• 소성∙가공
• /
• 제29권4호
• /
• pp.218-228
• /
• 2020

The prediction of final mass and optimized process conditions of injection molded products using Artificial Neural Network (ANN) were demonstrated. The ANN was modeled with 10 input parameters and one output parameter (mass). The input parameters, i.e.; melt temperature, mold temperature, injection speed, packing pressure, packing time, cooling time, back pressure, plastification speed, V/P switchover, and suck back were selected. To generate training data for the ANN model, 77 experiments based on the combination of orthogonal sampling and random sampling were performed. The collected training data were normalized to eliminate scale differences between factors to improve the prediction performance of the ANN model. Grid search and random search method were used to find the optimized hyper-parameter of the ANN model. After the training of ANN model, optimized process conditions that satisfied the target mass of 41.14 g were predicted. The predicted process conditions were verified through actual injection molding experiments. Through the verification, it was found that the average deviation in the optimized conditions was 0.15±0.07 g. This value confirms that our proposed procedure can successfully predict the optimized process conditions for the target mass of injection molded products.

• 대한화장품학회지
• /
• 제24권3호
• /
• pp.105-110
• /
• 1998

In the make-up product, Eye-shadow products have several purposes of enhancing product quality such as providing the beauty (variation of shape, clean appearance), feeling, continuity and adhesion. In this paper, newly developed wet type back injection press process is introduced so as to increase higher value products which providing various the beauty. The solvent takes an essential role to provide the fluidity of the powder bulk during the pressed-process of wet type pressed product. In this study, the effect of solvent in the oil binder was investigated, And the higher quality condition of the wet type pressed product was built to apply cosmetic preparation. Firstly, the system was designed powder phase as non treated pigment. The oil binder phase is categorized as hydrocarbons(Mineral oil, Squalane), Silicones(Methicone, Dimethicone ), esters (Octyldodecanol, Octyl Dodecyl Myristate). The solvent phase used was C 7-8 isoparaffin and Isopropyl Alcohol. The interaction of oil binder and solvent is investigated by measuring mass of final oil binder and the each solubility parameter. It was found that the higher the solubility the higher the degree of change in the final composition of the oil binder. In order to maintain the quality of the final product, the solvent used in pressed-process should be hydrophobic with oil binder.

• 전기학회논문지P
• /
• 제52권2호
• /
• pp.56-60
• /
• 2003

In this work, we have fabricated thin film resistors using the DC/RF magnetron sputter of 51wt%Ni-41wt%Cr-8wt%Si alloy target and studied the effect of the process parameters on the electrical properties. In fabrication process, sputtering power, substrate temperature and annealing temperature have been varied as controllable parameters. TCR decreases with increasing the substrate temperature, but TCR increases over 300 [$^{\circ}C$]. The films are annealed to 400 [$^{\circ}C$] in air atmosphere, TCR increases with increasing the annealing temperature. The resistivity was 172 [${\mu}{\Omega}{\cdot}cm$] and 209 [${\mu}{\Omega}{\cdot}cm$] for the RF and DC as a sputtering power sources, respectively. Also, TCR was -52 [$ppm/^{\circ}C$] and -25 [$ppm/^{\circ}C$]. As a results of them, it is suggested that the sheet resistance and TCR of thin films can be controlled by variation of sputter process parameter and annealing of thin film.

• Design & Manufacturing
• /
• 제16권3호
• /
• pp.1-8
• /
• 2022

In this study, an artificial neural network(ANN) was constructed to establish the relationship between process condition prameters and the qualities of the injection-molded product in the injection molding process. Six process parmeters were set as input parameter for ANN: melt temperature, mold temperature, injection speed, packing pressure, packing time, and cooling time. As output parameters, the mass, nominal diameter, and height of the injection-molded product were set. Two learning structures were applied to the ANN. The single-task learning, in which all output parameters are learned in correlation with each other, and the multi-task learning structure in which each output parameters is individually learned according to the characteristics, were constructed. As a result of constructing an artificial neural network with two learning structures and evaluating the prediction performance, it was confirmed that the predicted value of the ANN to which the multi-task learning structure was applied had a low RMSE compared with the single-task learning structure. In addition, when comparing the quality specifications of injection molded products with the prediction values of the ANN, it was confirmed that the ANN of the multi-task learning structure satisfies the quality specifications for all of the mass, diameter, and height.

