• Earthquakes and Structures
• /
• 제3권1호
• /
• pp.17-35
• /
• 2012

Simplified equations for fundamental period of vibration of skeletal structures provided by most seismic design provisions suffer from the absence of any associated confidence levels and of any reference to their empirical basis. Therefore, such equations may typically give a sector of designers the false impression of yielding a fairly accurate value of the period of vibration. This paper, although not addressing simplified codes equations, introduces a set of mathematical equations utilizing the theory of error propagation and First-Order Second-Moment (FOSM) techniques to determine bounds on the relative error in theoretically calculated fundamental period of vibration of skeletal structures. In a complementary step, and for verification purposes, Monte Carlo simulation technique has been also applied. The latter, despite involving larger computational effort, is expected to provide more precise estimates than FOSM methods. Studies of parametric uncertainties applied to reinforced concrete frame bents - potentially idealized as SDOF systems - are conducted demonstrating the effect of randomness and uncertainty of various relevant properties, shaping both mass and stiffness, on the variance (i.e. relative error) in the estimated period of vibration. Correlation between mass and stiffness parameters - regarded as random variables - is also thoroughly discussed. According to achieved results, a relative error in the period of vibration in the order of 19% for new designs/constructions and of about 25% for existing structures for assessment purposes - and even climbing up to about 36% in some special applications and/or circumstances - is acknowledged when adopting estimates gathered from the literature for relative errors in the relevant random input variables.

• 한국지반공학회지:지반
• /
• 제13권2호
• /
• pp.77-90
• /
• 1997

지하수 거동에 대한 불확실성을 극복하기 위해서 암반 지반의 투수계수를 예측할 수 있는 신뢰적이고 경제적인 방법이 필요하다. 이러한 목적을 위하여 암반의 투수계수 예측 방법에 대한 연구가 수행되어졌다. 인공 신경망 이론을 적용한 투수계수 예측 방법에 대한 일환으로 오차역 전파 학습알고리즘을 이용한 투수계수 예측 방법에 대하여 연구를 수행하였으며, 이 방법의 타당성 검토를 위하여 현장투수시험 결과와 지반물성치들에 적용하여 검증을 실시하였다. 검증결과 평균오차 범위가 작아 비교적 정착한 투수계수 예측방법임을 보여주었다.

• 대한화학회지
• /
• 제47권1호
• /
• pp.31-37
• /
• 2003

측정 및 화학분석에 많이 이용되는 한 점 교정식에 대하여 측정 불확도 표기 지침서(GUM)의 근사법과 Monte-Carlo Simulation에 의해 계산된 각각의 확장불확도를 비교하였다. 이 비교를 위하여 임의의 자료들을 여러 농도 수준에서 정규 분포 또는 t-분포로 가정하여 계산하였다. 나눗셈에 의한 한 점 교정식의 비선형성과 t-분포 형식을 함에 따른 입력량의 과도한 퍼짐으로 인하여, 경우에 따라서, GUM의 근사법으로 계산된 불확도가 Monte-Carlo Simulation에 의해 계산된 것보다 약 50% 이상의 오류가 있다는 것이 확인되었다. 그러나, 검출 하한을 계산하기 위하여 한 점 교정식을 이용하는 경우, 반응량의 표준불확도가 상대적으로 매우 크고 비선형성에 희한 계산 오류가 상대적으로 무시되므로 근사식에 따른 계산 오류가 발생하지 않았다.

• 한국수자원학회:학술대회논문집
• /
• 한국수자원학회 2012년도 학술발표회
• /
• pp.39-39
• /
• 2012

Data assimilation techniques have received growing attention due to their capability to improve prediction in various areas. Despite of their potentials, applicable software frameworks to probabilistic approaches and data assimilation are still limited because the most of hydrologic modelling software are based on a deterministic approach. In this study, we developed a hydrological modelling framework for sequential data assimilation, namely MPI-OHyMoS. MPI-OHyMoS allows user to develop his/her own element models and to easily build a total simulation system model for hydrological simulations. Unlike process-based modelling framework, this software framework benefits from its object-oriented feature to flexibly represent hydrological processes without any change of the main library. In this software framework, sequential data assimilation based on the particle filters is available for any hydrologic models considering various sources of uncertainty originated from input forcing, parameters and observations. The particle filters are a Bayesian learning process in which the propagation of all uncertainties is carried out by a suitable selection of randomly generated particles without any assumptions about the nature of the distributions. In MPI-OHyMoS, ensemble simulations are parallelized, which can take advantage of high performance computing (HPC) system. We applied this software framework for several catchments in Japan using a distributed hydrologic model. Uncertainty of model parameters and radar rainfall estimates is assessed simultaneously in sequential data assimilation.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1997년도 한국자동제어학술회의논문집; 한국전력공사 서울연수원; 17-18 Oct. 1997
• /
• pp.375-378
• /
• 1997

