• Journal of Advanced Marine Engineering and Technology
• /
• v.40 no.2
• /
• pp.138-145
• /
• 2016

This paper proposes an improved tabu search method for subset selection in multiple linear regression models. Variable selection is a vital combinatorial optimization problem in multivariate statistics. The selection of the optimal subset of variables is necessary in order to reliably construct a multiple linear regression model. Its applications widely range from machine learning, timeseries prediction, and multi-class classification to noise detection. Since this problem has NP-complete nature, it becomes more difficult to find the optimal solution as the number of variables increases. Two typical metaheuristic methods have been developed to tackle the problem: the tabu search algorithm and hybrid genetic and simulated annealing algorithm. However, these two methods have shortcomings. The tabu search method requires a large amount of computing time, and the hybrid algorithm produces a less accurate solution. To overcome the shortcomings of these methods, we propose an improved tabu search algorithm to reduce moves of the neighborhood and to adopt an effective move search strategy. To evaluate the performance of the proposed method, comparative studies are performed on small literature data sets and on large simulation data sets. Computational results show that the proposed method outperforms two metaheuristic methods in terms of the computing time and solution quality.

• Steel and Composite Structures
• /
• v.44 no.2
• /
• pp.241-254
• /
• 2022

The cables in a cable-stayed bridge are critical load-carrying parts. The potential damage to cables should be identified early to prevent disasters. In this study, an efficient deep learning model is proposed for the damage identification of cables using both a multi-layer perceptron (MLP) and a graph neural network (GNN). Datasets are first generated using the practical advanced analysis program (PAAP), which is a robust program for modeling and analyzing bridge structures with low computational costs. The model based on the MLP and GNN can capture complex nonlinear correlations between the vibration characteristics in the input data and the cable system damage in the output data. Multiple hidden layers with an activation function are used in the MLP to expand the original input vector of the limited measurement data to obtain a complete output data vector that preserves sufficient information for constructing the graph in the GNN. Using the gated recurrent unit and set2set model, the GNN maps the formed graph feature to the output cable damage through several updating times and provides the damage results to both the classification and regression outputs. The model is fine-tuned with the original input data using Adam optimization for the final objective function. A case study of an actual cable-stayed bridge was considered to evaluate the model performance. The results demonstrate that the proposed model provides high accuracy (over 90%) in classification and satisfactory correlation coefficients (over 0.98) in regression and is a robust approach to obtain effective identification results with a limited quantity of input data.

• Journal of the Korean GEO-environmental Society
• /
• v.13 no.6
• /
• pp.51-57
• /
• 2012

This study investigated on the photolytic degradation of Triclosan by E-beam process. The optimization of process was investigated during a series of batch experiments by design of experiments(DOEs). The DOE was one of the statistical application that was used for designed the response surface to determine the effects of each parameters. The responses were applied as removal rate of Triclosan(%, $Y_1$) and TOC removal rate(%, $Y_2$). Two independent variables were concentration of Triclosan and irradiation intensity that were designed as "$x_1$" and irradiation intensity was designed as "$x_2$". The regression equation in coded parameter between the Triclosan removal efficiencies(%) and TOC removal efficiencies(%) was $Y_1=63-12.4335x_1+15.1835x_2+5.8125x{_1}^2-5.6875x{_2}^2-0.75x_1x_2(R^2=95.1%,\;R^2(Adj)=91.7%)$ and $Y_2=46-8.8462x_1+11.7175x_2-0.75x{_1}^2-6.25x{_2}^2(R^2=98.7%,\;R^2(Adj)=97.7%)$, respectively. The model predictions agreed well with the experimentally observed results $R^2$ and $R^2(Adj)$ over 90% within both of $Y_1$ and $Y_2$. This result shows that the regression model express well about the effects of parameters on E-beam process and the statistical method was successfully applied.

• Journal of Surface Science and Engineering
• /
• v.43 no.3
• /
• pp.159-164
• /
• 2010

Surface roughness of deposited or etched film strongly depends on ion bombardment. Relationships between ion bombardment variables and surface roughness are too complicated to model analytically. To overcome this, an empirical neural network model was constructed and applied to a deposition process of silicon nitride (SiN) films. The films were deposited by using a pulsed plasma enhanced chemical vapor deposition system in $SiH_4$-$NH_4$ plasma. Radio frequency source power and duty ratio were varied in the range of 200-800 W and 40-100%. A total of 20 experiments were conducted. A non-invasive ion energy analyzer was used to collect ion energy distribution. The diagnostic variables examined include high (or) low ion energy and high (or low) ion energy flux. Mean surface roughness was measured by using atomic force microscopy. A neural network model relating the diagnostic variables to the surface roughness was constructed and its prediction performance was optimized by using a genetic algorithm. The optimized model yielded an improved performance of about 58% over statistical regression model. The model revealed very interesting features useful for optimization of surface roughness. This includes a reduction in surface roughness either by an increase in ion energy flux at lower ion energy or by an increase in higher ion energy at lower ion energy flux.

