• Title/Summary/Keyword: Complex Failure Prediction

Search Result 36, Processing Time 0.038 seconds

Bayesian reliability prediction under event tree (Event tree하에서 베이지안 기법을 이용한 신뢰도 예측)

  • 박철순;전치혁;양희중;장수영
    • Proceedings of the Korean Operations and Management Science Society Conference
    • /
    • 1993.10a
    • /
    • pp.24-30
    • /
    • 1993
  • When modeling a complex system we use an event tree to analyze propagation of failure. An event tree cannot represent the statistical interrelationships among parameters, but it can be represented as a statistically identical influence diagram so that parameter updating can be easily performed. After updating parameters we can calculate posterior distribution of the failure rate for each path. But exact distribution requires considerably complex numerical integration. We propose an approximation method to calculate the posterior and derive the predictive distribution of the time to next failure. Finally we introduce the system which implements our methodology.

  • PDF

Development of Downstream Flood Damage Prediction Model Based on Probability of Failure Analysis in Agricultural Reservoir (3차원 수리모형을 이용한 농업용 저수지의 파괴확률에 따른 하류부 피해예측 모델 개발)

  • Jeon, Jeong Bae;Yoon, Seong Soo;Choi, Won
    • Journal of The Korean Society of Agricultural Engineers
    • /
    • v.62 no.3
    • /
    • pp.95-107
    • /
    • 2020
  • The failures of the agricultural reservoirs that most have more than 50 years, have increased due to the abnormal weather and localized heavy rains. There are many studies on the prediction of damage from reservoir collapse, however, these referenced studies focused on evaluating reservoir collapse as single unit and applyed to one and two dimensional hydrodynamic model to identify the fluid flow. This study is to estimate failure probability of spillway, sliding, bearing capacity and overflowing targeting small and medium scale agricultural reservoirs. In addition, we calculate failure probability by complex mode. Moreover, we predict downstream flood damage by reservoir failure applying three dimensional hydrodynamic model. When the reservoir destroyed, the results are as follows; (1) the flow of fluid proceeds to same stream direction and to a lower slope by potential and kinetic energy; (2) The predicted damage in downstream is evaluated that damage due to building destruction is the highest.

Risk Assessment for the Failure of an Arch Bridge System Based upon Response Surface Method(I): Component Reliability (응답면 기법에 의한 아치교량 시스템의 붕괴 위험성평가(I): 요소신뢰성)

  • Cho, Tae-Jun;Bang, Myung-Seok
    • Journal of the Korean Society of Safety
    • /
    • v.21 no.6 s.78
    • /
    • pp.74-81
    • /
    • 2006
  • Probabilistic Risk Assessment considering statistically random variables is performed for the preliminary design of a Arch Bridge. Component reliabilities of girders have been evaluated using the response surfaces of the design variables at the selected critical sections based on the maximum shear and negative moment locations. Response Surface Method(RSM) is successfully applied for reliability analyses for this relatively small probability of failure of the complex structure, which is hard to be obtained by Monte-Carlo Simulations or by First Order Second Moment Method that can not easily calculate the derivative terms of implicit limit state functions. For the analysis of system reliability, parallel resistance system composed of girders is changed into parallel series connection system. The upper and lower probabilities of failure for the structural system have been evaluated and compared with the suggested prediction method for the combination of failure modes. The suggested prediction method for the combination of failure modes reveals the unexpected combinations of element failures in significantly reduced time and efforts compared with the previous permutation method or system reliability analysis method.

A Study on the Method of Combining Empirical Data and Deterministic Model for Fuel Failure Prediction (핵연료 파손 예측을 위한 경험적 자료와 결정론적 모델의 접합 방법)

  • Cho, Byeong-Ho;Yoon, Young-Ku;Chang, Soon-Heung
    • Nuclear Engineering and Technology
    • /
    • v.19 no.4
    • /
    • pp.233-241
    • /
    • 1987
  • Difficulties are encountered when the behavior of complex systems (i.e., fuel failure probability) that have unreliable deterministic models is predicted. For more realistic prediction of the behavior of complex systems with limited observational data, the present study was undertaken to devise an approach of combining predictions from the deterministic model and actual observational data. Predictions by this method of combining are inferred to be of higher reliability than separate predictions made by either model taken independently. A systematic method of hierarchical pattern discovery based on the method developed in the SPEAR was used for systematic search of weighting factors and pattern boundaries for the present method. A sample calculation was performed for prediction of CANDU fuel failures that had occurred due to power ramp during refuelling process. It was demonstrated by this sample calculation that there exists a region of feature space in which fuel failure probability from the PROFIT model nearly agree with that from observational data.

