• Journal of the Society of Disaster Information
• /
• v.18 no.1
• /
• pp.146-153
• /
• 2022

Purpose: In the case of domestic port facilities, port structures that have been in use for a long time have many problems in terms of safety performance and functionality due to the enlargement of ships, increased frequency of use, and the effects of natural disasters due to climate change. A big data analysis method was studied to develop an approximate model that can predict the aging pattern of a port facility based on the maintenance history data of the port facility. Method: In this study, member-level maintenance history data for caisson-type quay walls were collected, defined as big data, and based on the data, a predictive approximation model was derived to estimate the aging pattern and deterioration of the facility at the project level. A state-based aging pattern prediction model generated through Gaussian process (GP) and linear interpolation (SLPT) techniques was proposed, and models suitable for big data utilization were compared and proposed through validation. Result: As a result of examining the suitability of the proposed method, the SLPT method has RMSE of 0.9215 and 0.0648, and the predictive model applied with the SLPT method is considered suitable. Conclusion: Through this study, it is expected that the study of predicting performance degradation of big data-based facilities will become an important system in decision-making regarding maintenance.

• International Journal of Highway Engineering
• /
• v.16 no.3
• /
• pp.43-50
• /
• 2014

PURPOSES : The dynamic modulus can be determined by applying the various theories from the Impact Resonance Testing(IRT) Method. The objective of this paper is to determine the best theory to produce the dynamic modulus that has the lowest error as the dynamic modulus data obtained from these theories(Complex Wave equation Resonance Method related to either the transmissibility loss or not, Dynamic Stiffness Resonance Method) compared to the results for dynamic modulus determined by using the Universal Testing Machine. The ultimate object is to develop the predictive model for the dynamic modulus of a Linear Visco-Elastic specimen by using the Complex Wave equation Resonance Method(CWRM) came up for an existing study(S. O. Oyadiji; 1985) and the Optimization. METHODS : At the destructive test which uses the Universal Testing Machine, the dynamic modulus results along with the frequency can be used for determining the sigmoidal master curve function related to the reduced frequency by applying Time-Temperature Superposition Principle. RESULTS : The constant to be solved from Eq. (11) is a value of 14.13. The reduced dynamic modulus obtained from the IRT considering the loss factor related to the impact transmissibility has RMSE of 367.7MPa, MPE of 3.7%. When the predictive dynamic modulus model was applied to determine the master curve, the predictive model has RMSE of 583.5MPa, MPE of 3.5% compared to the destructive test results for the dynamic modulus. CONCLUSIONS : Because we considered that the results obtained from the destructive test had the most highest source credibility in this study, the dynamic modulus data obtained respectively from DSRM, CWRM were compared to the results obtained from the destructive test by using th IRT. At the result, the reduced dynamic modulus derived from DSRM has the most lowest error.

• Food Science and Preservation
• /
• v.6 no.1
• /
• pp.29-36
• /
• 1999

Using Fuji apple fruits cultivated in Kyungpook prefecture, the calibration model for firmness evaluation of fruits by near infrared(NIR) reflectance spectroscopy was developed, and the various influence factors such as instrument variety, measuring method, sample group, apple peel and selection of firmness point were investigated. Spectra of sample were recorded in wavelength range of 1100∼2500nm using NIR spectrometer (InfraAlyzer 500), and data were analyzed by stepwise multiple linear regression of IDAS program. The accuracy of calibration model was the highest when using sample group with wide range, and the firmness mean values obtained in graph by texture analyser(TA) were used as standard data. Chemometrics models were developed using a calibration set of 324 samples and an independent validation set of 216 samples to evaluate the predictive ability of the models. The correlation coefficients and standard error of prediction were 0.84 and 0.094kg, respectively. Using developed calibration model, it was possible to monitor the firmness change of fruits during storage frequently. Time, which was reached to firmness high value in graph by TA, is possible to use as new parameter for freshness of fruit surface during storage.

