대전시 도시경관관리제도 현황과 개선방안에 관한 연구

김종대 한남대학교 1998 국내석사

르네상스時代의 人體表現에 대한 美術解剖學的 考察

김종대 용인대학교 대학원 2001 국내석사

이중다결정 실리콘게이트 NMOS구조의 채널주입형 EAROM

김종대 경북대학교 1983 국내석사

「은혜 갚은 꿩」의 說話的 特性과 學習指導 方案 硏究

김종대 부산교육대학교 교육대학원 2003 국내석사

「은혜 갚은 꿩」의 說話的 特性과 學習指導 方案 硏究

韓國 도깨비譚 硏究

김종대 中央大學校 大學院 1994 국내박사

새로운 자가확장형 전방십자인대 고정장치의 생체역학적 특성 고찰

김종대 전북대학교 대학원 2008 국내박사

This thesis proposes a self-expansion type anterior cruciate ligament fixation device. The proposed fixation device provides graft fixation force by maintaining contact with the bone tunnel. Since the device maintains contact with the bone tunnel by the force that expands by the self-driven elastic force of the device, the main design parameters that determine the performance of this device are the ring thickness and expansion angle. Three-dimensional finite element models of the fixation devices and a bone are developed. By simulation with the developed finite element model, the influence of the main design parameters of the device on the maximum stress around the ring is studied when grasping the fixation device. Through the analysis of the stress on the bone tunnel wall, when the fixation device comes in contact with the bone tunnel, this thesis shows the influence of the main design parameters of the fixation device on the contact condition. In addition, through the analysis of the migration that occurred upon application of the pull-out load, the influence of the main design parameters on the initial fixation stability of the fixation device is investigated. To perform the pull-out test, the Ti6Al4V fixation device are heated after inserting in a jig for 1 hour at 800℃ in a protective argon gas atmosphere and allowed to cool to room temperature in the furnace. In order to investigate the influence of the expansion angle and the thickness of the ring on the biomechanical properties of the fixation device, the maximum pull-out load, stiffness and slippage of the ACL fixation device are measured using the pull-out tester, and statistical analyses are also executed. The results shows that the design parameters gave significant effect on the performance of the self-expansion type of fixation device. The pull-out load of the ACL fixation devices significantly increase as the thicknesses of the ring section are increased, having a similar trend for both expansion angles. The ACL fixation device shows about 2.5 times higher value of the pull-out load than that of the minimum load (500N) required for the "accelerated rehabilitation". The optimum θ and tR values of this ACL fixation device are suggested to have sufficient resistance against the pull-out force, high stiffness and relatively low slippage after ACL reconstruction.

신경망을 이용한 화학공정의 온라인 적응 제어

김종대 전남대학교 대학원 2001 국내박사

본 논문에서 시간 지연이 있는 연속교반탱크가열기(CSTH)의 온도와 비선형적인 중화 공정의 pH를 제어하기 위한 온라인 신경망 적응 제어기법을 개발하고, 결과를 고찰하였다. 신경망 제어 시스템은 신경망 식별기와 신경망 제어기로 구성되었다. 온라인 신경망 적응 제어의 성능을 향상시키기 위해 퍼지 논리, 예측 알고리즘과 PID 제어기를 포함시킨 여러 가지 신경망 제어기법을 제시하고, 컴퓨터 시뮬레이션과 실험장치를 통해 결과를 고찰하였다. 공정의 동특성을 실시간으로 나타내기 위해 공정 Jacobian을 신경망 식별기에서 미분연쇄법칙에 의해 해석적으로 구하였다. 신경망 제어기에서는 오차 역전파 학습알고리즘에 의해 공정 오차를 줄여가는 방향으로 제어량을 출력하였다. 선형모델 예측 알고리즘과 신경망 제어기의 파라미터에 대한 퍼지논리는 온라인 신경망 제어에서 발생되는 큰 오버슈트와 긴 응답시간을 감소시키는데 이용되었다. 실시간으로 샘플링한 공정데이타를 이용한 선형모델 예측 알고리즘은 시간 지연이 존재하는 연속교반탱크가열기의 신경망 온도제어에서 초기 오버슈트와 응답시간을 감소시켰다. 제시한 제어기법은 약산-강염기계 pH 중화공정의 급격한 반응을 일으키는 적정구간에서 효과적인 제어 성능을 보였다. 신경망의 학습율과 활성함수의 기울기는 학습속도과 관련되어 제어량의 변화를 결정하였다. 본 연구에서 시스템의 동특성과 학습속도에 따라서 퍼지논리에 의해 학습율과 기울기를 조절하였다. 그 결과, 온도제어에서 초기 오버슈트를 감소시키고, 응답시간을 단축하였다. pH 제어에서는 적정구간에서의 응답특성이 효과적으로 개선되었다. 또한, 신경망에 의해 PID 제어기의 파라미터를 자체 조정하는 신경망 PID 제어기를 설계하여 연속교반탱크가열기의 온도와 중화공정에서 pH를 제어한 결과 더 향상된 제어성능을 보였다. 시스템에 외란이 작용하는 경우에도 제시한 모든 신경망 제어에서 양호한 제어성능을 보여주었다. 제시한 신경망 제어 중 가장 효과적인 응답특성을 보인 것은 신경망 자체 조정 PID 제어였다. 연구 결과 온라인 신경망 적응제어에서 퍼지 논리, 예측알고리즘, PID 제어기를 함께 이용하여 시간지연과 비선형성이 있는 공정에서 더욱 향상된 제어결과를 얻을 수 있었다. In this paper, the on-line neural adaptive control method for temperature of continuous stirrer tank heater(CSTH) with time delay and pH control of a nonlinear neutralization process has been developed and evaluated. The neural control system is composed of two neural networks, the neural identifier and the neural controller. To improve the performance of neural adaptive control method, various neural control algorithms involved fuzzy logic, predictive, and PID control algorithm were proposed and tested with computer simulation and experimental system. The process dynamics was described in real time with the Jacobian of the process analytically obtained from neural identifier with derivative chain rule. The control output from neural controller was generated to reduce system error with error backpropagation learning algorithm. Linear model predictive algorithm and fuzzy logic for parameters of neural controller have been used to reduce large overshoots and long response time resulting from on-line neural adaptive control. Linear model predictive algorithm with real time data reduced the overshoot and response time of the initial process control. The developed control method showed good performance for the pH neutralization process of weak acid and strong base system with explosion reaction in titration. The learning rate and the slope of the activation function determined the amount of control output, and also affects control response. The response time and overshoot of the control output were reduced with the developed fuzzy logic for the learning rate and the slope of the activation function. The self tuning PID controller based on neural networks has been developed and applied for the pH control of neutralization process and the temperature control of CSTH. In this case, the control performance was better improved. The control output of the proposed controller with significant disturbance system has shown robust control response by numerical simulation and experiment. Neural self tuning PID control method gave the best performance for the most of cases. On-line neural adaptive control with fuzzy logic, predictive algorithm, and PID controller showed the better performance for the temperature and pH control with time delay and nonlinear processes.

