• 한국전산구조공학회:학술대회논문집
• /
• 한국전산구조공학회 1997년도 가을 학술발표회 논문집
• /
• pp.134-141
• /
• 1997

A geometrically non-linear finite element formulation of spatial cable networks is presented using three cable elements. Firstly, derivation procedures of tangent stiffness and mass matrices for the space truss element and the elastic catenary cable element, and the isoparametric cable element are summarized. The load incremental method based on Newton-Raphson iteration method and the dynamic relaxation method are presented in order to determine the initial static state of cable nets subjected to self-weights and support motions. Furthermore, static non-linear analysis of cable structures under additional live loads are performed based on the initial configuration. Challenging example problems are presented and discussed in order to demonstrate the feasibility of the present finite element method and investigate static non-linear behaviors of cable nets.

• Macromolecular Research
• /
• 제15권2호
• /
• pp.109-113
• /
• 2007

The surface molecular motion of monodisperse polystyrene (PS) films was examined using scanning vis-coelasticity microscopy (SVM) in conjunction with lateral force microscopy (LFM). The dynamic storage modulus, E', and loss tangent, $tan\delta$, at a PS film surface with number-average molecular weights, $M_n$, smaller than 30 k were found to be smaller and larger than those for the bulk sample, even at room temperature, meaning that the PS surface is in a glass-rubber transition or fully rubbery sate at this temperature when the $M_n$ is small. In order to quantitatively elucidate the dynamics of the molecular motion at the PS surface, SVM and LFM measurements were performed at various temperatures. The glass transition temperature, $T_g$, at the surface was found to be markedly lower than the bulk $T_g$, and this discrepancy between the surface and bulk became larger with decreasing $M_n$. Such an intensive activation of the thermal molecular motion at the PS surfaces can be explained in terms of an excess free volume in the vicinity of the film surface induced by the preferential segregation of the chain end groups.

• 한국농공학회지
• /
• 제42권3호
• /
• pp.107-113
• /
• 2000

This study is performed to examine the physical and mechanical properties of concrete with fly ash. Test results show that the unit weights of concrete with fly ash are decreased 1-3% and the highest strength is achieved by 10% filled fly ash concrete with it is increased 7% than that of the normal cement concrete. the ultrasonic pulse velocity is in the range of 3.705~4.204m/s and the dynamic and static modulus of elasticity is in the range of 271$\times$103 ~289$\times$103kgf/cm2 and 208$\times$103 ~262$\times$103kgf/cm2 respectively. The acid-resistance is increased with increase of the content of fly ash. It is 1.2 times of the normal cement concrete by 10% filled fly ash concrete and 1.7 times by 30% filled fly ash concrete respectively.

• ETRI Journal
• /
• 제38권6호
• /
• pp.1190-1196
• /
• 2016

We propose a new bandpass filter (BPF)-based online channel normalization method to dynamically suppress channel distortion when the speech and channel noise components are unknown. In this method, an adaptive modulation frequency filter is used to perform channel normalization, whereas conventional modulation filtering methods apply the same filter form to each utterance. In this paper, we only normalize the two mel frequency cepstral coefficients (C0 and C1) with large dynamic ranges; the computational complexity is thus decreased, and channel normalization accuracy is improved. Additionally, to update the filter weights dynamically, we normalize the learning rates using the dimensional power of each frame. Our speech recognition experiments using the proposed BPF-based blind channel normalization method show that this approach effectively removes channel distortion and results in only a minor decline in accuracy when online channel normalization processing is used instead of batch processing

• 한국공작기계학회:학술대회논문집
• /
• 한국공작기계학회 1999년도 춘계학술대회 논문집
• /
• pp.210-215
• /
• 1999

In this paper, a new approach to the dynamic control technique for track vehicle system using fuzzy-neural network control technique is proposed. The proposed control scheme uses a Gaussian function as a unit function in the neural network-fuzzy, and back propagation algorithm to train the fuzzy-neural network controller in the framework of the specialized learning architecture. It is proposed a learning controller consisting of two neural network-fuzzy based on independent reasoning and a connection net with fixed weights to simply the neural networks-fuzzy. The performance of the proposed controller is shown by simulation for trajectory tracking of the speed and azimuth of a track vehicle.

