• 대한기계학회논문집A
• /
• 제30권9호
• /
• pp.1116-1123
• /
• 2006

In this study, the effective method for reliability estimation is proposed using tow-staged kriging metamodel and genetic algorithm. Kriging metamodel can be determined by appropriate sampling range and the number of sampling points. The first kriging metamodel is made based on the proposed sampling points. The advanced f'=rst order reliability method is applied to the first kriging metamodel to determine the reliability and most probable failure point(MPFP) approximately. Then, the second kriging metamodel is constructed using additional sampling points near the MPFP. These points are selected using genetic algorithm that have the maximum mean squared error. The Monte-Carlo simulation is applied to the second kriging metamodel to estimate the reliability. The proposed method is applied to numerical examples and the results are almost equal to the reference reliability.

• 대한전기학회논문지:전력기술부문A
• /
• 제55권11호
• /
• pp.453-462
• /
• 2006

This paper evaluates the reliability of KEPCO system using TRELSS, which is a probabilistic reliability evaluation program for large-scaled power system. In order to reflect the characteristic of KEPCO system, the sensitivity of reliability indices such as LOLP, EDLC, EENS and Energy Curtailment for variations of TRELSS parameter and input data was analyzed. Additionally, probabilistic reliability of KEPCO system reflecting sensitivity analysis results was systematically evaluated and simulated. Finally, maximum acceptable FOR of KEPCO system to satisfy reliability criterion, which meet in process of establishing the basic plan for long-term electricity supply and demand is suggested.

• Communications for Statistical Applications and Methods
• /
• 제1권1호
• /
• pp.10-19
• /
• 1994

We study the change-point problem in the context of system reliability models. The maximum likelihood estimators are obtained based on the Jelinski and Moranda model. To find the approximate distribution of the change-point estimator, we suggest of parametric bootstrap method in which the estimators are substituted in the assumed model. Through an example we illustrate the proposed method.

• 품질경영학회지
• /
• 제27권1호
• /
• pp.80-90
• /
• 1999

In this paper, reliability estimation using Gibbs sampler is considered for the mixture model with Gamma family, Gibbs sampler is derived to compute the features for the posterior distribution. By simulation study, the maximum likelihood estimator and the Gibbs estimator are obtained. A numerical study with a simulated data is provided.

• 품질경영학회지
• /
• 제13권1호
• /
• pp.31-40
• /
• 1985

In this paper, statistical estimation of the parameters of the bivariate exponential distribution are studied. Bayes estimators of the parameters are obtained and compared with the maximum likelihood estimators which are introduced by Freund. We know that the method of moments estimators coincide with the maximum likelihood estimators and Bayes estimators are more efficient than the maximum likelihood estimators in moderate samples. The asymptotic distributions of the maximum likelihood estimators and the estimator of mean time to system failure are obtained.

• 한국항공우주학회지
• /
• 제47권10호
• /
• pp.747-752
• /
• 2019

신뢰성 성장 시험을 수행하며 획득하게 되는 고장 정보와 누적 시험수행시간을 이용하면 신뢰성 성장 모델의 모수 추정이 가능하며, 모수 추정을 통해 해당 제품의 MTBF를 예측할 수 있다. 그러나 시험에 대한 비용, 시간 혹은 제품의 특성 등의 여러 제약으로 인해 고장 정보가 구간적으로 획득되거나, 획득한 고장 정보의 샘플 데이터(Sample Data)의 수가 작을 수 있다. 이는 신뢰성 성장 모델의 모수 추정의 오차를 커지게 하는 원인이 될 수 있다. 본 논문에서는 샘플 데이터의 수가 작을 경우 신뢰성 성장 모델의 모수 추정 시 베이지안 기법 기반의 모수 추정 방법의 적용에 대해 연구를 수행하였다. 시뮬레이션 결과 신뢰성 성장 모델의 모수를 추정할 때, MLE를 적용하여 추정하는 방법보다 베이지안 기법을 적용하는 방법이 추정 정확도가 높음을 확인하였다.

