• Journal of Information Processing Systems
• /
• v.20 no.2
• /
• pp.263-272
• /
• 2024

A model based on genetic algorithm optimization, GA-SVM, is proposed to warn university students of their status. This model improves the predictive effect of support vector machines. The genetic optimization algorithm is used to train the hyperparameters and adjust the kernel parameters, kernel penalty factor C, and gamma to optimize the support vector machine model, which can rapidly achieve convergence to obtain the optimal solution. The experimental model was trained on open-source datasets and validated through comparisons with random forest, backpropagation neural network, and GA-SVM models. The test results show that the genetic algorithm-optimized radial basis kernel support vector machine model GA-SVM can obtain higher accuracy rates when used for early warning in university learning.

• Journal of Welding and Joining
• /
• v.25 no.5
• /
• pp.64-71
• /
• 2007

The aim of this paper is to investigate the effects of welding parameters on the weld shape in seal-welding of STS304L capsule for manufacturing a radioisotope source which is widely used in nondestructive testing of metal structures using gamma ray. Pulsed gas tungsten arc (Pulsed GTA) welding is performed for thin cross sectional area of the capsule. Seven welding parameters including current waveform parameters and arc length etc. are selected as main process parameters using design of experiment. The weld shape such as bead width, penetration depth, weld area, aspect ratio and area rate is investigated to assess the effects of welding parameters. As results, the combination of pulse duty/welding speed largely affects on bead width, penetration depth, area and aspect ratio. Finally, it is concluded that the key parameters are the combination of pulse duty/welding speed, base current and arc length, and their optimal conditions are 50%/1.77mm/s, 6.4A and 1 mm.

• International Journal of Korean Welding Society
• /
• v.1 no.1
• /
• pp.17-22
• /
• 2001

Because the quantitative relationships between welding parameters and welding result are not yet blown, optimal values of welding parameters for $CO_2$ robotic arc welding is a difficult task. Using the various artificial data processing methods may solve this difficulty. This research aims to develop an expert system for $CO_2$ robotic arc welding to recommend the optimal values of welding parameters. This system has three main functions. First is the recommendation of reasonable values of welding parameters. For such work, the relationships in between the welding parameters are investigated by the use of regression analysis and fuzzy system. The second is the estimation of bead shape by a neural network system. In this study the welding current voltage, speed, weaving width, and root gap are considered as the main parameters influencing a bead shape. The neural network system uses the 3-layer back-propagation model and a generalized delta rule as teaming algorithm. The last is the optimization of the parameters for the correction of undesirable weld bead. The causalities of undesirable weld bead are represented in the form of rules. The inference engine derives conclusions from these rules. The conclusions give the corrected values of the welding parameters. This expert system was developed as a PC-based system of which can be used for the automatic or semi-automatic $CO_2$ fillet welding with 1.2, 1.4, and 1.6mm diameter the solid wires or flux-cored wires.

• Geophysics and Geophysical Exploration
• /
• v.25 no.3
• /
• pp.99-108
• /
• 2022

Among induced polarization (IP) methods, spectral IP (SIP) uses alternating current as a transmission source to measure amplitudes and phase of complex electrical resistivity at each source frequency, which disperse with respect to source frequencies. The frequency dependence, which can be explained by a relaxation model such as Cole-Cole model or equivalent models, is analyzed to estimate SIP parameters from dispersion curves of complex resistivity employing multi-objective optimization (MOO). The estimation uses a generic algorithm to optimize two objective functions minimizing data misfits of amplitude and phase based on Cole-Cole model, which is most widely used to explain IP relaxation effects. The MOO-based estimation properly recovered Cole-Cole model parameters for synthetic examples but hardly fitted for the real laboratory measures ones, which have relatively smaller values of phases (less than about 10 mrad). Discrepancies between scales for data misfits of amplitude and phase, used as parameters of MOO method, and it is in necessity to employ other methods such as machine learning, which can deal with the discrepancies, to estimate SIP parameters from dispersion curves of complex resistivity.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2007.04a
• /
• pp.233-236
• /
• 2007

A new evolution method for strong exploration and strong exploitation termed queen-bee and mutant-bee evolution is proposed based on the previous queen-bee evolution [1]. Even though the queen-bee evolution has shown very good performances, two parameters for strong mutation are added to the genetic algorithms. This makes the application of genetic algorithms with queen-bee evolution difficult because the values of the two parameters are empirically decided by a trial-and-error method without a systematic method. The queen-bee and mutant-bee evolution has no this problem because it does not need additional parameters for strong mutation. Experimental results with typical problems showed that the queen-bee and mutant-bee evolution produced nearly similar results to the best ones of queen-bee evolution even though it didn't need to select proper values of additional parameters.