• 한국해양공학회지
• /
• 제8권2호
• /
• pp.131-140
• /
• 1994

Dynamics of AUV has heavy nonlinearities and many unknown parameters due to its bluff shape and low cruising speed. Intelligent algorithms, therefore, are required to overcome these nonlinearities and unknown system dynamics. Several identification techniques have been suggested for the application of control of underwater vehicles during last decade. This paper applies the neural network to identification and motion control problem of AUVs. Nonlinear dynamic systems of an AUV are identified using feedforward neural network. Simulation results show that the learned neural network can generate the motion of AUV. This paper, also, suggest an adaptive control scheme up-dates the controller weights with reference model and feedforward neural network using error back propagation.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2006년도 추계학술발표대회
• /
• pp.35-38
• /
• 2006

다층 퍼셉트론(MLP) 학습 이론인 오류 역전파 알고리즘은 델타룰과 최급 하강법을 사용하기 때문에 학습시 많은 시간이 소요된다는 단점을 가지고 있다. 때문에 신경망에서의 잘못된 초기 가중치 선택은 오류 역전파 알고리즘을 사용하는 신경망에서의 현격한 학습 성능저하를 발생시키게 된다. 본 논문에서는 학습시 오류 역전파 알고리즘의 수렴시간을 개선하기 위한 신경망의 동적 초기 가중치 선택 알고리즘을 제안한다. 이 알고리즘은 학습전 기존의 선택 가중치와 모든 가중치가 1.0 또는 -1.0 값을 가지는 가중치 집합에서 가중치 변동률을 선측정하여 이들 중 가장 변동률이 큰 경우를 초기 가중치 집합으로 선정하게 된다. 즉, 초기의 가중치 변동률을 차후 성능을 판단하는 지표로 사용하여 잘못된 가중치 선택으로 인한 최악의 학습효율의 가능성을 배제시키고 다층 신경망의 학습특성상 평균 이상의 학습효율을 보장하는 초기 가중치 선택방법이다.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2003년도 하계학술대회 논문집 D
• /
• pp.2414-2416
• /
• 2003

In this raper, we present a Wavelet Neural Network(WNN) approach to the solution of the tracking problem for mobile robots that possess complexity, nonlinearity and uncertainty. The neural network is constructed by the wavelet orthogonal decomposition to form a wavelet neural network that can overcome the problems caused by local minima of optimization and various uncertainties. This network structure is helpful to determine the number of the hidden nodes and the initial value of weights with compact structure. In our control method, the control signals are directly obtained by minimizing the difference between the reference track and the pose of a mobile robot that is controlled through a wavelet neural network. The control process is a dynamic on-line process that uses the wavelet neural network trained by the gradient-descent method. Through computer simulations, we demonstrate the effectiveness and feasibility of the proposed control method.

• 한국시스템다이내믹스연구
• /
• 제6권1호
• /
• pp.177-192
• /
• 2005

A system dynamics model is developed to investigate policies for the school-age child weights and heights. The model is based upon the system dynamic model of Soonhee lee(2003), the purpose of which was policy analysis for obese control of adult people. Although the purpose and target people are different, the main structure can be applied to in the same way. Besides the carbohydrate, protein, and fat material, the new model covers hormone and heights with new input mechanisms for foods and activities. The simulation results matches well with the average school-age children in Korea with the average inputs data (foods and activities). The model can be used for various purposes such as policy analyses, care plan for obese children, etc.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2001년도 하계학술대회 논문집 D
• /
• pp.2320-2322
• /
• 2001

This paper describes the design of a robust adaptive neural-net(NN) observer for uncertain nonlinear dynamical system. The Lyapunov synthesis approach is used to guarantee a uniform ultimate boundedness property of the state estimation error, as well as of all other signals in the closed-loop system. Especially, for reducing the dynamic oder of the observer, we propose a new method in which no strictly positive real(SPR) condition is needed with on-line estimation of weights of the NNs. No a priori knowledge of an upper bounds on the uncertain terms is required. The theoretical results are illustrated through a simulation example.