• Journal of Information Processing Systems
• /
• v.9 no.3
• /
• pp.365-378
• /
• 2013

In Wireless Mesh Networks (WMNs), we usually deploy multiple Internet Gateways (IGWs) to improve the capacity of WMNs. As most of the traffic is oriented towards the Internet and may not be distributed evenly among different IGWs, some IGWs may suffer from bottleneck problem. To solve the IGW bottleneck problem, we propose an efficient scheme to balance the load among different IGWs within a WMN. Our proposed load-balancing scheme consists of two parts: a traffic load calculation module and a traffic load migration algorithm. The IGW can judge whether the congestion has occurred or will occur by using a linear smoothing forecasting method. When the IGW detects that the congestion has occurred or will occur, it will firstly select another available IGW that has the lightest traffic load as the secondary IGW and then inform some mesh routers (MPs) which have been selected by using the Knapsack Algorithm to change to the secondary IGW. The MPs can return to their primary IGW by using a regression algorithm. Our Qualnet 5.0 experiment results show that our proposed scheme gives up to 18% end-to-end delay improvement compared with the existing schemes.

• Journal of Welding and Joining
• /
• v.17 no.5
• /
• pp.98-106
• /
• 1999

In GMAW, the spatters are generated according to the variation of the arc. Of the arc is stable, Few spatters are generated. But if unstable, too many spatters are generated. So, this means the spatters are dependent on the arc state. The aim of this study is to accurately estimate the arc state. To do this, the generated spatters were captured under the some welding conditions, and the waveforms of the arc voltage and welding current were collected. From the collected signals, the waveform factors and their standard deviations were extracted. Using these factors as input parameters of multi-layer artificial neural network, the learning for the weight of the generated spatters is performed and the estimation results to the real spatter are assessed. Obtained results are as follow: the linear correlation coefficient between the estimated result and the real spatters was 0.9986. And although the average convergence error was set 0.002, the estimated error to the real spatter was within 0.1 gr/min at each welding condition. In the estimation for the weight generated spatters, the result with multi-layer neural network was far better than with multiple regression analysis. Especially, even though under the welding condition which the arc state is unstable (the spatter is generated much more), very excellent estimation performance was shown.

• Journal of the Society of Naval Architects of Korea
• /
• v.57 no.5
• /
• pp.297-304
• /
• 2020

In this study, the prediction method was reviewed to process a cylindrical plate forming using machine learning as a data-driven approach by roll bending equipment. The calculation of the forming variables was based on the analysis using the mechanical relationship between the material properties and the roll bending machine in the bending process. Then, by applying the finite element analysis method, the accuracy of the deformation prediction model was reviewed, and a large number data set was created to apply to machine learning using the finite element analysis model for deformation prediction. As a result of the application of the machine learning model, it was confirmed that the calculation is slightly higher than the linear regression method. Applicable results were confirmed through the machine learning method.

• Journal of Ecology and Environment
• /
• v.40 no.1
• /
• pp.34-44
• /
• 2016

Background: Little is known about how chemical water quality is associated with ecological stream health in relation to landuse patterns in a watershed. We evaluated spatial characteristics of water quality characteristics and the ecological health of Dongjin-River basin, Korea in relation to regional landuse pattern. The ecological health was assessed by the multi-metric model of Index of Biological Integrity (IBI), and the water chemistry data were compared with values obtained from the health model. Results: Nutrient and organic matter pollution in Dongjin-River basin, Korea was influenced by land use pattern and the major point sources, so nutrients of TN and TP increased abruptly in Site 4 (Jeongeup Stream), which is directly influenced by wastewater treatment plants along with values of electric conductivity (EC), bacterial number, and sestonic chlorophyll-a. Similar results are shown in the downstream (S7) of Dongjin River. The degradation of chemical water quality in the downstream resulted in greater impairment of the ecological health, and these were also closely associated with the landuse pattern. Forest region had low nutrients (N, P), organic matter, and ionic content (as the EC), whereas urban and agricultural regions had opposite in the parameters. Linear regression analysis of the landuse (arable land; $A_L$) on chemicals indicated that values of $A_L$ had positive linear relations with TP ($R^2=0.643$, p < 0.01), TN ($R^2=0.502$, p < 0.05), BOD ($R^2=0.739$, p < 0.01), and suspended solids (SS; ($R^2=0.866$, p < 0.01), and a negative relation with TDN:TDP ratios ($R^2=0.719$, p < 0.01). Conclusions: Chemical factors were closely associated with land use pattern in the watershed, and these factors influenced the ecological health, based on the multimetric fish IBI model. Overall, the impairments of water chemistry and the ecological health in Dongjin-River basin were mainly attributes to point-sources and land-use patterns.

• Journal of the Korean Data and Information Science Society
• /
• v.26 no.1
• /
• pp.65-75
• /
• 2015

This paper analyzes the determinants of wage for the college and university graduates utilizing both individual-level and industry-level variables. We note that wage determination has multi-level structure in the sense that individual wage is influenced by individual-level variables (level-1) and industry-level (level-2) variables. Then, the assumption that individual wage is independent in the classical regression is violated. Therefore, this paper utilizes the hierarchical linear model (HLM). The major results are the followings. First, the multiple correspondence analysis including level-1 and 2 variables reveals that both level 1 and level 2 variables affects individual wages judging from the fact that the values of level 1 and level 2 variables differ across the different level of individual wage groups. Second, the decision tree analysis including level-1 and 2 variables shows that the most influential variable in wage determination is industry-level wage and the next is industry-level working hour, ages and sex in the decling order in. This suggests that the utilization of the HLM is appropriate since the characteristics of industry is important in determining the individual wage. Third, it is shown that the HLM model is the best compared to the other models which do not take level-1 and level-2 variables simultaneously into account.

