• Title/Summary/Keyword: Remaining Useful Life Prediction

Search Result 41, Processing Time 0.022 seconds

Deep Learning based Machine Remaining Useful Life Prediction System (딥러닝 기반의 기계 잔존 수명 예측 시스템)

  • Lee, Se-Hoon;Kim, Han-Sol;Jung, Chan-Young;Lee, Tae-Hyeong;Kim, Ji-Tae;Song, Kyung-Hwan;Sohn, Jung-Mo
    • Proceedings of the Korean Society of Computer Information Conference
    • /
    • 2020.07a
    • /
    • pp.15-16
    • /
    • 2020
  • 본 논문에서는 산업 현장에서 사용되는 기계들의 건전성을 유지하고 예측하는 시스템을 개선할 수 있는 연구 결과를 비교하고 설명한다. 이번 연구에서는 딥러닝 기술을 이용함으로서 특정장치에 종속되지 않고 범용적으로 수집된 소음데이터를 사용하여 현장 적용의 유연성을 높이고, 딥러닝 모델 중 GRU를 이용하여 기존 연구 결과와 비교 실험을 하여 더 우수한 결과를 얻었다.

  • PDF

A Long-term Durability Prediction for RC Structures Exposed to Carbonation Using Probabilistic Approach (확률론적 기법을 이용한 탄산화 RC 구조물의 내구성 예측)

  • Jung, Hyun-Jun;Kim, Gyu-Seon
    • Journal of the Korea institute for structural maintenance and inspection
    • /
    • v.14 no.5
    • /
    • pp.119-127
    • /
    • 2010
  • This paper provides a new approach for durability prediction of reinforced concrete structures exposed to carbonation. In this method, the prediction can be updated successively by a Bayes' theorem when additional data are available. The stochastic properties of model parameters are explicitly taken into account in the model. To simplify the procedure of the model, the probability of the durability limit is determined based on the samples obtained from the Latin Hypercube Sampling(LHS) technique. The new method may be very useful in design of important concrete structures and help to predict the remaining service life of existing concrete structures which have been monitored. For using the new method, in which the prior distribution is developed to represent the uncertainties of the carbonation velocity using data of concrete structures(3700 specimens) in Korea and the likelihood function is used to monitor in-situ data. The posterior distribution is obtained by combining a prior distribution and a likelihood function. Efficiency of the LHS technique for simulation was confirmed through a comparison between the LHS and the Monte Calro Simulation(MCS) technique.

Prediction Remaining Useful Life of Aircraft Turbofans Using Transfer Learning Based CNN-LSTM (전이학습 기반 CNN-LSTM을 통한 항공기 터보팬 잔여 유효 수명 예측)

  • Kim Jeong Min;Kang Hyeon Woo;Cho Young Ki;Kwon Gi Hyuk;An Seo Yeon;Kim Hun Kee
    • The Transactions of the Korea Information Processing Society
    • /
    • v.13 no.12
    • /
    • pp.700-709
    • /
    • 2024
  • The objective of research in the field of prognostics and health management is to predict the Remaining Useful Life of aircraft engines, a critical component of analysis within this domain. Nevertheless, there are difficulties in acquiring dependable failure information, and the limited availability of defect data hinders the development of predictive models. Current data augmentation techniques are utilized to enhance the insufficient defect data; however, the heuristic approaches might oversimplify the data characteristics, ultimately decreasing predictive accuracy. This study suggests a hybrid model that combines Transfer Learning, specifically integrating Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM). The hybrid CNN-LSTM model integrates the CNN's feature extraction capabilities with the LSTM's long-term time series learning capacity, facilitating the representation of intricate dynamic characteristics and temporal fluctuations in aircraft engine sensor data. The performance of predictive techniques is enhanced by applying data learned from various source domains to target domain data through transfer learning. The results obtained by applying this model to the C-MAPSS aircraft engine simulator dataset developed by the National Aeronautics and Space Administration (NASA) corroborate the idea that employing a pre-trained model through transfer learning improves predictive accuracy in comparison to the standard mixed model. Furthermore, the proposed model demonstrates improved predictive abilities when compared to various leading predictive models in the PHM field.

Durability Prediction for Concrete Structures Exposed to Chloride Attack Using a Bayesian Approach (베이지안 기법을 이용한 염해 콘크리트구조물의 내구성 예측)

  • Jung, Hyun-Jun;Zi, Goang-Seup;Kong, Jung-Sik;Kang, Jin-Gu
    • Journal of the Korea Concrete Institute
    • /
    • v.20 no.1
    • /
    • pp.77-88
    • /
    • 2008
  • This paper provides a new approach for predicting the corrosion resistivity of reinforced concrete structures exposed to chloride attack. In this method, the prediction can be updated successively by a Bayesian theory when additional data are available. The stochastic properties of model parameters are explicitly taken into account into the model. To simplify the procedure of the model, the probability of the durability limit is determined from the samples obtained from the Latin hypercube sampling technique. The new method may be very useful in designing important concrete structures and help to predict the remaining service life of existing concrete structures which have been monitored.

Framework Development for Fault Prediction in Hot Rolling Mill System (열간 압연 설비의 고장 예지를 위한 프레임워크 구축)

  • Son, J.D.;Yang, B.S.;Park, S.H.
    • Transactions of the Korean Society for Noise and Vibration Engineering
    • /
    • v.21 no.3
    • /
    • pp.199-205
    • /
    • 2011
  • This paper proposes a framework to predict the mechanical fault of hot rolling mill system (HRMS). The optimum process of HRMS is usually identified by the rotating velocity of working roll. Therefore, observing the velocity of working roll is relevant to early know the HRMS condition. In this paper, we propose the framework which consists of two methods namely spectrum matrix which related to case-based fast Fourier transform(FFT) analysis, and three dimensional condition monitoring based on novel visualization. Validation of the proposed method has been conducted using vibration data acquired from HRMS by accelerometer sensors. The acquired data was also tested by developed software referred as hot rolling mill facility analysis module. The result is plausible and promising, and the developed software will be enhanced to be capable in prediction of remaining useful life of HRMS.

