• Journal of the Semiconductor & Display Technology
• /
• v.19 no.4
• /
• pp.126-129
• /
• 2020

We introduce a lifetime assessment technique using deep learning algorithm with complex electrical parameters such as resistivity, permittivity, impedance parameters as integrated indicators for predicting the degradation of the organic molecules. The evaluation system consists of fully automated in-situ measurement system and multiple layer perceptron learning system with five hidden layers and 1011 perceptra in each layer. Prediction accuracies are calculated and compared depending on the physical feature, learning hyperparameters. 62.5% of full time-series data are used for training and its prediction accuracy is estimated as r-square value of 0.99. Remaining 37.5% of the data are used for testing with prediction accuracy of 0.95. With k-fold cross-validation, the stability to the instantaneous changes in the measured data is also improved.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.46 no.3
• /
• pp.123-138
• /
• 2023

The Fourth Industrial Revolution and sensor technology have led to increased utilization of sensor data. In our modern society, data complexity is rising, and the extraction of valuable information has become crucial with the rapid changes in information technology (IT). Recurrent neural networks (RNN) and long short-term memory (LSTM) models have shown remarkable performance in natural language processing (NLP) and time series prediction. Consequently, there is a strong expectation that models excelling in NLP will also excel in time series prediction. However, current research on Transformer models for time series prediction remains limited. Traditional RNN and LSTM models have demonstrated superior performance compared to Transformers in big data analysis. Nevertheless, with continuous advancements in Transformer models, such as GPT-2 (Generative Pre-trained Transformer 2) and ProphetNet, they have gained attention in the field of time series prediction. This study aims to evaluate the classification performance and interval prediction of remaining useful life (RUL) using an advanced Transformer model. The performance of each model will be utilized to establish a health index (HI) for cutting blades, enabling real-time monitoring of machine health. The results are expected to provide valuable insights for machine monitoring, evaluation, and management, confirming the effectiveness of advanced Transformer models in time series analysis when applied in industrial settings.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.43 no.1
• /
• pp.123-131
• /
• 2020

There have been a lot of studies in the past for the method of predicting the failure of a machine, and recently, a lot of researches and applications have been generated to diagnose the physical condition of the machine and the parts and to calculate the remaining life through various methods. Survival models are also used to predict plant failures based on past anomaly cycles. In particular, special machine that reflect the fluid flow and process characteristics of chemical plants are connected to hundreds or thousands of sensors, so there are not many factors that need to be considered, such as process and material data as well as application of derivative variables. In this paper, the data were preprocessed through time series anomaly detection based on unsupervised learning to predict the abnormalities of these special machine. Next, clustering results reflecting clustering-based data characteristics were applied to produce additional variables, and a learning data set was created based on the history of past facility abnormalities. Finally, the prediction methodology based on the supervised learning algorithm was applied, and the model update was confirmed to improve the accuracy of the prediction of facility failure. Through this, it is expected to improve the efficiency of facility operation by flexibly replacing the maintenance time and parts supply and demand by predicting abnormalities of machine and extracting key factors.

• Journal of Institute of Control, Robotics and Systems
• /
• v.20 no.7
• /
• pp.730-735
• /
• 2014

This paper introduces a GP (Genetic Programming) based robust technique for the prediction of a heavy rain newsflash. The nature of prediction for precipitation is very complex, irregular and highly fluctuating. Especially, the prediction of heavy precipitation is very difficult. Because not only it depends on various elements, such as location, season, time and geographical features, but also the case data is rare. In order to provide a robust model for precipitation prediction, a nonlinear and symbolic regression method using GP is suggested. The remaining part of the study is to evaluate the performance of prediction for a heavy rain newsflash using a GP based nonlinear regression technique in Korean regions. Analysis of the feature selection is executed and various fitness functions are proposed to improve performances. The KLAPS data of 2006-2010 is used for training and the data of 2011 is adopted for verification.

• Proceedings of the Korea Society for Energy Engineering kosee Conference
• /
• 1999.05a
• /
• pp.171-176
• /
• 1999

Nondestructive inspection and accompanying life analysis based on fracture mechanics were the major conventional methods for evaluating remaining life of critical high temperature components in power plants. By using these conventional methods, it has been difficult to perform in-service inspection for life prediction. Also, quantitative damage evaluation due to unexpected abrupt changes in operating temperature was almost impossible. Thus, many efforts have been made for evaluating remaining life during operation of the plants and predicting real-time life usage values based on the shape of structures, operating history, and material properties. In this study, a core software for on-line life monitoring system which carries out real-time life evaluation of a critical component in power boiler(high temperature steam headers) is developed. The software is capable of evaluating creep and fatigue life usage from the real-time stress data calculated by using temperature/stress transfer Green functions derived for the specific headers and by counting transient cycles. The major benefits of the developed software lie in determining future operating schedule, inspection interval, and replacement plan by monitoring real-time life usage based on prior operating history.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.17 no.4
• /
• pp.183-189
• /
• 2024

In order to effectively predict time series data, this study proposed a hybrid prediction model that decomposes the data into trend, seasonality, and residual components using Seasonal-Trend Decomposition on Loess, and then applies ARIMA to the trend component, Fourier Series Regression to the seasonality component, and XGBoost to the remaining components. In addition, performance comparison experiments including ARIMA, XGBoost, LSTM, EMD-ARIMA, and CEEMDAN-LSTM models were conducted to evaluate the prediction performance of each model. The experimental results show that the proposed hybrid model outperforms the existing single models with the best performance indicator values in MAPE(3.8%), MAAPE(3.5%), and RMSE(0.35) metrics.