This research deals with a problem of reconstructing 3D surface structures from their 2D projections, which is an important research topic in computer vision. In order to provide robust reconstruction algorithm, that is reliable even in the presence of uncertainty in the range images, we first present a detailed model and analysis of several error sources and their effects on measuring three-dimensional surface properties using the space encoded range imaging technique. Our approach has two key elements. The first is the error modeling for the space encoding range sensor and its propagation to the 3D surface reconstruction problem. The second key element in our approach is the algorithm for removing outliers in the range image. Such analyses, to our knowledge, have never attempted before. Experimental results show that our approach is significantly reliable.

• 제어로봇시스템학회논문지
• /
• 제12권3호
• /
• pp.201-207
• /
• 2006

A robust Kalman filtering method for uncertain stochastic systems is suggested by adopting a perturbation estimation process which is to reconstruct total uncertainty with respect to the nominal state transition equation. The predictor and corrector of discrete Kalman filter are reformulated with the perturbation estimator. Successively, the state and perturbation estimation error dynamics and the corresponding error covariance propagation equations are derived as well. Finally we have the recursive algorithm of Combined Kalman Filter-Perturbation Estimator (CKF). The proposed combined Kalman filter-perturbation estimator has the property of integrating innovations and the adaptation capability to system uncertainties. A numerical example is shown to demonstrate the effectiveness of the proposed scheme.

• 제어로봇시스템학회논문지
• /
• 제2권4호
• /
• pp.312-317
• /
• 1996

This paper proposes a neural network controller with the computed torque method. The neural network is used not to learn the inverse dynamic model but to compensate the uncertainties of robotic manipulators. When training the neural network, we use the signals present in the proposed controller, which is simpler than that proposed by Ishiguro et al., whose teaching signals of the neural network come from the robot model.

• 대한기계학회논문집
• /
• 제17권12호
• /
• pp.3124-3134
• /
• 1993

Diagnosis is composed of two different but interrelated steps ; retrieving the sensory responses f the system and reasoning the state of the system through the given sensor data. This paper explains the probabilistic nature of reasoning involved in the diagnosis when the uncertainties are inevitably included in experts' diagnostic decision making. Uncertainties in decision making are experts' personal experiences, preferences, and system's coherent characteristics. In order to ensure a consistent decision based on the same responses from the system, expert system technology is adopted with the Bayesian reasoning scheme.

• Smart Structures and Systems
• /
• 제13권1호
• /
• pp.155-171
• /
• 2014

Various monitoring techniques are now available for structural health monitoring and Acoustic Emission (AE) is one of them. One of the major advantages of the AE technique is its capability to locate active cracks in structural members. AE crack locating approaches are affected by the signal attenuation and dispersion of elastic waves due to inhomogeneity and geometry of reinforced concrete (RC) members. In this paper, a novel technique is described based on signal processing and sensor arrangement to process multisensory AE data generated by the onset and propagation of cracks and is validated with experimental results from an in-situ load test. Considering the sources of uncertainty in the AE crack location process, a methodology is proposed to capture and locate events generated by cracks. In particular, the relationship between AE events and load is analyzed, and the feasibility of using the AE technique to evaluate the cracking behavior of two RC slab strips during loading to failure is studied.

• 제어로봇시스템학회논문지
• /
• 제6권3호
• /
• pp.256-262
• /
• 2000

This paper proposes a new sliding mode controller combined with a multi-layer neural network using the error back propagation learning algorithm,, The network acts as a compensator of the conventional sliding mode controller to improve the control performance when initial assumptions of uncertainty bounds of system parameters are violated. The proposed controller can reduce th steady state error of conventional sliding mode controller with the boundary layer technique Computer simulation results show that the proposed method is effective to control dynamic systems with unexpectably large uncertainties.