• Water for future
• /
• v.14 no.4
• /
• pp.19-26
• /
• 1981

The auto-and cross-correlation function, power spectrum, coherence function and Markov model are applied to investigate the statistical characteristics of discharge and each factor of water quality and the interrelation-ship between the variation of discharge and water quality factors. The analysis of discharge, dissolved oxygen and electric conductivity, which were only obtainable data at the Indogyo gagining station in the downstream of the Han River, clearly showed that they hace distinct period of 12 months and three different periods of 6, 4 and 3 months weaker than the former. The cross-correlation between the discharge and water quality(DO, COND) is rather weak and the crosscorrelation function has its peak at lag one. It is considered therefrom that the variation of discharge behaves on water quality facotrs with one day's difference. In the examination of linear regression model for the serial generation and predictive measures, discharge series is fit to first and second order Markov model and DO, COND to first order Markov model.

• International Journal of Control, Automation, and Systems
• /
• v.1 no.2
• /
• pp.194-202
• /
• 2003

In this paper, we introduce a category of Multi-FNN (Fuzzy-Neural Networks) models, analyze the underlying architectures and propose a comprehensive identification framework. The proposed Multi-FNNs dwell on a concept of fuzzy rule-based FNNs based on HCM clustering and evolutionary fuzzy granulation, and exploit linear inference being treated as a generic inference mechanism. By this nature, this FNN model is geared toward capturing relationships between information granules known as fuzzy sets. The form of the information granules themselves (in particular their distribution and a type of membership function) becomes an important design feature of the FNN model contributing to its structural as well as parametric optimization. The identification environment uses clustering techniques (Hard C - Means, HCM) and exploits genetic optimization as a vehicle of global optimization. The global optimization is augmented by more refined gradient-based learning mechanisms such as standard back-propagation. The HCM algorithm, whose role is to carry out preprocessing of the process data for system modeling, is utilized to determine the structure of Multi-FNNs. The detailed parameters of the Multi-FNN (such as apexes of membership functions, learning rates and momentum coefficients) are adjusted using genetic algorithms. An aggregate performance index with a weighting factor is proposed in order to achieve a sound balance between approximation and generalization (predictive) abilities of the model. To evaluate the performance of the proposed model, two numeric data sets are experimented with. One is the numerical data coming from a description of a certain nonlinear function and the other is NOx emission process data from a gas turbine power plant.

• Journal of the Korean GEO-environmental Society
• /
• v.25 no.7
• /
• pp.5-19
• /
• 2024

During the blasting process, a fracture zone is formed in the vicinity of the blast hole. Any damage that extends beyond the excavation boundary line necessitates the implementation of an additional support system to assure safety. Typically, fracture zone radius is estimated from blast hole pressure using theoretical methods due to its simplicity. However, linear charge concentration (kg/m) is used for tunnel blasting. This paper compiles Swedish experimental datasets to estimate the radius of fracture zones based on linear charge concentration. Further numerical analyses are performed in LS-DYNA for coupled single-hole blasting. The Riedel-Hiermaier-Thoma (RHT) model has been selected as the constitutive model for this investigation. The numerical model is validated against small-scale laboratory tests. Parametric studies are conducted to predict fracture zones in granite and sandstone rocks using two kinds of explosives, PETN and AFNO. The analyses evaluate ten types of blast hole sizes, ranging from 17 to 100 mm. The results indicate that granite has a larger fracture zone than sandstone, and the PETN explosive predicts more damage than ANFO. Smaller blast holes exhibit smaller fracture zones in comparison to larger blast holes. Wave propagation is more rapidly attenuated in granite than in sandstone. Subsequently, the predicted fracture zone outcomes are compared with the empirical dataset. Fracture zones of medium blast hole diameter align well with the experimental data set. A predictive equation is derived from the data set, which may be used to evaluate blast design to manage fracture zones beyond the excavation line.

• Journal of Korean Neurosurgical Society
• /
• v.53 no.3
• /
• pp.167-173
• /
• 2013

