shaking table experiments are conducted on a three-story steel shear building structure to verify the performance of the PCA-ATMD. The results indicate that the absolute acceleration of the structure is well suppressed whether considering peak or root-mean-square responses. The experiment also demonstrates that the control of the PCA-ATMD can be decentralized, so that it is convenient to apply and maintain to real high-rise building structures.

• Design & Manufacturing
• /
• 제17권4호
• /
• pp.72-78
• /
• 2023

Injection molding is a process widely used in various industries because of its high production speed and ease of mass production during the plastic manufacturing process, and the product is molded by injecting molten plastic into the mold at high speed and pressure. Since process conditions such as resin and mold temperature mutually affect the process and the quality of the molded product, it is difficult to accurately predict quality through mathematical or statistical methods. Recently, studies to predict the quality of injection molded products by applying artificial neural networks, which are known to be very useful for analyzing nonlinear types of problems, are actively underway. In this study, structural optimization of neural networks was conducted by applying multi-task learning techniques according to the characteristics of the input and output parameters of the artificial neural network. A structure reflecting the characteristics of each process step was applied to the input parameters, and a structure reflecting the quality characteristics of the injection molded part was applied to the output parameters using multi-tasking learning. Building an artificial neural network to predict the three qualities (mass, diameter, height) of injection-molded product under six process conditions (melt temperature, mold temperature, injection speed, packing pressure, pacing time, cooling time) and comparing its performance with the existing neural network, we observed enhancements in prediction accuracy for mass, diameter, and height by approximately 69.38%, 24.87%, and 39.87%, respectively.