Duality for nanlinear mulitobjective programming with cone constraints

김종대 부경대학교 2007 국내석사

五積散이 Estradiol Valerate로 유발된 다낭성 난소 모델에서 난포의 성숙 및 NGF발현에 미치는 영향

김종대 세명대학교 대학원 2011 국내석사

purpose : This study was designed to investigate the effect of Ojeoksan(OJS) on the estradiol valerate(EV)-induced polycystic ovaries(PCO) in rats. methods : PCO was induced by single intramuscular injection with estradiol valerate(EV)(4mg) in female rats. Normal group(n=8) were injected with sesame oil and orally administrated distilled water for sixty days. PCO control group(n=8) were injected with EV and orally administrated distilled water for sixty days. OJS treated group(n=8) were injected with EV and orally administrated OJS for sixty days. Then, we measured weights of body, ovaries, adrenal glands and uterus of rats. The histomorphometrical changes of ovaries were also evaluated. And we observed the NGF and CRF expression by immunohistochemistry. results : The results were as follows - The weights(㎎) of ovaries in OJS treated group(57.4±9.4) were significantly increased(p<0.05) compared with PCO control group(42.3±8.5). - The numbers of mature follicles in OJS treated group(9.3±2.5) were significantly increased(p<0.05) compared with PCO control group(6.1±2.1). - The expressions of NGF-immunoreactive cells in the ovarian granulosa cells in OJS treated group were weaker than PCO control group. Conclusion : From the these results, we concluded that Ojeoksan(OJS) has inhibitory effect on the development of EV-induced polycystic ovary. And it's effect maybe related with decreased NGF activities in the ovary. Key words : Ojeoksan, polycystic ovaries, estradiol valerate, nerve growth factor, ovarian histopathology

굴뚝효과를 이용한 자연대류 히트싱크의 열성능 향상

김종대 忠南大學校 大學院 2018 국내석사

The natural convection is a method using buoyancy force by temperature difference. Because an additional device is not needed, the maintenance of the heat sink is convenient, and the noise from fan is not generated. However, due to fast development of electronic devices and increment of heat flux, traditional natural convection cooling methods consisted only heat sink are not enough to meet the required cooling performance. One method for increasing the thermal performance of heat sink is to apply chimney effect to heat sink. The chimney increases airflow into the heat sink, enhance the heat transfer rate in heat sink as a result. In this study, the chimney to enhance the natural convection heat transfer was applied to heat sink. The heat sink consists of heat pipe and aluminum fin, and cylindrical chimney was installed on heat sink. CPU and motherboard of PC was used for experiment, CPU temperature is inspected by digital thermocouple. Experiment shows that heat transfer of heat sink in natural convection was enhanced by chimney effect. In order to compare to the experimental result, numerical analysis was performed by CFD code. Result of numerical simulation was compared to experimental result, and it was shown that numerical analysis is reasonable. In conclusion, it was verified that the chimney improves heat transfer of heat sink in natural convection.