• 한국지리정보학회지
• /
• 제12권1호
• /
• pp.106-118
• /
• 2009

1m급 전후의 고해상도 위성영상들이 지구관측 및 모니터링으로부터 국토의 디지털 지도제작에 이르기까지 폭넓게 활용되고 있다. IKONOS 영상의 경우 Pro와 Precision 제품은 상당히 고가이므로 정확한 지도를 제작하는데 저가의 Geo제품과 영상공급자에 의해 제공된 RPC계수를 이용하는 것은 바람직하다. IKONOS 고해상도 영상은 엄밀한 센서모델 대신 RF에 의해 설명되어진다. 본 연구에서는 RF 모델을 기반으로 추출되는 지상좌표의 정확도를 개선하기 위하여 대상물공간과 상공간에서 각각 정의된 4개의 모델, 즉 선형이동 모델, 축척 및 선형이동 모델, Affine 모델, 2차 다항식 모델을 제시하였다. RF 모델을 기반으로 한 지상좌표 산출 알고리즘과 산출된 지상좌표의 정확도 개선 알고리즘을 개발하고 실험을 통하여 다항식 보정모델별 개선효과를 분석하였다. 또한 지상기준점의 수와 배치유형, 측량의 정확도와 같은 여러 가지 지도제작 매개변수들이 지상좌표의 정확도에 미치는 효과를 평가하였다. 적용실험 결과, RF 모델에 의해 1차적으로 산출된 3차원 지상좌표의 RMSE는 X 방향에서 8.035m, Y 방향에서 10.020m, Z 방향에서 13.318m이었으나 다항식 보정 알고리즘을 통하여 X 방향 2.791m, Y 방향 2.520m, Z 방향 1.441m까지 RMSE를 낮춤으로써 수평위치에서 약 65%, 수직방향에서 약 89%까지 정확도를 크게 개선할 수 있었다.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2000년도 제15차 학술회의논문집
• /
• pp.15-15
• /
• 2000

In the past couple of years, there has been increasing interest in the fusion of neural networks and fuzzy logic. Most of the existing fused models have been proposed to implement different types of fuzzy reasoning mechanisms and inevitably they suffer from the dimensionality problem when dealing with complex real-world problem. To overcome the problem, we propose the self-organizing networks with activation nodes based on fuzzy inference and polynomial function. The proposed model consists of two parts, one is fuzzy nodes which each node is operated as a small fuzzy system with fuzzy implication rules, and its fuzzy system operates with Gaussian or triangular MF in Premise part and constant or regression polynomials in consequence part. the other is polynomial nodes which several types of high-order polynomials such as linear, quadratic, and cubic form are used and are connected as various kinds of multi-variable inputs. To demonstrate the effectiveness of the proposed method, time series data for gas furnace process has been applied.

• 한국농공학회논문집
• /
• 제46권2호
• /
• pp.67-75
• /
• 2004

This study aims to investigate the dynamic behaviour of a parabolic arch with initial deflection by using the elasto-plastic finite element model where the von-Mises yield criteria have been adopted. The initial deflection of arch was assumed by the high order polynomial of ${\omega}_i$ = ${\omega}_o$${(1-{(2x/L)}^m)}^n$) and the sinusoidal profile of ${\omega}_i$ = ${\omega}_o$$\sin$(n$\pi$x/L). Several numerical examples were tested considering symmetric initial deflection modes when the maximum initial deflection of an arch is fixed as L/500, L/1000, L/2000 or L/5000. The effects of polynomials order on the dynamic behavior of arch were not conspicuous. The most unfavorite dynamic response occurs when the maximum initial deflection varies from L/1000 to L/4000 if the initial deflection mode is represented by high order polynomials.

• 대한전기학회논문지
• /
• 제34권12호
• /
• pp.472-484
• /
• 1985

A design of robust voltage controller for high speed excitation of synchronous machine was carried out by pole assignment techniques. An affine map from characteristic polynomial coefficients to feedback parameters is formulated in order to place the system eigen values in the desired region. The feedback parameters determined from linearized model are tested on nonlinear model subjecting it to small disturbances and system faults to show the effectiveness of the controller designed by the proposed technique. The results obtained indicate that the controller presented improves the dynamic stability and system performances of conventionally controlled synchronous machine significantly.

• 전기학회논문지P
• /
• 제54권1호
• /
• pp.34-40
• /
• 2005

This paper presents a frequency transfer function synthesis for simplifying a high-order model with time delay to a low-order model. A model reduction is based on minimizing the error function weighted by the numerator polynomial of reduced systems. The proposed method provides better low frequency fit and a computer aided algorithm. And in this paper, we present a design method of PID controller for achieving the desired specifications via the reduced model. The proposed method identifies the parameter vector of PID controller from a linear system that develops from rearranging the two dimensional input matrices and output vectors obtained from the frequency bounds.