• Journal of Institute of Control, Robotics and Systems
• /
• v.11 no.6
• /
• pp.550-557
• /
• 2005

The multivariate statistical analysis methods, using both multiple linear regression(MLR) and partial least square(PLS), have been applied to predict the reactor properties and the product quality of a direct esterification reactor for polyethylene terephthalate(PET) synthesis. On the basis of the set of data including the flow rate of water vapor, the flow rate of EG vapor, the concentration of acid end groups of a product and other operating conditions such as temperature, pressure, reaction times and feed monomer mole ratio, two multi-variable analysis methods have been applied. Their regression and prediction abilities also have been compared. The prediction results are critically compared with the actual plant data and the other mathematical model based results in reliability. This paper shows that PLS method approach can be used for the reasonably accurate prediction of a product quality of a direct esterification reactor in PET synthesis process.

• International Journal of Control, Automation, and Systems
• /
• v.6 no.5
• /
• pp.639-650
• /
• 2008

Electricity price forecasting has become an integral part of power system operation and control. In this paper, a wavelet transform (WT) based neural network (NN) model to forecast price profile in a deregulated electricity market has been presented. The historical price data has been decomposed into wavelet domain constitutive sub series using WT and then combined with the other time domain variables to form the set of input variables for the proposed forecasting model. The behavior of the wavelet domain constitutive series has been studied based on statistical analysis. It has been observed that forecasting accuracy can be improved by the use of WT in a forecasting model. Multi-scale analysis from one to seven levels of decomposition has been performed and the empirical evidence suggests that accuracy improvement is highest at third level of decomposition. Forecasting performance of the proposed model has been compared with (i) a heuristic technique, (ii) a simulation model used by Ontario's Independent Electricity System Operator (IESO), (iii) a Multiple Linear Regression (MLR) model, (iv) NN model, (v) Auto Regressive Integrated Moving Average (ARIMA) model, (vi) Dynamic Regression (DR) model, and (vii) Transfer Function (TF) model. Forecasting results show that the performance of the proposed WT based NN model is satisfactory and it can be used by the participants to respond properly as it predicts price before closing of window for submission of initial bids.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.12
• /
• pp.5765-5781
• /
• 2018

Extreme learning machine (ELM) is emerging as a powerful machine learning method in a variety of application scenarios due to its promising advantages of high accuracy, fast learning speed and easy of implementation. However, how to select the optimal hidden layer of ELM is still an open question in the ELM community. Basically, the number of hidden layer nodes is a sensitive hyperparameter that significantly affects the performance of ELM. To address this challenging problem, we propose to adopt multiple kernel learning (MKL) to design a multi-hidden-layer-kernel ELM (MHLK-ELM). Specifically, we first integrate kernel functions with random feature mapping of ELM to design a hidden-layer-kernel ELM (HLK-ELM), which serves as the base of MHLK-ELM. Then, we utilize the MKL method to propose two versions of MHLK-ELMs, called sparse and non-sparse MHLK-ELMs. Both two types of MHLK-ELMs can effectively find out the optimal linear combination of multiple HLK-ELMs for different classification and regression problems. Experimental results on seven data sets, among which three data sets are relevant to classification and four ones are relevant to regression, demonstrate that the proposed MHLK-ELM achieves superior performance compared with conventional ELM and basic HLK-ELM.

• 제어로봇시스템학회:학술대회논문집
• /
• 2003.10a
• /
• pp.250-255
• /
• 2003

The position of a 3-dimensional(3D) point can be measured by using calibrated stereo camera. To obtain more accurate measurement ,more accurate camera calibration is required. There are many existing methods to calibrate camera. The simple linear methods are usually not accurate due to nonlinear lens distortion. The nonlinear methods are accurate more than linear method, but it increase computational cost and good initial guess is needed. The multi step methods need to know some camera parameters of used camera. Recent years, these explicit model based camera calibration work with the development of more precise camera models involving correction of lens distortion. But these explicit model based camera calibration have disadvantages. So implicit camera calibration methods have been derived. One of the popular implicit camera calibration method is to use neural network. In this paper, we propose implicit stereo camera calibration method for 3D reconstruction using support vector machine. SVM can learn the relationship between 3D coordinate and image coordinate, and it shows the robust property with the presence of noise and lens distortion, results of simulation are shown in section 4.

• Geomechanics and Engineering
• /
• v.22 no.1
• /
• pp.97-108
• /
• 2020

The time-consuming and less objectivity are the main problems of conventional micromechanical parameters calibration method of Particle Flow Code simulations. Thus this study aims to address these two limitation of the conventional "trial-and-error" method. A new calibration method for the linear parallel bond model (CM-LPBM) is proposed. First, numerical simulations are conducted based on the results of the uniaxial compression tests on limestone. The macroscopic response of the numerical model agrees well with the results of the uniaxial compression tests. To reduce the number of the independent micromechanical parameters, numerical simulations are then carried out. Based on the results of the orthogonal experiments and the multi-factor variance analysis, main micromechanical parameters affecting the macro parameters of rocks are proposed. The macro-micro parameter functions are ultimately established using multiple linear regression, and the iteration correction formulas of the micromechanical parameters are obtained. To further verify the validity of the proposed method, a case study is carried out. The error between the macro mechanical response and the numerical results is less than 5%. Hence the calibration method, i.e., the CM-LPBM, is reliable for obtaining the micromechanical parameters quickly and accurately, providing reference for the calibration of micromechanical parameters.