• 한국소성가공학회:학술대회논문집
• /
• 한국소성가공학회 1999년도 압출 및 인발 심포지엄
• /
• pp.56-59
• /
• 1999

In the present study, in order to investigate the effect of the corner filling in the drawing of the rectangular rod from a round bar, the drawing of the square rod from a round bar has been simulated by using rigid-plastic finite element method and artificial neural network has been introduced to reduce the number of simulation. The experimental investigation has been also implemented to verify the efficiency of the application of results of present and previous study. According to the results of present and pervious study, the combination of semi-die angle gives a great effect on the corner filling in case of the irregular shaped drawing process, but, in case of the regular shaped drawing process, the main process variable on the corner filling is reduction in area.

• 한국소성가공학회:학술대회논문집
• /
• 한국소성가공학회 2006년도 춘계학술대회 논문집
• /
• pp.70-73
• /
• 2006

In the present study, the flow-softening behavior occurring during high temperature deformation of AZ31 Mg alloy was investigated. Flow softening of AZ31 Mg alloy was attributed to (1) thermal softening by deformation heating and (2) microstructural softening by dynamic recrystallization. Artificial neural networks method was used to derive the accurate amounts of thermal softening by deformation heating. A series of mechanical tests (High temperature compression and load relaxation tests) was conducted at various temperatures ($250^{\circ}C{\sim}500^{\circ}C$) and strain rates ($10^{-4}/s{\sim}100/s$) to formulate the recrystallization kinetics and grain size relation. The effect of DRX kinetics on microstructure evolution (fraction of recrystallization) was evaluated by the unified SRX/DRX (static recrystallization/dynamic recrystallization) approaches

• 한국소성가공학회:학술대회논문집
• /
• 한국소성가공학회 1999년도 제3회 압연심포지엄 논문집 압연기술의 미래개척 (Exploitation of Future Rolling Technologies)
• /
• pp.438-446
• /
• 1999

Crop minimization of the top and bottom ends of hot rolled plate, in a plate, in a plate mill, has been investigated. The existing model to determine the edging pattern at the finishing rolling pass was not reasonable to get high width accuracy and rolling yields. New models including width prediction have been formulated by using neural network model of back propagation learning algorithm and statistical analysis based on the actual production rolling data to give the optimal pattern for minimizing trimming loss. Using these models, at a given rolling condition of broadside pass and finishing pass and the permissible condition of width variation, it was possible to minimize crip at the top and bottom ends according to optimum procedure in plate mill. An application to improve the plan view pattern reduced width variation by 23% and crop length by 30% on average with an effective fishtail crop shape.

• 한국소성가공학회:학술대회논문집
• /
• 한국소성가공학회 1997년도 춘계학술대회논문집
• /
• pp.144-147
• /
• 1997

Process design of multi-step wire drawing process, conducted by means of finite element analysis and ANN(Artificial Neural Network), has been considered. The investigated problem involves the adequate selection of the drawing die angle and the correspondent reduction rate sequence in the condition of desired initial and final diameter. Combinations of the process parameters which are used in finite element simulation are selected by using orthogonal array. Also the orthogonal array and the results of finite element simulation which are related to the process energy are used as train data of ANN. In this study, it is shown that the new technique using ANN is useful method in application to the wide range of metal forming process.

• 한국소성가공학회:학술대회논문집
• /
• 한국소성가공학회 1999년도 춘계학술대회논문집
• /
• pp.84-87
• /
• 1999

This paper presents a preform design method that combines the analytic method and inference of known knowledge with neural network. The analytic method is a finite element method that is used to simulate backward extrusion with pre-defined process parameters. The multi-layer network and back-propagation algorithm are utilized to learn the training examples from the simulation results. The design procedures are utilized to learn the training examples from the simulation results. The design procedures are two methods the first the neural network infer the deformed shape from the pre-defined processes parameters. The other the network infer the processes parameters from deformed shape. Especially the latest method is very useful to design the preform From the desired feature it is possible to determine the processes parameters such as friction stroke and tooling geometry. The proposed method is useful for shop floor to decide the processes parameters and preform shapes for producing sound product.

• 한국소성가공학회:학술대회논문집
• /
• 한국소성가공학회 1999년도 춘계학술대회논문집
• /
• pp.99-102
• /
• 1999

In this study, to investigate the effect of process variables such as reduction in area, semi-die angle and the rectangular ratio to the corner filling which influences the dimensional accuracy of the final product in the drawing of the cluadrangle rod from a round bar, it has been simulated by three dimensional rigid-plastic finite element method. In order to reduce the number of simulation artificial neural network has been introduced. Also, through the experimental investigation, the present results have been implemented on the industrial product. In results, the main process variable is the combination of the semi-die angle in case of the irregular shaped drawing process and reduction in area in the event of regular shaped drawing process, respectively.

• 한국소성가공학회:학술대회논문집
• /
• 한국소성가공학회 2009년도 춘계학술대회 논문집
• /
• pp.357-360
• /
• 2009

The cooling stage greatly affects the product quality in the injection molding process. The cooling system that minimizes temperature variance in the product surface will improve the quality and the productivity of products. In this research, we tried the back-propagation algorithm of artificial neural network to find an optimum solution in the cooling system design of injection mold. The cooling system optimization problem that was once solved by a response surface method with 4 design variables was solved by applying the back-propagation algorithm, resulting in a solution with a sufficient accuracy. Furthermore the number of training points was much reduced by applying the fractional factorial design without losing solution accuracy.

• 한국소성가공학회:학술대회논문집
• /
• 한국소성가공학회 2005년도 춘계학술대회 논문집
• /
• pp.389-392
• /
• 2005

In the modern days, a galvannealed sheet steel(GA) instead of a cold rolled steel sheet has been widely used as an alternative to extend the life of automotive body. Accordingly, the mechanical properties of GA for automobiles were taken Into account and studied by comparing with the temperature variation on annealing in this study. To clarify the effect of surface features in the mechanical properties of GA, the several tests such as nanoindentation and FE-analysis were executed. For this goal use is made of the method of neural networks. The developed neural networks apply also to obtain reliable mechanical properties of the thin films. Load-displacement curve was computed by the analysis procedure and compared with experimental results.

• 로봇학회논문지
• /
• 제1권2호
• /
• pp.163-171
• /
• 2006

This paper introduces the research progress on the artificial brain in the Telerobotics and Control Laboratory at KAIST. This series of studies is based on the assumption that it will be possible to develop an artificial intelligence by copying the mechanisms of the animal brain. Two important brain mechanisms are considered: spike-timing dependent plasticity and dopaminergic plasticity. Each mechanism is implemented in two coding paradigms: spike-codes and rate-codes. Spike-timing dependent plasticity is essential for self-organization in the brain. Dopamine neurons deliver reward signals and modify the synaptic efficacies in order to maximize the predicted reward. This paper addresses how artificial intelligence can emerge by the synergy between self-organization and reinforcement learning. For implementation issues, the rate codes of the brain mechanisms are developed to calculate the neuron dynamics efficiently.

• 소성∙가공
• /
• 제30권6호
• /
• pp.311-322
• /
• 2021