• Proceedings of the Korean Society of Machine Tool Engineers Conference
• /
• 2004.04a
• /
• pp.298-303
• /
• 2004

parameters in surface grinding. Taguchi method which is one of the design of experiments has been introduced in achieving the aims. The process parameters were the grain size, the wheel speed, the depth of cut and the table speed. The effect of the process parameters on the geometric error was examined and an optimal set of the parameters was selected to minimize the geometric error within the controllable range of the used grinding machine. The reliability of the results was evaluated by the ANOVA.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.26 no.11
• /
• pp.2262-2269
• /
• 2002

Process optimization is carried out to determine process parameters which satisfy the given design requirement and constraint conditions in sheet metal forming processes. Sensitivity -based-approach is utilized for the optimum searching of process parameters in sheet metal forming precesses. The scheme incorporates an elasto-plastic finite element method with shell elements . Sensitivities of state variables are calculated from the direct differentiation of the governing equation for the finite element analysis. The algorithm developed is applied to design of the variablc blank holding force in deep drawing processes. Results show that determination of process parameters is well performed to control the major strain for preventing fracture by tearing or to decrease the amount of springback for improving the shape accuracy. Results demonstrate that design of process parameters with the present approach is applicable to real sheet metal forming processes.

• Proceedings of the KWS Conference
• /
• v.43
• /
• pp.73-75
• /
• 2004

The purpose of this study is to optimize Nd:YAG laser welding parameters to seaz a Rf source into a Ti micro capsule. Ti tube ends can be sealed as some length of ぉbe end is melted and coalesced. The exact control of the melted length is the most important to get sound sealing. The Nozzle type, tube rotating speed, tilt angle, focal position, pumping voltage, pulse frequency and pulse width were selected as the Nd:YAG laser welding Parameters. These Parameters were optimized by the Taguchi experimental method using 115 orthogonal array. Appearance and cross section of the seated tube ends were examined by SEM.

• Magazine of the Korean Society of Agricultural Engineers
• /
• v.33 no.4
• /
• pp.61-72
• /
• 1991

A lumped deterministic model(DAWAST model) was developed to predict the daily streamflow. Since the streamflow is dominantly determined by the soil water storage in the watershed, the model takes the soil water accounting procedures which are based on three linear reservoirs representing the surface, unsaturated, and saturated soil layers. The variation of soil water storage in the unsaturated zone is traced from the soil water balance on a daily basis. DAWAST model consists of 5 parameters for water balance and 3 parameters for routing. A optimization technique of unconstrained nonlinear Simplex method was applied for the determination of the optimal parameters for water balance. Model verification was carried out to the 7 hydrologic watersheds with areas of 5.89-7,126km$^2$ and the results were generally satisfactory. The daily streamflow can be arbitrarily simulated with the input data of daily rainfall and pan evaporation by the DAWAST model at the station where the observed streamflow data of short periods are available to calibrate the model parameters.

• Communications for Statistical Applications and Methods
• /
• v.11 no.2
• /
• pp.355-368
• /
• 2004

In order to use Kriging, one has to estimate three parameters(nugget, sill and range) of semivariogram, which shows the relationship in the given two sites. A visual fit of the semivariogram parameters to a few standard models is widely used. But, it does not give the suitable results and not provide the automated process of Kriging. The gradient based nonlinear least squares is another choices to estimate three parameters, but it has some problems such as initial value problem. In this paper, we suggest the genetic algorithm as a compatible alternative method to solve the above mentioned problem. Finally, we estimate three parameters of semivariogram of rain-fall by adapting the genetic algorithm, compute Kriging estimate and conclude its effectiveness and compatibility.