• 제어로봇시스템학회:학술대회논문집
• /
• 2005.06a
• /
• pp.920-924
• /
• 2005

This paper presents adaptive learning data of evolvable neural networks (ENNs) for time series prediction of nonlinear dynamic systems. ENNs are a special class of neural networks that adopt the concept of biological evolution as a mechanism of adaptation or learning. ENNs can adapt to an environment as well as changes in the environment. ENNs used in this paper are L-system and DNA coding based ENNs. The ENNs adopt the evolution of simultaneous network architecture and weights using indirect encoding. In general just previous data are used for training the predictor that predicts future data. However the characteristics of data and appropriate size of learning data are usually unknown. Therefore we propose adaptive change of learning data size to predict the future data effectively. In order to verify the effectiveness of our scheme, we apply it to chaotic time series predictions of Mackey-Glass data.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.18 no.2
• /
• pp.146-153
• /
• 2015

An unmanned aerial photography method by using an unmanned helicopter is useful method for measuring of the water-entry traces of small falling objects into the sea. Pixel sizes on the aerial photograph may be too large due to a limit of camcorder resolution and a wide shooting area. If the pixel size is too large, identification of water-entry trace is impossible. Thus an accurate prediction of water-entry trace size is required. The traces of water-entry could be classified into three types such as splash, water column, and bubble. Diameters of each trace are predicted by water-entry impact pressure theories, cavity theories, and trial test results. The results are verified by drop tests using an unmanned helicopter at two water-entry speeds. As a result, prediction and test results showed sufficient similarity to evaluate the identifiability of water-entry trace.

• International Journal of Naval Architecture and Ocean Engineering
• /
• v.13 no.1
• /
• pp.321-339
• /
• 2021

Response Surface Method (RSM) has been widely used for flammable cloud size prediction as it can reduce computational intensity for further Explosion Risk Analysis (ERA) especially during the early design phase of offshore platforms. However, RSM encounters the overfitting problem under very limited simulations. In order to overcome the disadvantage of RSM, Bayesian Regularization Artificial Neural (BRANN)-based model has been recently developed and its robustness and efficiency have been widely verified. However, for ERA during the early design phase, there seems to be room to further reduce the computational intensity while ensuring the model's acceptable accuracy. This study aims to develop an integrated method, namely the combination of Center Composite Design (CCD) method with Bayesian Regularization Artificial Neural Network (BRANN), for flammable cloud size prediction. A case study with constant and transient leakages is conducted to illustrate the feasibility and advantage of this hybrid method. Additionally, the performance of CCD-BRANN is compared with that of RSM. It is concluded that the newly developed hybrid method is more robust and computational efficient for ERAs during early design phase.

• Journal of the Korean Society of Clothing and Textiles
• /
• v.46 no.3
• /
• pp.454-463
• /
• 2022

This study aims to provide basic size data for the development of a hip dislocation prevention garment for cerebral palsy (CP) children and useful information for the design of garment products for CP children through identifying differences in body shape between CP and non-CP children and reviewing the tibial-stature prediction formulas of previous studies. Forty-seven Korean children with CP aged 2 to 14 years were measured for body size from October 2019 to August 2020. Body measurements of 18 sites, including greater trochanter length, which is an important site for a hip dislocation prevention garment, were collected and analyzed. Data of non-CP children were taken from same age of Size Korea and compared. Tibial-stature prediction formulas suggested in four previous studies were also reviewed. CP children had significantly lower stature as well as circumferential dimensions when compared to non-CP children. Greater trochanter length is difficult to predict through other body dimensions. Thus, direct measurement is required. Of the general key dimensions used in the clothing industry, only hip circumference could explain the body shape of CP children. Tibial-stature prediction formulas cannot always but tend to largely predict the actual stature of CP children.

• 제어로봇시스템학회:학술대회논문집
• /
• 1986.10a
• /
• pp.547-549
• /
• 1986

This paper deals with large scale distributed packet switching system which is modeled by state space form and optimizing routing algorithms and buffer size via a hierachical system optimization method, the interaction prediction method.

• Journal of Broadcast Engineering
• /
• v.17 no.6
• /
• pp.976-989
• /
• 2012

In this paper, we propose an efficient motion prediction architecture and DPB design mechanism for scalable multi-view video coding which is implemented by integrating SVC and MVC coding algorithms. In the proposed motion prediction architecture, we employ pictures associated with other views as a candidate for reference picture for improved motion prediction performance. By the proposed prediction architecture, we could enormously reduce the size of compressed video data. When performing scalable multi-view video coding, an integrated DPB design mechanism is also proposed. It is shown by various simulations that the proposed motion prediction architecture for scalable multi-view video coding can result in reduced data size in the compressed bitstream.

• Journal of the Korean Data and Information Science Society
• /
• v.18 no.4
• /
• pp.945-953
• /
• 2007

In this paper, we propose a smoothing and thus noise-reducing method of data sequences for stock price prediction with hidden Markov models, HMMs. The suggested method just uses simple moving average. A proper average size is obtained from forecasting experiments with stock prices of bank sector of Korean Exchange. Forecasting method with HMM and moving average smoothing is compared with a conventional method.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.30 no.11 s.254
• /
• pp.1455-1460
• /
• 2006

The relationship between fatigue crack growth behavior and cyclic crack tip opening displacement is studied. An elastic-plastic finite element analysis (FEA) is performed to examine the growth behavior of fatigue crack, where the contact elements are used in the mesh of the crack tip area. We investigate the relationship between the reversed plastic zone size and the changes of the cyclic crack tip opening displacement along the crack growth. We investigate the effect of the element size when predict fatigue crack opening behavior using the cyclic crack tip opening displacement obtained from FEA. The cyclic crack tip opening displacement is related to fatigue crack opening behavior.

• Computers and Concrete
• /
• v.18 no.5
• /
• pp.937-950
• /
• 2016

This study applied autoclave expansion and heat curing to accelerate the hydration of concrete and investigated how these methods affect the expansion rate, crack pattern, aggregate size effect, and expansion of electric arc furnace oxidizing slag (EOS)-containing concrete. An expansion prediction model was simulated to estimate the expansion behavior over a long period and to establish usage guidelines for EOS aggregates. The results showed that the EOS content in concrete should range between 20% and 30% depending on the construction conditions, and that coarse aggregates with a diameter of ${\geq}4.75-mm$ are not applicable to construction engineering. By comparison, aggregates with a size of 1.18-0.03 mm resulted in higher expansion rates; these aggregates can be used depending on the construction conditions. On Day 21, the prediction model attained a coefficient of determination ($R^2$) of at least 0.9.

• Journal of Power System Engineering
• /
• v.18 no.6
• /
• pp.13-20
• /
• 2014

The steady-state, incompressible and three-dimensional numerical analysis was carried out to evaluate turbulent models on the aerodynamic performance of a small-size axial fan(SSAF). The prediction performance on the static pressure of all turbulent models is going downhill at the high static pressure and low flowrate region, but has improved at the axial flow region. In consequence, all turbulent models predict the static pressure coefficient with an error performance less than about 4% after the region of the flowrate coefficient of about 0.14. Especially, the turbulent model of SST $k-{\omega}$ shows the best prediction performance equivalent to an error performance less than about 2% on the static pressure.