• Carbon letters
• /
• 제22권
• /
• pp.14-24
• /
• 2017

Carboxylated multi-wall carbon nanotubes (MWCNTs-COOH) have been used as efficient adsorbents for the removal of picric acid from aqueous solutions under stirring and ultrasound conditions. Batch experiments were conducted to study the influence of the different parameters such as pH, amount of adsorbents, contact time and concentration of picric acid on the adsorption process. The kinetic data were fitted with pseudo-first order, pseudo-second-order, Elovich and intra-particle diffusion models. The kinetic studies were well described by the pseudo-second-order kinetic model for both methods. In addition, the adsorption isotherms of picric acid from aqueous solutions on the MWCNTs were investigated using six two-parameter models (Langmuir, Freundlich, Tempkin, Halsey, Harkins-Jura, Fowler-Guggenheim), four three-parameter models (Redlich-Peterson, Khan, Radke-Prausnitz, and Toth), two four-parameter equations (Fritz-Schlunder and Baudu) and one five-parameter equation (Fritz-Schlunder). Three error analysis methods, correlation coefficient, chi-square test and average relative errors, were applied to determine the best fit isotherm. The error analysis showed that the models with more than two parameters better described the picric acid sorption data compared to the two-parameter models. In particular, the Baudu equation provided the best model for the picric acid sorption data for both methods.

• Journal of Mechanical Science and Technology
• /
• 제20권11호
• /
• pp.1801-1812
• /
• 2006

In this paper, a 4-wheel vehicle model including the effects of tire slip was considered, along with variable parameter sliding control, pushrod force as the end control parameter, and an antilock sliding control, in order to improve the performance of the vehicle longitudinal response. The variable sliding parameter is made to be proportional to the square root of the pressure derivative at the wheel, in order to compensate for large pressure changes in the brake cylinder. A typical tire force-relative slip curve for dry road conditions was used to generate an analytical tire force-relative slip function, and an antilock sliding control process based on the analytical tire force-relative slip function was used. A retrofitted brake system, with the pushrod force as the end control parameter, was employed, and an average decay function was used to suppress the simulation oscillations. Simulation results indicate that the velocity and spacing errors were slightly larger than the results that without considering wheel slip effect, the spacing errors of the lead and follower were insensitive to the adhesion coefficient up to the critical wheel slip value, and the limit for the antilock control on non-constant adhesion road condition was determined by the minimum of the equivalent adhesion coefficient.

• 대한전기학회논문지:전기기기및에너지변환시스템부문B
• /
• 제51권10호
• /
• pp.573-580
• /
• 2002

This paper presents neural load disturbance observer that used to deadbeat load torque observer and regulation of the compensation gain by parameter estimator As a result, the response of PMSM follows that of the nominal plant. The load torque compensation method is compose of a neural deadbeat observer. To reduce of the noise effect, the post-filter, which is implemented by MA process, is proposed. The parameter compensator with RLSM(recursive least square method) parameter estimator is suggested to increase the performance of the load torque observer and main controller. The proposed estimator is combined with a high performance neural torque observer to resolve the problems. As a result, the proposed control system becomes a robust and precise system against the load torque and the parameter variation. A stability and usefulness, through the verified computer simulation and experiment, are shown in this paper.

• Structural Engineering and Mechanics
• /
• 제71권1호
• /
• pp.45-56
• /
• 2019

A modified mechanical model of pre-pressed spring self-centering energy dissipation (PS-SCED) brace is proposed, and the hysteresis band is distinguished by the indication of relevant state variables. The MDOF frame system equipped with the braces is formulated in an incremental form of linear acceleration method. A multi-objective genetic algorithm (GA) based brace parameter optimization method is developed to obtain an optimal solution from the primary design scheme. Parameter sensitivities derived by the direct differentiation method are used to modify the change rate of parameters in the GA operator. A case study is conducted on a steel braced frame to illustrate the effect of brace parameters on node displacements, and validate the feasibility of the modified mechanical model. The optimization results and computational process information are compared among three cases of different strategies of parameter change as well. The accuracy is also verified by the calculation results of finite element model. This work can help the applications of PS-SCED brace optimization related to parameter sensitivity, and fulfill the systematic design procedure of PS-SCED brace-structure system with completed and prospective consequences.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2003년도 ICCAS
• /
• pp.84-89
• /
• 2003

Fuzzy controller is one of the succeed controller used in the process control in case of model uncertainties. But it my be difficult to fuzzy controller to articulate the accumulated knowledge to encompass all circumstance. Hence, it is essential to provide a tuning capability. There are many parameters in fuzzy controller can be adapted, scale factor tuning of normalized fuzzy controller is one of the adaptation parameter. Two adaptation methods are implemented in this work on an experimental thermal process, which simulate heating process in liquefied petroleum gases (LPG) recovery process in one of petrochemical industries: Gradient decent (GD) adaptation method; supervisory fuzzy controller. A comparison between the two methods is discussed.

• 한국연소학회:학술대회논문집
• /
• 한국연소학회 2000년도 제20회 KOSCO SYMPOSIUM 논문집
• /
• pp.76-84
• /
• 2000

A numerical study was conducted for the investigation of thermal choking process in a model scramjet engine based on the experimental results at the Australian National University. The results of numerical simulation showed that thermal choking process could be related to the interaction between hypersonic flow and fuel-air mixing process. Especially, we could make sure that turbulent mixing was most important parameter to the thermal choking process.