• Journal of Welding and Joining
• /
• 제31권3호
• /
• pp.44-48
• /
• 2013

Artificial neural network (ANN) model is applied to predict arc welding process window for automotive steel plate. Target weldment was various automotive steel plate combination with lap fillet joint. The accuracy of prediction was evaluated through comparison experimental result to ANN simulation. The effect of ANN variables on the accuracy is investigated such as number of hidden layers, perceptrons and transfer function type. A static back propagation model is established and tested. The result shows comparatively accurate predictability of the suggested ANN model. However, it restricts to use nonlinear transfer function instead of linear type and suggests only one single hidden layer rather than multiple ones to get better accuracy. In addition to this, obvious fact is affirmed again that the more perceptrons guarantee the better accuracy under the precondition that there are enough experimental database to train the neural network.

• 대한토목학회논문집
• /
• 제30권4D호
• /
• pp.387-393
• /
• 2010

본 논문에서는 아스팔트 포장의 균열 성장을 분석하기 위해서 확장유한요소법을 사용하였다. 또한 아스팔트의 점탄성 효과를 고려하기 위하여 맥스웰 체인을 이용한 점탄성 구성방정식을 사용하였으며, 균열 모델로는 선형점성균열 모델을 사용하였다. 특히 점탄성 구성방정식을 구성할 때 측정을 통해 얻어지는 온도별 변형계수와 지연시간을 Prony 급수를 이용해 재구성한 크리프 곡선을 직접 사용하지 않고 연속적인 미분이 가능한 멱승 로그 식으로 대체하여 사용하였다. 멱승 로그 식으로 완화시간 스펙트럼(relaxation spectrum)을 계산하여 맥스웰 체인의 부분탄성계수(partial moduli)를 도출하였다. 멱승 로그 적정 식을 통해 구한 맥스웰 체인의 부분 탄성계수는 크리프 곡선을 직접 이용하는 방법으로 구한 부분 탄성계수 보다 안정적인 형태의 곡선을 나타내어 해석을 용이하게 해준다. 개발된 정적균열 해석 모듈을 이용하여 아스팔트 시편의 온도별 정적균열 성장 실험 결과를 성공적으로 모사할 수 있었다.

• 한국항공우주학회지
• /
• 제34권6호
• /
• pp.83-91
• /
• 2006

현대의 고성능 전투기는 공력성능 및 조종성능의 향상을 위하여 대부분 세로축 방향으로 항공기를 불안정하게 설계하는 정안정성 완화 개념을 채택하고 있다. 비행제어법칙의 설계 작업은 불안정하게 설계된 항공기에 안정성을 부여하고, 주어진 비행임무에 대하여 만족스런 조종성능을 발휘할 수 있도록 비행성능을 조작하는 일련의 과정이다. 세로축 무게중심은 무장형상, 연료상태 및 착륙장치의 위치에 영향을 받으며 항공기 안정성에 많은 영향을 미친다. 따라서 무게중심의 이동은 세로축 안정도 여유에 영향을 미친다. 본 논문에서는 운용 시에 발생 가능한 최대 후방 무게중심에 대해 항공기 안정성을 해석하였고, 비행시험을 통해 최종적으로 검증하였다. 선형해석 항목은 세로축 단주기 모드 특성 및 안정도 여유에 관하여 행하였으며, 비선형 해석 항목은 단주기 모드를 해석하기 위해 세로축 가진 입력에 대한 항공기 응답특성을 분석하였다. 또한, 최대 후방 무게중심에서 수행된 고받음각 비행시험 자료를 제시함으로써 T-50 고등훈련기의 비행 안정성을 제시하였다.

• 한국안전학회지
• /
• 제35권5호
• /
• pp.39-48
• /
• 2020

This study proposes a fast estimation method of dynamic reliability indices or failure probability for SDOF structure subjected to earthquake excitations. The proposed estimation method attempts to derive coefficient function for correcting dynamic effects from static reliability analysis in order to estimate the dynamic reliability analysis results. For this purpose, a total of 60 cases of structures with various characteristics of natural frequency and damping ratio under various allowable limits were taken into account, and various types of approximation coefficient functions were considered as potential candidate models for dynamic effect correction. Each reliability index was computed by directly performing static and dynamic reliability analyses for the given 60 cases, and nonlinear curve fittings for potential candidate models were performed from the computed reliability index data. Then, the optimal estimation model was determined by evaluating the accuracy of the dynamic reliability analysis results estimated from each candidate model. Additional static and dynamic reliability analyses were performed for new models with different characteristics of natural frequency, damping ratio and allowable limit. From these results, the accuracy and numerical efficiency of the optimal estimation model were compared with the dynamic reliability analysis results. As a result, it was confirmed that the proposed model can be a very efficient tool of the dynamic reliability estimation for seismically excited SDOF structure since it can provide very fast and accurate reliability analysis results.