Neural Network based Aircraft Engine Health Management using C-MAPSS Data (C-MAPSS 데이터를 이용한 항공기 엔진의 신경 회로망 기반 건전성관리)

  • Yun, Yuri;Kim, Seokgoo;Cho, Seong Hee;Choi, Joo-Ho
    • Journal of Aerospace System Engineering
    • /
    • v.13 no.6
    • /
    • pp.17-25
    • /
    • 2019
  • PHM (Prognostics and Health Management) of aircraft engines is applied to predict the remaining useful life before failure or the lifetime limit. There are two methods to establish a predictive model for this: The physics-based method and the data-driven method. The physics-based method is more accurate and requires less data, but its application is limited because there are few models available. In this study, the data-driven method is applied, in which a multi-layer perceptron based neural network algorithms is applied for the life prediction. The neural network is trained using the data sets virtually made by the C-MAPSS code developed by NASA. After training the model, it is applied to the test data sets, in which the confidence interval of the remaining useful life is predicted and validated by the actual value. The performance of proposed method is compared with previous studies, and the favorable accuracy is found.

Durability Assesment for Concrete Structures Exposed to Chloride Attack Using a Bayesian Approach (베이지안 기법을 이용한 염해 콘크리트 구조물의 내구성 평가)

  • Jung, Hyun-Jun;Zi, Goang-Seup
    • Proceedings of the Computational Structural Engineering Institute Conference
    • /
    • 2007.04a
    • /
    • pp.589-594
    • /
    • 2007
  • This paper is shown new method for durability assesment and design have been noticed to be very valuable has been successfully applied to predict concrete structures. This paper provides that a new approach for predicting the corrosion durability of reinforced concrete structures exposed to chloride attack. In this method, the prediction can be updated successive1y by the Bayesian theory when additional data are available. The stochastic properties of model parameters are explicitly taken into account into the model the probability of the durability limit is determined from the samples obtained from the Latin hypercube sampling technique. The new method may be very useful in designing important concrete structures and help to predict the remaining service life of existing concrete structures under chloride attack environments.

  • PDF

A Study on Performance Reliability Analysis Device of Primary Battery (1차 전지의 성능 신뢰도 분석 장치에 관한 연구)

  • Kim, Yon Soo;Chung, Young-Bae
    • Journal of Korean Society of Industrial and Systems Engineering
    • /
    • v.37 no.2
    • /
    • pp.70-76
    • /
    • 2014
  • In industrial situation, electronic and electro-mechanical systems have been using different type of batteries in rapidly increasing numbers. These systems commonly require high reliability for long periods of time. Wider application of battery for low-power design as a prime power source requires us knowledge of failure mechanism and reliability of batteries in terms of load condition, environment condition and other explanatory variables. Battery life is an important factor that affects the reliability of such systems. There is need for us to understand the mechanism leading to the failure state of battery with performance characteristic and develop a method to predict the life of such battery. The purpose of this paper is to develope the methodology of monitoring the health of battery and determining the condition or fate of such systems through the performance reliability to predict the remaining useful life of primary battery with load condition, operating condition, environment change in light of battery life variation. In order to evaluate on-going performance of systems and subsystems adopting primary batteries as energy source, The primitive prototype for performance reliability analysis device was developed and related framework explained.

Machine Condition Prognostics Based on Grey Model and Survival Probability

  • Tangkuman, Stenly;Yang, Bo-Suk;Kim, Seon-Jin
    • International Journal of Fluid Machinery and Systems
    • /
    • v.5 no.4
    • /
    • pp.143-151
    • /
    • 2012
  • Predicting the future condition of machine and assessing the remaining useful life are the center of prognostics. This paper contributes a new prognostic method based on grey model and survival probability. The first step of the method is building a normal condition model then determining the error indicator. In the second step, the survival probability value is obtained based on the error indicator. Finally, grey model coupled with one-step-ahead forecasting technique are employed in the last step. This work has developed a modified grey model in order to improve the accuracy of prediction. For evaluating the proposed method, real trending data of low methane compressor acquired from condition monitoring routine were employed.

A study on the multiple health monitoring indicator for remaining useful life prediction of battery (리튬이온 배터리의 잔여 수명 예측을 위한 다중 건전성 모니터링 지표 연구)

  • Kwon, Sanguk;Kim, Kyutae;Yoon, Sunghyun;Lim, Cheolwoo;Kim, Jonghoon
    • Proceedings of the KIPE Conference
    • /
    • 2020.08a
    • /
    • pp.130-132
    • /
    • 2020
  • 배터리 시스템은 어플리케이션의 대영화에 따른 데이터 저장공간 문제 및 연속적인 배터리 신뢰성 문제 해결을 위한 건전성 예측 및 관리기술 접목에 관한 문제에 직면해 있으며, 이러한 문제 해결을 위해서는 배터리 시스템 신호를 통해 추출 가능한 건전성 지표 수립이 중요하다. 본 논문은 건전성 지표를 물리적, 간접적 지표로써 정의하고, 사이클 노화 데이터를 통해 건전성 지표로써의 성능을 검증하였다.

  • PDF