• 조명전기설비학회논문지
• /
• 제18권6호
• /
• pp.176-182
• /
• 2004

본 논문에서는 시간지연을 갖는 고차모델을 저차모델로 간소화하기 위한 주파수 전달함수 합성법을 제안한다. 모델축소는 축소시스템의 분자다항식에 의해 가중된 오차함수를 최소화하는 것에 기반을 두고 있다. 제안된 방법은 보다 우수한 저주파수 적합도를 제공한다. 그리고 네 개의 예제가 제안된 방식의 유용성을 나타내기 위해서 주어진다.

• 보건교육건강증진학회지
• /
• 제12권2호
• /
• pp.36-47
• /
• 1995

This research was obtained from analyzing how the physiques of the 3rd grade students of high school for males and females and developed for the last eleven years(from 1983 to 1993). By the physiques and nutritional index of physical growth and development, Relative Body Weight of 36.62 exceeded the standard, on the other hand females showed lower records than the standard. Relative Chest Girth Index belonged to the normal type of males and females in all, in the comparison of the records between 1983 and 1993, males increased in average 0.29 and females in average 0.55. Relative Chest Girth Index of females was greater than that of females. By the results of Relative Sitting Height Index, growth of the lower body for males and females was greater than that of males. In case of Vervaeck Index, males increased in average 2.04 but females increased in average 1, 20 relatively less than males. These phenomena provided for the evidence of the deficient nutrition in females. In the regression models of body height and body weight within a certain period, statistical regression model types which best indicated chronological average changes of body height and body weight, took 3rd Order Polynomial Regression Model rather than linear regression model. In females, statistical regression model types which best is suitable for chronological average change of body height and body weight, took 4th and 2nd Order Polynomial Regression Model respectively. The prediction value of 1995 by estimated polynomial regression model anticipated that body height of 3rd grade year students of high school of males in 1993 went on increasing from 170.87cm to 171.79cm in average 0.92cm growth and that of females from 158.99cm to 160.79cm in average 1.80cm growth. In addition, body weight of males seemed to increase from 62.58kg to 64.52kg in average 1.94kg growth and that of females seemed to increase from 54.05kg to 54.19kg in average 0.14kg growth. Linear Regression Model was suitable for the regression model of body weight for body height. Prediction on increase of an average body weight for body height was that, according to growth of body height 1cm in males, body weight increased 1.41kg averagely and that of females 0.86kg. For that reason, we came to conclusion that body weight increase for body height 1cm in males was greater than that in females on average.

• 한국환경과학회지
• /
• 제30권5호
• /
• pp.399-406
• /
• 2021

The dissolution of ionized gas in dielectric barrier plasma, similar to the principle of ozone generation, is a major performance-affecting factor. In this study, the plasma gas dissolving performance of a gas mixing-circulation plasma process was evaluated using an experimental design methodology. The plasma reaction is a function of four parameters [electric current (X₁), gas flow rate (X₂), liquid flow rate (X₃) and reaction time (X₄)] modeled by the Box-Behnken design. RNO (N, N-Dimethyl-4-nitrosoaniline), an indictor of OH radical formation, was evaluated using a quadratic response surface model. The model prediction equation derived for RNO degradation was shown as a second-order polynomial. By pooling the terms with poor explanatory power as error terms and performing ANOVA, results showed high significance, with an adjusted R² value of 0.9386; this indicate that the model adequately satisfies the polynomial fit. For the RNO degradation, the measured value and the predicted values by the model equation agreed relatively well. The optimum current, gas flow rate, liquid flow rate and reaction time were obtained for the highest desirability for RNO degradation at 0.21 A, 2.65 L/min, 0.75 L/min and 6.5 min, respectively.

• 한국정보통신설비학회:학술대회논문집
• /
• 한국정보통신설비학회 2008년도 정보통신설비 학술대회
• /
• pp.218-222
• /
• 2008

RF power amplifier requires linearization in order to reduce adjacent channel interference. And most of the existing linearization algorithms assume that a PA has memory-less nonlinearity. But for the wider bandwidth signal, the memory effect of PA cannot be ignored. This paper investigates digital pre-distortion by use of a memory polynomial model which compensates for amplifier nonlinearity and memory effect. The look-up table based implementation scheme is used to reduce the computational complexity of the pre-distortion block. The linearization performance is demonstrated on wideband CDMA signal and class AB high power amplifier.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2003년도 하계학술대회 논문집 D
• /
• pp.2519-2521
• /
• 2003

In this paper, we propose Self-Organizing Fuzzy Polynomial Neural Networks(SOFPNN) architecture for optimal model identification and discuss a comprehensive design methodology supporting its development. It is shown that this network exhibits a dynamic structure as the number of its layers as well as the number of nodes in each layer of the SOFPNN are not predetermined (as this is the case in a popular topology of a multilayer perceptron). As the form of the conclusion part of the rules, especially the regression polynomial uses several types of high-order polynomials such as linear, quadratic, and modified quadratic. As the premise part of the rules, both triangular and Gaussian-like membership function are studied and the number of the premise input variables used in the rules depends on that of the inputs of its node in each layer. We introduce two kinds of SOFPNN architectures, that is, the basic and modified one with both the generic and the advanced type. The superiority and effectiveness of the proposed SOFPNN architecture is demonstrated through nonlinear function numerical example.