• Smart Structures and Systems
• /
• 제25권5호
• /
• pp.543-557
• /
• 2020

This paper presents a series of experimental and numerical investigations on a vertical isolation system with quasi-zero stiffness (QZS) property. The isolation system comprises a linear helical spring and disk spring. The disk spring is designed to provide variable stiffness to the system. Orthogonal static tests with different design parameters are conducted to verify the mathematical and mechanical models of the isolation system. The deviations between theoretical and test results influenced by the design parameters are summarized. Then, the dynamic tests for the systems with different under-load degrees are performed, including the fast sweeping tests, harmonic excitation tests, and half-sine impact tests. The displacement transmissibility, vibration reduction rate, and free vibration response are calculated. Based on the test results, the variation of the transmission rule is evaluated and the damping magnitudes and types are identified. In addition, the relevant numerical time history responses are calculated considering the nonlinear behavior of the system. The results indicate that the QZS isolation system has a satisfactory isolation effect, while a higher damping level can potentially promote the isolation performance in the low-frequency range. It is also proved that the numerical calculation method accurately predicts the transmission character of the isolation system.

• 전력전자학회논문지
• /
• 제3권1호
• /
• pp.77-84
• /
• 1998

본 논문에서는 비선형상태피드백을 적용한 무효전력보상기의 제어시스템을 설계하여 시뮬레이션과 축소모형 실험에 의한 성능해석에 관해 기술하였다. 무효전력 보상기의 수리모형을 3상 등가회로와 대표적인 동작점에 대한 상태방정식을 이용하여 도출하였으며, 비선형 상태피드백을 고려한 무효전력 보상기의 동특성을 표현하는 전달함수를 유도하였다. 또한 근궤적을 이용한 시스템 안정도 해석을 실시하여 제어시스템 설계를 실시하였다. 고안된 제어시스템의 성능해석은 EMTP를 이용한 시뮬레이션과 축소모형 실험을 통해 검증하였는데, 검증결과 고안된 제어시스템은 송전계통의 무효전력보상에 우수한 성능을 갖음이 확인되었다.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2000년도 하계학술대회 논문집 D
• /
• pp.2488-2491
• /
• 2000

A Neural network is possible to nonlinear function mapping and parallel processing. Therefore It has been developing for a Diagnosis system of nuclear plower plant. In general Neural Networks is a static mapping but Dynamic Neural Network(DNN) is dynamic mapping. When a fault occur in system, a state of system is changed with transient state. Because of a previous state signal is considered as a information. DNN is better suited for diagnosis systems than static neural network. But a DNN has many weights, so a real time implementation of diagnosis system is in need of a rapid network architecture. This paper presents a algorithm for RCP monitoring Alarm diagnosis system using Self Dynamic Neural Network(SDNN). SDNN has considerably fewer weights than a general DNN. Since there is no interlink among the hidden layer. The effectiveness of Alarm diagnosis system using the proposed algorithm is demonstrated by applying to RCP monitoring in Nuclear power plant.

• Structural Engineering and Mechanics
• /
• 제33권6호
• /
• pp.765-779
• /
• 2009

Quantitative condition assessment of structures has been traditionally using proof load test leading to an indication of the load-carrying capacity. Alternative approaches using ultrasonic, dynamics etc. are based on the unloaded state of the structure and anomalies may not be fully mobilized in the load resisting path and thus their effects are not fully included in the measured responses. This paper studies the effect of the load carried by a reinforced concrete beam on the assessment result of the crack damage. This assessment can only be performed with an approach based on static measurement. The crack damage is modelled as a crack zone over an area of high tensile stress of the member, and it is represented by a damage function for the simulation study. An existing nonlinear optimization algorithm is adopted. The identified damage extent from a selected high level load and a low load level are compared, and it is concluded that accurate assessment can only be obtained at a load level close to the one that creates the damage.

• 한국기계가공학회지
• /
• 제13권6호
• /
• pp.88-98
• /
• 2014

A moving mass on a skewed linear guideway model to analyze the friction-induced stick-slip behavior of ball-screw-driven slides is proposed. To describe the friction force, a friction coefficient function is modelled as a third-order polynomial of the relative velocity between the slide mass and a guideway. A nonlinear differential equation of motion is derived and an approximate solution is obtained using a perturbation method for the amplitudes and base frequencies of both pure-slip and stick-slip oscillations. The results are presented with time responses, phase plots, and amplitude plots, which are compared adequately with those obtained by Runge Kutta 4th-order numerical integration, as long as the difference between the static and kinematic friction coefficients is small. However, errors in the results by the approximate solution increase and are not negligible if the difference between the friction coefficients exceeds approximately 40% of the static friction coefficient.

• 한국정밀공학회지
• /
• 제16권12호
• /
• pp.204-213
• /
• 1999

This paper suggests a numerical method of finding optimal geometric path and minimum-time motion for a manipulator arm. To find the minimum-time motion, the optimal geometric path is searched first, and the minimum-time motion is searched on this optimal path. In the algorithm finding optimal geometric path, the objective function is minimizing the combination of joint velocities, joint-jerks, and actuator forces as well as avoiding several static obstacles, where global search is performed by adjusting the seed points of the obstacle models. In the minimum-time algorithm, the traveling time is expressed by the linear combinations of finite-term quintic B-splines and the coefficients of the splines are obtained by nonlinear programming to minimize the total traveling time subject to the constraints of the velocity-dependent actuator forces. These two search algorithms are basically similar and their convergences are quite stable.