• Food Science and Preservation
• /
• v.10 no.3
• /
• pp.320-325
• /
• 2003

This study was conducted to optimize sugaring process of yam for development of new snack product and enhancement acceptability. Three variables by five level central composite design and response surface methodology were used to determine optimum conditions for sugaring time, temperature and concentration. Optimization of the process was conducted using the combination of the moisture content, solid content, color and rehydration ratio. The regression polynomial model was suitable (P>0.05) model by Lack-of-Fit analysis with highly significant. To optimize the process, based on surface response and contour plots, superimposing the individual contour plots for the response variables. The optimum conditions for this process were 5.5 hours and 58％ at 40$^{\circ}C$ under the optimum of restricted variables as moisture content was 66 to 70, solid content was 25 to 30％, L value was above 75, a value was -2.1 to -2.4, b value was above 5 and rehydration ratio was 200 to 250.

• The Transactions of the Korean Institute of Electrical Engineers D
• /
• v.53 no.8
• /
• pp.561-570
• /
• 2004

In this paper, new architectures and comprehensive design methodologies of Genetic Algorithms(GAs) based Genetically optimized Neurofuzzy Networks(GoNFN) are introduced, and a series of numeric experiments are carried out. The proposed GoNFN is based on the rule-based Neurofuzzy Networks(NFN) with the extended structure of the premise and the consequence parts of fuzzy rules being formed within the networks. The premise part of the fuzzy rules are designed by using space partitioning in terms of fuzzy sets defined in individual variables. In the consequence part of the fuzzy rules, three different forms of the regression polynomials such as constant, linear and quadratic are taken into consideration. The structure and parameters of the proposed GoNFN are optimized by GAs. GAs being a global optimization technique determines optimal parameters in a vast search space. But it cannot effectively avoid a large amount of time-consuming iteration because GAs finds optimal parameters by using a given space. To alleviate the problems, the dynamic search-based GAs is introduced to lead to rapidly optimal convergence over a limited region or a boundary condition. In a nutshell, the objective of this study is to develop a general design methodology o GAs-based GoNFN modeling, come up a logic-based structure of such model and propose a comprehensive evolutionary development environment in which the optimization of the model can be efficiently carried out both at the structural as well as parametric level for overall optimization by utilizing the separate or consecutive tuning technology. To evaluate the performance of the proposed GoNFN, the models are experimented with the use of several representative numerical examples.

• Tunnel and Underground Space
• /
• v.28 no.6
• /
• pp.651-669
• /
• 2018

This study suggests the prediction model to estimate the specific energy of a pick cutter using a gene expression programming (GEP) and particle swarm optimization (PSO). Estimating the performance of mechanical excavators is of crucial importance in early design stage of tunnelling projects, and the specific energy (SE) based approach serves as a standard performance prediction procedure that is applicable to all excavation machines. The purpose of this research, is to investigate the relationship between UCS and BTS, penetration depth, cut spacing, and SE. A total of 46 full-scale linear cutting test results using pick cutters and different values of depth of cut and cut spacing on various rock types was collected from the previous study for the analysis. The Mean Squared Error (MSE) associated with the conventional Multiple Linear Regression (MLR) method is more than two times larger than the MSE generated by GEP-PSO algorithm. The $R^2$ value associated with the GEP-PSO algorithm, is about 0.13 higher than the $R^2$ associated with MLR.

• Journal of Biosystems Engineering
• /
• v.12 no.2
• /
• pp.38-43
• /
• 1987

An experimental study was conducted to develop a thin-layer rewetting equation of short grain rough rice of Akihikari variety. Four thin-layer rewetting equations were experimentally determined from $25^{\circ}C$ to $45^{\circ}C$ and 70%RH to 85%RH conditions. Diffusion, Henderson, Page, and Thompson equations widely used as thin-layer drying equations were selected. Experimental data were fitted to these equations using linear regression analysis except diffusion equation. The diffusivity in the diffusion equation was determined by optimization method. Four equations were highly significant. In order to compare the goodness of fit of each equation, the error mean square of each equawas calculated. The diffusion model was not a very good model because the error mean square was very large. The other three models showed the same level or error mean square and could predict satisfactorily the rewetting rate or short grain rough rice.

• Journal of Biosystems Engineering
• /
• v.34 no.6
• /
• pp.434-438
• /
• 2009

This study was conducted to design an cone penetrometer type impedance sensor that can measure soil water content in realtime. The best width between electrical probe was determined by 5 mm. For optimization about realtime application device, linear regression analysis was applied between soil water content and impedance signal. It was concluded that proper combination of excitation frequency, impedance parameter, and model would provide acceptible performance of a soil waler content sensoe. Best model was obatained at a 36.5 MHz with |Z| as a predictor variable, with a coefficient of determination of 0.96 (RMSE=1.35, RPD=4.98).

• Proceedings of the KIEE Conference
• /
• 1987.07b
• /
• pp.981-985
• /
• 1987

In this paper, we predicted the quantities of ass output messages with the generalized estimation equation based on regression model. And, to know the generalization of equation, we measured the deviation of errors between the observed and the estimated values. As a result, the proposed equation applied to sample data showed linear characteristics in some cases.