• 한국콘텐츠학회논문지
• /
• 제4권3호
• /
• pp.99-105
• /
• 2004

이 논문은 다층퍼셉트론의 학습을 빠르게 하기 위한 최적 학습률을 제안한다. 이 학습률은 한 뉴런에 연결된 가중치들에 대한 학습률과, 중간층에 가상의 목표값을 설정하기 위한 학습률로 나타난다. 그 결과, 중간층 가중치의 최적 학습률은 가상의 중간층 목표값 할당 성분과 중간층 오차함수를 최소화 시키고자하는 성분의 곱으로 나타난다. 제안한 방법은 고립단어인식과 필기체 숫자 인식 문제의 시뮬레이션으로 효용성을 확인하였다.

• 전자공학회논문지C
• /
• 제34C권12호
• /
• pp.61-69
• /
• 1997

This paper proposes a new method for improving the performances of the neural network for optimization using a hyubrid of gradient descent method and dynamic tunneling system. The update rule of gradient descent method, which has the fast convergence characteristic, is applied for high-speed optimization. The update rule of dynamic tunneling system, which is the deterministic method with a tunneling phenomenon, is applied for global optimization. Having converged to the for escaping the local minima by applying the dynamic tunneling system. The proposed method has been applied to the travelling salesman problems and the optimal task partition problems to evaluate to that of hopfield model using the update rule of gradient descent method.

• International Journal of Air-Conditioning and Refrigeration
• /
• 제12권3호
• /
• pp.123-130
• /
• 2004

Control algorithms for an absorption air conditioning system may be developed by using dynamic models of the system. The simplified effective dynamic models, which can predict the dynamic behaviors of the system, may help to develop effective control algorithms for the system. In this study, control algorithms for an absorption air conditioning system were developed by using a dynamic simulation program. A cooling water inlet temperature control algorithm, a chilled water outlet temperature control algorithm, and a supply air temperature control algorithm, were developed and analyzed. The steepest descent method was used as an optimal algorithm. Simulation results showed energy savings and the effective controls of an absorption air conditioning system.

• 한국항공운항학회지
• /
• 제30권4호
• /
• pp.1-8
• /
• 2022

In this study, we propose trajectory-based arrival management using CDO (Continuos Descent Operations). The operational procedures with TBO (Trajectory-Based Operations) concept were established to allow aircraft and ground system to share the trajectories with each other in real time. The proposed operational concept was validated in the air traffic control simulation environment, which consists of controller working position, pseudo pilot system, air traffic generation system, and controllers' decision support system for arrival management using CDO. Simulation results compared with actual flight data indicate that proposed concept could improve the efficiency of traffic flow management in terms of total descending time and fuel consumption. And it was confirmed that if there is a system that can share and utilize the synchronized trajectory, it can be helpful to control arrival aircraft and apply CDO concept.

• 한국운동역학회지
• /
• 제16권3호
• /
• pp.65-74
• /
• 2006

This study was attempted to Kinematical characteristics of the Endo $360^{\circ}$El-grip Swing on the horizontal bar. To do this, we selected and analyzed the performance of two athletes who did Endo $360^{\circ}$El-grip Swing in the horizontal bar competition of male artistic gymnastic in the 22nd Universiad Games 2003 Daegu. We drew the conclusions from the kinematical factors that were came out through analyzing three-dimensional cinematography of the athletes' movements, by using two video cameras. In point of analyzing the actual competition situation, it is expected that gymnastics and coaches have the effective informations, and the following conclusion had resulted. 1. When performing Endo $360^{\circ}$El-Grip, the average for entire required time was $1.93{\pm}0.06sec$. The average for descent phase time was $0.24{\pm}0.02sec$, ascent phase time was 0.22${\pm}0.07sec$, connecting phase time was $0.87{\pm}0.07sec$, and El-Grip phase time was $0.61{\pm}0.02sec$. The descent phase need short period of time but however to have a stable performance, ensuring ascent and connecting phase time are needed. El-Grip phase need short period of time to have a stable re-grasp. 2. To have a convenient preparation for El-Grip in descent and ascent phase, lowering CM, and ease up in sway and plunge from the High Bar would make descent and ascent even more faster and would have increase effect in trunk rotation. 3. In descent and ascent phase, if shoulder angle and arm slope is dwindling then it would effect rotation angle so might risk it from hitting a Bar when putting legs in and out. 4. In connecting phase, it requires some time to show stable performance when El-Grip phase is continued by using hip angle which would make trunk rotation angle bigger and make descent and ascent time slower. 5. In El-Grip phase, when doing motions like hand standing. using hip angle more than maximum would make CM even faster and it is stable position while performing.