  • PDF

Risk Assessment for a Bridge System Based upon Response Surface Method Compared with System Reliability (체계신뢰성 평가와 비교한 응답면기법에 의한 교량시스템의 위험성평가)

  • Cho, Tae-Jun;Moon, Jae-Woo;Kim, Jong-Tae
    • Proceedings of the Computational Structural Engineering Institute Conference
    • /
    • 2007.04a
    • /
    • pp.295-300
    • /
    • 2007
  • Probabilistic Risk Assessment considering statistically random variables is performed for the preliminary design of a Arch Bridge. Component reliabilities of girders have been evaluated using the response surfaces of the design variables at the selected critical sections based on the maximum shear and negative moment locations. Response Surface Method (RSM) is successfully applied for reliability analyses for this relatively small probability of failure of the complex structure, which is hard to be obtained by Monte-Carlo Simulations or by First Order Second Moment Method that can not easily calculate the derivative terms of implicit limit state functions. For the analysis of system reliability, parallel resistance system composed of girders is changed into parallel series connection system. The upper and lower probabilities of failure for the structural system have been evaluated and compared with the suggested prediction method for the combination of failure modes. The suggested prediction method for the combination of failure modes reveals the unexpected combinations of element failures in significant]y reduced time and efforts compared with the previous permutation method or system reliability analysis method.

  • PDF

Prediction of Dynamic Expected Time to System Failure

  • Oh, Deog-Yeon;Lee, Chong-Chul
    • Proceedings of the Korean Nuclear Society Conference
    • /
    • 1997.10a
    • /
    • pp.244-250
    • /
    • 1997
  • The mean time to failure (MTTF) expressing the mean value of the system life is a measure of system effectiveness. To estimate the remaining life of component and/or system, the dynamic mean time to failure concept is suggested. It is the time-dependent Property depending on the status of components. The Kalman filter is used to estimate the reliability of components using the on-line information (directly measured sensor output or device-specific diagnostics in the intelligent sensor) in form of the numerical value (state factor). This factor considers the persistency of the fault condition and confidence level in measurement. If there is a complex system with many components, each calculated reliability's or components are combined, which results in the dynamic MTTF or system. The illustrative examples are discussed. The results show that the dynamic MTTF can well express the component and system failure behaviour whether any kinds of failure are occurred or not.

  • PDF

Development of a complex failure prediction system using Hierarchical Attention Network (Hierarchical Attention Network를 이용한 복합 장애 발생 예측 시스템 개발)

  • Park, Youngchan;An, Sangjun;Kim, Mintae;Kim, Wooju
    • Journal of Intelligence and Information Systems
    • /
    • v.26 no.4
    • /
    • pp.127-148
    • /
    • 2020
  • The data center is a physical environment facility for accommodating computer systems and related components, and is an essential foundation technology for next-generation core industries such as big data, smart factories, wearables, and smart homes. In particular, with the growth of cloud computing, the proportional expansion of the data center infrastructure is inevitable. Monitoring the health of these data center facilities is a way to maintain and manage the system and prevent failure. If a failure occurs in some elements of the facility, it may affect not only the relevant equipment but also other connected equipment, and may cause enormous damage. In particular, IT facilities are irregular due to interdependence and it is difficult to know the cause. In the previous study predicting failure in data center, failure was predicted by looking at a single server as a single state without assuming that the devices were mixed. Therefore, in this study, data center failures were classified into failures occurring inside the server (Outage A) and failures occurring outside the server (Outage B), and focused on analyzing complex failures occurring within the server. Server external failures include power, cooling, user errors, etc. Since such failures can be prevented in the early stages of data center facility construction, various solutions are being developed. On the other hand, the cause of the failure occurring in the server is difficult to determine, and adequate prevention has not yet been achieved. In particular, this is the reason why server failures do not occur singularly, cause other server failures, or receive something that causes failures from other servers. In other words, while the existing studies assumed that it was a single server that did not affect the servers and analyzed the failure, in this study, the failure occurred on the assumption that it had an effect between servers. In order to define the complex failure situation in the data center, failure history data for each equipment existing in the data center was used. There are four major failures considered in this study: Network Node Down, Server Down, Windows Activation Services Down, and Database Management System Service Down. The failures that occur for each device are sorted in chronological order, and when a failure occurs in a specific equipment, if a failure occurs in a specific equipment within 5 minutes from the time of occurrence, it is defined that the failure occurs simultaneously. After configuring the sequence for the devices that have failed at the same time, 5 devices that frequently occur simultaneously within the configured sequence were selected, and the case where the selected devices failed at the same time was confirmed through visualization. Since the server resource information collected for failure analysis is in units of time series and has flow, we used Long Short-term Memory (LSTM), a deep learning algorithm that can predict the next state through the previous state. In addition, unlike a single server, the Hierarchical Attention Network deep learning model structure was used in consideration of the fact that the level of multiple failures for each server is different. This algorithm is a method of increasing the prediction accuracy by giving weight to the server as the impact on the failure increases. The study began with defining the type of failure and selecting the analysis target. In the first experiment, the same collected data was assumed as a single server state and a multiple server state, and compared and analyzed. The second experiment improved the prediction accuracy in the case of a complex server by optimizing each server threshold. In the first experiment, which assumed each of a single server and multiple servers, in the case of a single server, it was predicted that three of the five servers did not have a failure even though the actual failure occurred. However, assuming multiple servers, all five servers were predicted to have failed. As a result of the experiment, the hypothesis that there is an effect between servers is proven. As a result of this study, it was confirmed that the prediction performance was superior when the multiple servers were assumed than when the single server was assumed. In particular, applying the Hierarchical Attention Network algorithm, assuming that the effects of each server will be different, played a role in improving the analysis effect. In addition, by applying a different threshold for each server, the prediction accuracy could be improved. This study showed that failures that are difficult to determine the cause can be predicted through historical data, and a model that can predict failures occurring in servers in data centers is presented. It is expected that the occurrence of disability can be prevented in advance using the results of this study.