• Smart Structures and Systems
• /
• v.21 no.3
• /
• pp.305-320
• /
• 2018

The second half of the 20th century was marked with a significant raise in amount of railway bridges in Austria made of reinforced concrete. Today, many of these bridges are slowly approaching the end of their envisaged service life. Current methodology of assessment and evaluation of structural condition is based on visual inspections, which, due to its subjectivity, can lead to delayed interventions, irreparable damages and additional costs. Thus, to support engineers in the process of structural evaluation and prediction of the remaining service life, the Austrian Federal Railways (${\ddot{O}}$ BB) commissioned the formation of a concept for an anticipatory life cycle management of engineering structures. The part concerning concrete bridges consisted of forming a bridge management system (BMS) in a form of a web-based analysis tool, known as the LeCIE_tool. Contrary to most BMSs, where prediction of a condition is based on Markovian models, in the LeCIE_tool, the time-dependent deterioration mechanisms of chloride- and carbonation-induced corrosion are used as the most common deterioration processes in transportation infrastructure. Hence, the main aim of this article is to describe the background of the introduced tool, with a discussion on exposure classes and crucial parameters of chloride ingress and carbonation models. Moreover, the article presents a verification of the generated analysis tool through service life prediction on a dozen of bridges of the Austrian railway network, as well as a case study with a more detailed description and implementation of the concept applied.

• Structural Engineering and Mechanics
• /
• v.49 no.2
• /
• pp.203-223
• /
• 2014

The extensive use of prestressed reinforced concrete (PSC) highway bridges in marine environment drastically increases the sensitivity to both fatigue-and corrosion-induced damage of their critical structural components during their service lives. Within this scenario, an integrated method that is capable of evaluating the fatigue reliability, identifying a condition-based maintenance, and predicting the remaining service life of its critical components is therefore needed. To accomplish this goal, a procedure for fatigue reliability prediction of PSC highway bridges is proposed in the present study. Vehicle-bridge coupling vibration analysis is performed for obtaining the equivalent moment ranges of critical section of bridges under typical fatigue truck models. Three-dimensional nonlinear mathematical models of fatigue trucks are simplified as an eleven-degree-of-freedom system. Road surface roughness is simulated as zero-mean stationary Gaussian random processes using the trigonometric series method. The time-dependent stress-concentration factors of reinforcing bars and prestressing tendons are accounted for more accurate stress ranges determination. The limit state functions are constructed according to the Miner's linear damage rule, the time-dependent S-N curves of prestressing tendons and the site-specific stress cycle prediction. The effectiveness of the methodology framework is demonstrated to a T-type simple supported multi-girder bridge for fatigue reliability evaluation.

• Journal of Korean Society for Quality Management
• /
• v.35 no.2
• /
• pp.45-52
• /
• 2007

The danger of self-ignition of single base propellants will increase with time. Therefore, a good prediction of the safe storage time is very important. In order to determine the remaining shelf-life of the propellants, the content of stabilizer is determined. The propellants stored under normal storage conditions about 10 to 18 years were investigated and accelerated aging test was carried out by storing propellant sample at higher temperature. Finally, we analyzed the results by various methods in order to show the best way to predict the realistic shelf-life. The safe storage life of the propellants will be 24 years, at least 15 years. In case of applying Arrhenius's law, using the reaction rate constant at 28$^{\circ}C$ to 30$^{\circ}C$ to predict the shelf-life by accelerated aging test is reasonable for a good prediction.

• Proceedings of the Korean Society for Quality Management Conference
• /
• 2006.11a
• /
• pp.321-326
• /
• 2006

The danger of self-ignition of single base propellants will increase with time. Therefore, a good prediction of the safe storage time is very important In order to determine the remaining shelf1ife of the propellants, the content of stabilizer is determined. The propellants stored under normal storage conditions about 10 to 18 years were investigated and accelerated aging test was carried out by storing propellant sample at higher temperature. Finally, we analyzed the results by various methods in order to show the best way to predict the realistic shelflife. The safe storage life of the propellants will be 24 years, at least 15 years. In case of applying Arrhenius's law, using the reaction rate constant at $28^{\circ}C$ to $30^{\circ}C$ to predict the shelflife by accelerated aging test is reasonable for a good prediction.