• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.17 no.8
• /
• pp.741-746
• /
• 2007

In this study, transfer path identification and output estimation are simulated by multi-dimension spectral analysis method (MDSA). Multi -input/single-output system give expression the vehicle suspension which each inputs are correlated reciprocally. In case of correlating with inputs, the system needs separating the each input signal by MDSA. Main simulations are about finding effective input by coherent output spectrum and selecting optimal input's number by multiple coherence function. Also, by shielding transfer path of each input, transfer path characteristic is identified in terms of overall integrated contribution level.

• Proceedings of the Korea Concrete Institute Conference
• /
• 2000.04a
• /
• pp.479-484
• /
• 2000

In this study, optimal design of reinforced concrete piers under seismic load is numerically investigated. Object function is the area of the concreate-section. Design variables are the total area of reinforcement and concrete-section dimension(Circular section diameter). Constraints of the design strength of the column, longitudinal reinforcement ratio and lower and upper bounds on the design variables are imposed. The reinforcement concrete column is analysed and designed by the Ultimated Strength Design method and load combination involving dead, live, wind and seismic load is used. For numerical optimization, ADS(Garret N, Vanderplaats_ routine is used. From the result of numerical examples, the concrete-section dimension was reduced, but longitudinal reinforcement was not changed. The results show that confinement reinforcement was reduced and confinement reinforcement spacing is increased. The higher strength of reinforcement used, the more concrete-section area was reduced.

• Proceedings of the KIEE Conference
• /
• 1997.07c
• /
• pp.1144-1148
• /
• 1997

In this paper, a method for the daily maximum load forecasting which uses a chaotic time series in power system and artificial neural network. We find the characteristics of chaos in power load curve and then determine a optimal embedding dimension and delay time. For the load forecast of one day ahead daily maximum power, we use the time series load data obtained in previous year. By using of embedding dimension and delay time, we construct a strange attractor in pseudo phase plane and the artificial neural network model trained with the attractor mentioned above. The one day ahead forecast errors are about 1.4% for absolute percentage average error.

• Proceedings of the KIEE Conference
• /
• 1995.07b
• /
• pp.578-580
• /
• 1995

In this paper, a method for the daily maximum load forecasting which uses a chaotic time series in power system and artificial neural network. We find the characteristics of chaos in power load curve and then determine a optimal embedding dimension and delay time, For the load forecast of one day ahead daily maximum power, we use the time series load data obtained in previous year. By using of embedding dimension and delay time, we construct a strange attractor in pseudo phase plane and the artificial neural network model trained with the attractor font mentioned above. The one day ahead forecast errors are about 1.4% of absolute percentage average error.

• Proceedings of the KIEE Conference
• /
• 1996.07b
• /
• pp.835-837
• /
• 1996

In this paper, a method for the daily maximum load forecasting which uses a chaotic time series in power system and artificial neural network(Back-propagation) is proposed. We find the characteristics of chaos in power load curve and then determine a optimal embedding dimension and delay time. For the load forecast of one day ahead daily maximum power, we use the time series load data obtained in previous year. By using of embedding dimension and delay time, we construct a strange attractor in pseudo phase plane and the artificial neural network model trained with the attractor mentioned above. The one day ahead forecast errors are about 1.4% for absolute percentage average error.

• Journal of the Korean Society of Safety
• /
• v.18 no.3
• /
• pp.22-28
• /
• 2003

This study proposes the construction of chaos simulator for cutting characteristics evaluation of non-ferrous metals. Also this paper aims to find the optimal cutting conditions of diamond turning machine by measuring surface form and roughness to perform the cutting experiment of non-ferrous metals, which are aluminum, with diamond tool. As well, according to change cutting conditions such as fled rate, using diamond turning machine to perform cutting processing, by measuring cutting force and surface roughness and according to cutting conditions the aluminum about cutting properties. Trajectory changes in the attractor indicated a substantial difference in fractal characteristics. Constructed chaos simulator in this study can be used for cutting characteristics evaluation of non-ferrous metals.