Objective : There are many cases in which degenerative changes are prevalent in both the cervical and lumbar spine, and the relation between both spinal degenerative findings of MRI is controversial. The authors analyzed the prevalence of abnormal findings on MRI, and suggested a model to explain the relationship between cervical and lumbar disc in asymptomatic Korean subjects. Methods : We performed 3 T MRI sagittal scans on 102 asymptomatic subjects (50 men and 52 women) who visited our hospital between the ages of 14 and 82 years (mean age 46.3 years). Scores pertaining to herniation (HN), annular fissure (AF), and nucleus degeneration (ND) were analyzed. The total scores for the cervical and lumbar spine were analyzed using correlation coefficients and multiple linear regression with various predictive parameters, including weight, height, sex, age, smoking, occupation, and sedentary fashion. Results : The correlation coefficients of HN, AF, and ND were 0.44, 0.50, and 0.59, respectively. We made the best model for relationship by using multiple linear regression. Conclusion : The results of the current study showed that there was a close relationship between the cervical score (CS) and lumbar score (LS). In addition, the correlation between CS and LS, as well as the LS value itself, can be altered by other explanatory variables. Although not absolute, there was also a linear relationship between degenerative changes of the cervical and lumbar spine. Based on these results, it can be inferred that degenerative changes of the lumbar spine will be useful in predicting the degree of cervical spine degeneration in an actual clinical setting.

• Journal of Institute of Control, Robotics and Systems
• /
• v.8 no.2
• /
• pp.126-135
• /
• 2002

The study is concerned with an approach to the design of new architectures of fuzzy neural networks and the discussion of comprehensive design methodology supporting their development. We propose an Adaptive Fuzzy Polynomial Neural Networks(APFNN) based on Fuzzy Neural Networks(FNN) and Self-organizing Networks(SON) for model identification of complex and nonlinear systems. The proposed AFPNN is generated from the mutually combined structure of both FNN and SON. The one and the other are considered as the premise and the consequence part of AFPNN, respectively. As the premise structure of AFPNN, FNN uses both the simplified fuzzy inference and error back-propagation teaming rule. The parameters of FNN are refined(optimized) using genetic algorithms(GAs). As the consequence structure of AFPNN, SON is realized by a polynomial type of mapping(linear, quadratic and modified quadratic) between input and output variables. In this study, we introduce two kinds of AFPNN architectures, namely the basic and the modified one. The basic and the modified architectures depend on the number of input variables and the order of polynomial in each layer of consequence structure. Owing to the specific features of two combined architectures, it is possible to consider the nonlinear characteristics of process system and to obtain the better output performance with superb predictive ability. The availability and feasibility of the AFPNN are discussed and illustrated with the aid of two representative numerical examples. The results show that the proposed AFPNN can produce the model with higher accuracy and predictive ability than any other method presented previously.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2019.10a
• /
• pp.585-588
• /
• 2019

Predictions of Energy Consumption for Industries gain an important place in energy management and control system, as there are dynamic and seasonal changes in the demand and supply of energy. This paper presents and discusses the predictive models for energy consumption of the steel industry. Data used includes lagging and leading current reactive power, lagging and leading current power factor, carbon dioxide (tCO2) emission and load type. In the test set, four statistical models are trained and evaluated: (a) Linear regression (LR), (b) Support Vector Machine with radial kernel (SVM RBF), (c) Gradient Boosting Machine (GBM), (d) random forest (RF). Root Mean Squared Error (RMSE), Mean Absolute Error (MAE) and Mean Absolute Percentage Error (MAPE) are used to measure the prediction efficiency of regression designs. When using all the predictors, the best model RF can provide RMSE value 7.33 in the test set.

• Korean Chemical Engineering Research
• /
• v.52 no.5
• /
• pp.592-602
• /
• 2014

In the Integrated Gasification Combined Cycle (IGCC), the stability of the gasifier has strong influences on the rest of the plant as it supplies the feed to the rest of the power generation system. In order to ensure a safe and stable operation of the entrained-flow gasifier and for protection of the gasifier wall from the high internal temperature, the solid slag layer thickness should be regulated tightly but its control is hampered by the lack of on-line measurement for it. In this study, a previously published dynamic simulation model of a Shell-type gasifier is reproduced and two different linear model predictive control strategies are simulated and compared for multivariable control of the entrained-flow gasifier. The first approach is to control a measured secondary variable as a surrogate to the unmeasured slag thickness. The control results of this approach depended strongly on the unmeasured disturbance type. In other words, the slag thickness could not be controlled tightly for a certain type of unmeasured disturbance. The second approach is to estimate the unmeasured slag thickness through the Kalman filter and to use the estimate to predict and control the slag thickness directly. Using the second approach, the slag thickness could be controlled well regardless of the type of unmeasured disturbances.