A Basic Study on Sale Price Prediction Model of Apartment Building Projects using Machine Learning Technique (머신러닝 기반 공동주택 분양가 예측모델 개발 기초연구)

  • Son, Seung-Hyun;Kim, Ji-Myong;Han, Bum-Jin;Na, Young-Ju;Kim, Tae-Hee
    • Proceedings of the Korean Institute of Building Construction Conference
    • /
    • 2021.05a
    • /
    • pp.151-152
    • /
    • 2021
  • The sale price of apartment buildings is a key factor in the success or failure of apartment projects, and the factors that affect the sale price of apartments vary widely, including location, environmental factors, and economic conditions. Existing methods of predicting the sale price do not reflect the nonlinear characteristics of apartment prices, which are determined by the complex impact factors of reality, because statistical analysis is conducted under the assumption of a linear model. To improve these problems, a new analysis technique is needed to predict apartment sales prices by complex nonlinear influencing factors. Using machine learning techniques that have recently attracted attention in the field of engineering, it is possible to predict the sale price reflecting the complexity of various factors. Therefore, this study aims to conduct a basic study for the development of a machine learning-based prediction model for apartment sale prices.

  • PDF

Probabilistic Risk Assessment of a Cable-Stayed Bridge Based on the Prediction Method for the Combination of Failure Modes (붕괴모드 조합 예측법에 의한 PSC사장교의 위험도평가)

  • Park, Mi-Yun;Cho, Hyo-Nam;Cho, Taejun
    • KSCE Journal of Civil and Environmental Engineering Research
    • /
    • v.26 no.4A
    • /
    • pp.647-657
    • /
    • 2006
  • Probabilistic Risk Assessment considering statistically random variables is performed for the preliminary design of a Cable Stayed Bridge, which is Prestressed Concrete Bridge consisted of cable and plate girders, based on the method of Working Stress Design and Strength Design. Component reliabilities of cables and girders have been evaluated using the response surface of the design variables at the selected critical sections based on the maximum shear, positive and negative moment locations. Response Surface Method (RSM) is successfully applied for reliability analyses for this relatively small probability of failure of the complex structure, which is hard to obtain through Monte-Carlo Simulations. or through First Order Second Moment Method that can not easily calculate the derivative terms of implicit limit state functions. For the analysis of system reliability, parallel resistance system consisting of cables and plate girder is changed into series connection system and the result of system reliability of total structure is presented. As a system reliability, the upper and lower probabilities of failure for the structural system have been evaluated and compared with the suggested prediction method for the combination of failure modes. The suggested prediction method for the combination of failure modes reveals the unexpected combinations of element failures in significantly reduced time and efforts compared with the previous permutation method or system reliability analysis method, which calculates upper and lower bound failure probabilities.

The Prediction of Failure Probability of Bridges using Monte Carlo Simulation and Lifetime Functions (몬테칼로법과 생애함수를 이용한 교량의 파괴확률예측)

  • Seung-Ie Yang
    • Journal of the Korean Society of Safety
    • /
    • v.18 no.1
    • /
    • pp.116-122
    • /
    • 2003
  • Monte Carlo method is one of the powerful engineering tools especially to solve the complex non-linear problems. The Monte Carlo method gives approximate solution to a variety of mathematical problems by performing statistical sampling experiments on a computer. One of the methods to predict the time dependent failure probability of one of the bridge components or the bridge system is a lifetime function. In this paper, FORTRAN program is developed to predict the failure probability of bridge components or bridge system by using both system reliability and lifetime function. Monte Carlo method is used to generate the parameters of the lifetime function. As a case study, the program is applied to the concrete-steel bridge to predict the failure probability.