• International journal of advanced smart convergence
• /
• v.13 no.2
• /
• pp.80-87
• /
• 2024

This paper focuses on improving accuracy in constrained computing settings by employing the ReLU (Rectified Linear Unit) activation function. The research conducted involves modifying parameters of the ReLU function and comparing performance in terms of accuracy and computational time. This paper specifically focuses on optimizing ReLU in the context of a Multilayer Perceptron (MLP) by determining the ideal values for features such as the dimensions of the linear layers and the learning rate (Ir). In order to optimize performance, the paper experiments with adjusting parameters like the size dimensions of linear layers and Ir values to induce the best performance outcomes. The experimental results show that using ReLU alone yielded the highest accuracy of 96.7% when the dimension sizes were 30 - 10 and the Ir value was 1. When combining ReLU with the Adam optimizer, the optimal model configuration had dimension sizes of 60 - 40 - 10, and an Ir value of 0.001, which resulted in the highest accuracy of 97.07%.

• KOREAN JOURNAL OF PACKAGING SCIENCE & TECHNOLOGY
• /
• v.19 no.2
• /
• pp.61-65
• /
• 2013

Modified atmosphere (MA) of reduced $O_2$ and elevated $CO_2$ concentrations has been used for keeping the quality of fresh produce and extending the shelf life. As a way to attain the beneficial MA package around the produce, a gas diffusion tube or perforation can be attached onto the container and controlled on real time in its opening/closing responding to $O_2$ and $CO_2$ concentrations measured by gas sensors. The timely-controlled opening of the gas diffusion tube can work in harmony with the produce respiration and help to create the desired MA. By use of the mathematical modeling, the effect of tube dimension on the controlled container atmosphere was figured out in this study. Spinach and king oyster mushroom were used as typical commodities for designing the model container system (0.35 and 0.9 kg in 13 L, respectively) because of their respiration characteristics and the optimal MA condition ($O_2$ 7~10%/$CO_2$ 5~10% for spinach; $O_2$ 2~5%/$CO_2$ 10~15% for mushroom). With a control logic for the gas composition to stay as close as possible to optimum MA window without invading injurious low $O_2$ and/or high $CO_2$ concentrations, the atmosphere of the sensor-controlled container could stay at its lower $O_2$ boundary or upper $CO_2$ limit under certain tube dimensional conditions. There were found to be the ranges of the tube diameter and length allowing the beneficial MA. The desired range of the tube dimension for spinach consisted of combinations of larger diameter and shorter length in the window of 0.3~2 cm diameter and 0.2~10 cm length. Similarly, that for king oyster mushroom was combinations of larger diameter and shorter length in the window of 0.9~2 cm diameter and 0.2~3 cm in length. Clear picture on generally affordable tube dimension range may be formulated by further study on a wide variety of commodity and pack conditions.

• Journal of the Korea Society of Computer and Information
• /
• v.17 no.1
• /
• pp.161-169
• /
• 2012

In general, Euclidean minimum spanning tree is a tree connecting input nodes with minimum connecting cost. But the tree may not be optimal when applied to real world problems of three dimension. In this paper, three dimension Euclidean minimum spanning tree is proposed, connecting all input nodes of 3-dimensional space with minimum cost. In experiments for 30,000 input nodes with 100% space ratio, the tree produced by the proposed method can reduce 90.0% connection cost tree, compared with the tree by two dimension Prim's minimum spanning tree. In two dimension plane, the proposed tree increases 251.2% connecting cost, which is pointless in 3-dimensional real world. Therefore, the proposed method can work well for many connecting problems in real world space of three dimensions.

• Proceedings of the Korea Inteligent Information System Society Conference
• /
• 1999.10a
• /
• pp.365-373
• /
• 1999

Recently, numerous studies have demonstrated that artificial intelligence such as neural networks can be an alternative methodology for classification problems to which traditional statistical methods have long been applied. In building neural network model, the selection of independent and dependent variables should be approached with great care and should be treated as a model construction process. Irrespective of the efficiency of a learning procedure in terms of convergence, generalization and stability, the ultimate performance of the estimator will depend on the relevance of the selected input variables and the quality of the data used. Approaches developed in statistical methods such as correlation analysis and stepwise selection method are often very useful. These methods, however, may not be the optimal ones for the development of neural network models. In this paper, we propose a genetic algorithms approach to find an optimal or near optimal input variables for neural network modeling. The proposed approach is demonstrated by applications to bankruptcy prediction modeling. Our experimental results show that this approach increases overall classification accuracy rate significantly.