• 대한토목학회논문집
• /
• 제30권5D호
• /
• pp.497-504
• /
• 2010

본 연구에서는 최근 10년간의 수원과 의정부지역의 국도 포장구간의 수명 데이터를 기반으로 신뢰성 개념을 도입하여 포장 수명의 고장확률, 신뢰도, 평균 수명 등의 정보를 산정하는 방안을 제시하였다. 모수적 추론 방법인 최우추정법과 확률지를 이용하여 최적 수명분포형태가 대수정규분포임을 확인하였으며, 본 연구에서 제안한 대수정규수명분포에 의해 추정된 누적 고장확률이 실제 관측값과 차이가 거의 없음도 확인하였다. 나아가 추정된 모수를 기반으로 신설 포장과 덧씌우기 포장의 하중별 평균 수명을 산정한 결과, 각각 6.5년~7.9년과 7.3년~9.1년으로 추정되었으며 상대적으로 덧씌우기 구간의 평균수명이 더 길게 추정되었다. 본 연구에서 제안한 신뢰성 개념을 이용한 분석 방법은 포장관리 및 유지보수 데이터가 축적되어감에 따라 비교적 용이하게 계속 보완이 가능하며 보다 정확한 포장수명에 대한 신뢰도 값에 접근해 갈 수 있을 것으로 기대된다.

• Journal of Electrical Engineering and Technology
• /
• 제12권4호
• /
• pp.1397-1405
• /
• 2017

As the introduction of wind power is steadily increasing, negative effects of wind power become more important. To operate a power system more reliable, the system operator needs to recognize the maximum required capacity of available generators for a certain period. For recognizing the maximum capacity, this paper proposes a methodology to determine an optimal reserve requirement considering wind power, for the certain period in the mid-term perspective. As wind speed is predicted earlier, the difference of the forecasted and the actual wind speed becomes greater. All possible forecast errors should be considered in determining optimal reserve, and they are represented explicitly by the proposed matrix form in this paper. In addition, impacts of the generator failure are also analyzed using the matrix form. Through three main stages which are the scheduling, contingency and evaluation stages, costs associated with power generation, reserve procurement and the usage, and the reliability cost are calculated. The optimal reserve requirement is determined so as to minimize the sum of these costs based on the cost/reliability analysis. In case study, it is performed to analyze the impact of wind power penetration on the reserve requirement, and how major factors affect it.

• 상하수도학회지
• /
• 제24권5호
• /
• pp.509-517
• /
• 2010

Statistical distribution of annual maximum rainfall intensity of 18 cities in Korea was analyzed and applied to the reliability model which can calculate the probability of performance failure of storm sewer. After the analysis, it was found that distribution of annual maximum rainfall intensity of 18 cities in Korea is well matched with Gumbel distribution. Rational equation was used to estimate the load and Manning's equation was used to estimate the capacity in reliability function to calculate the probability of performance failure of storm sewer. Reliability analysis was performed by developed model applying to the real storm sewer. It was found that probability of performance failure is abruptly increased if the diameter is smaller than certain size. Therefore, cleaning the inside of storm sewer to maintain the original diameter can be one of the best ways to reduce the probability of performance failure. In the present study, probability of performance failure according to accumulation of debris in storm sewer was calculated. It was found that increasing the amount of debris seriously decrease the capacity of storm sewer and significantly increase the probability of performance failure.

• Journal of the Korean Data and Information Science Society
• /
• 제14권4호
• /
• pp.837-844
• /
• 2003

In this paper, we obtain the estimator of system reliability for the bivariate Pareto model with bivariate type 1 censored data. We obtain the estimators and approximated confidence intervals of the reliability for the parallel system based on likelihood function and the relative frequency, respectively. Also we present a numerical example by giving a data set which is generated by computer.