• 대한원격탐사학회지
• /
• 제38권6_1호
• /
• pp.1015-1023
• /
• 2022

착륙선의 진입-하강-착륙 과정에는 많은 환경적 및 기술적 어려움이 수반된다. 이러한 문제들을 해결하기 위한 방안으로, 최근 착륙선에는 지형상대항법 기술이 필수적으로 고려되고 있다. 지형상대항법은 하강하는 착륙선에서 수집되는 Inertial Measurement Unit (IMU) 데이터 및 영상 데이터를 기 구축된 참조 데이터와 비교하여 착륙선의 위치 및 자세를 추정하는 기술이다. 본 논문에서는 화성에서 활용할 지형상대항법 기술을 개발하기 위해 그 핵심 기술 요소로서 하강 데이터셋 생성 및 랜드마크 추출 방법을 제시한다. 제안방법은 화성착륙 시뮬레이션 궤적정보를 이용하여 하강하는 착륙선의 IMU 데이터를 생성하며, 이에 맞추어 고해상도 정사영상지도 및 수치표고모델로부터 ray tracing 기법을 통해 하강영상을 생성한다. 랜드마크 추출은 텍스쳐 정보가 부족한 화성 표면의 특성을 고려하여 영역 기반 추출 방식으로 이루어지며, 정합 정확도와 속도 향상을 위해 탐색영역 축소가 수행된다. 하강영상 생성 방법의 성능분석 결과는 제안방법으로 촬영 기하학적 조건을 만족시키는 영상 생성이 가능함을 보여주었으며, 랜드마크 추출 방법의 성능분석 결과는 제안방법을 통해 수 미터 수준의 위치 추정 정확도를 담보하면서 동시에 특징점 기반 방식만큼의 처리속도 확보가 가능함을 보여주었다.

• 한국통신학회:학술대회논문집
• /
• 한국통신학회 2006년도 하계종합학술발표회 논문 초록집
• /
• pp.564-564
• /
• 2006

• 호남수학학술지
• /
• 제36권4호
• /
• pp.755-766
• /
• 2014

In the discrete formulation of the bubble stabilized Legendre Galerkin methods, the system of equations includes the artificial viscosity term as the parameter. We investigate the estimation of this parameter to get the optimal solution which minimizes the maximum error. Some numerical results are reported.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 1999년도 하계종합학술대회 논문집
• /
• pp.1033-1036
• /
• 1999

This paper proposes an efficient fault diagnosis for digital circuits using multilayer neural networks. The efficient learning algorithm is also proposed for the multilayer neural network, which is combined the steepest descent for high-speed optimization and the dynamic tunneling for global optimization. The fault-diagnosis system using the multilayer neural network of the proposed algorithm has been applied to the parity generator circuit. The simulation results shows that the proposed system is higher convergence speed and rate, in comparision with system using the backpropagation algorithm based on the gradient descent.

• Journal of applied mathematics & informatics
• /
• 제29권3_4호
• /
• pp.603-619
• /
• 2011

To the best of our knowledge, it would probably be the first time in the literature that we clarify the relationship between Yamada's method and viscosity iteration correctly. We design iterative methods based on the hybrid steepest descent algorithms for solving variational inclusions, equilibrium problems. Our results unify, extend and improve the corresponding results given by many others.