• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2000년도 추계학술대회 논문집 학회본부 D
• /
• pp.666-668
• /
• 2000

In this paper, a fuzzy prediction method is proposed for time series data having uncertainty and non-stationary characteristics. Conventional methods, which use past data directly in prediction procedure, cannot properly handle non-stationary data whose long-term mean is floating. To cope with this problem, a data preprocessing technique utilizing the differences of original time series data is suggested. The difference sets are established from data. And the optimal difference set is selected for input of fuzzy predictor. The proposed method based the Takigi-Sugeno-Kang(TSK or TS) fuzzy rule. Computer simulations show improved results for various time series.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2005년도 ICCAS
• /
• 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.

• 산업기술연구
• /
• 제28권B호
• /
• pp.101-109
• /
• 2008

In general, non-stationary or chaos time series forecasting is very difficult since there exists a drift and/or nonlinearities in them. To overcome this situation, we suggest a new prediction method based on multiple model TS fuzzy predictors combined with preprocessing of time series data, where, instead of time series data, the differences of them are applied to predictors as input. In preprocessing procedure, the candidates of optimal difference interval are determined by using con-elation analysis and corresponding difference data are generated. And then, for each of them, TS fuzzy predictor is constructed by using k-means clustering algorithm and least squares method. Finally, the best predictor which minimizes the performance index is selected and it works on hereafter for prediction. Computer simulation is performed to show the effectiveness and usefulness of our method.

• 한국전산유체공학회:학술대회논문집
• /
• 한국전산유체공학회 2004년도 추계 학술대회논문집
• /
• pp.33-36
• /
• 2004

The ballast tank of a ship is a system that realizes the required shipping condition and controls the draft of a ship. The loading/unloading of the ballast tank is frequently operated during navigation and the accurate prediction of the loading/unloading time is very important. A numerical algorithm that predicts the loading/unloading time of the ballast tank has been developed and applied to the prediction of the loading/unloading time of the ballast tank with various piping systems. This algorithm can be useful in optimizing the ballast tank system in early design stage.

• 대한기계학회논문집A
• /
• 제24권11호
• /
• pp.2812-2822
• /
• 2000

Several methods have been developed to predict the rupture time of the boiler tubes in thermal power plant. However, existing life prediction methods give very conservative value at operating stress of power plant and rupture strain cannot be well estimated. Therefore, in this study, rupture time and strain prediction method accounting for creep, corrosion and heat transfer is newly proposed and compared with the current research results. The creep damage evolves by continuous cavity nucleation and constrained cavity growth. The corrosion damage evolves by steam side and fire side corrosion. The results showed good correlation between the theoretically predicted rupture time and the current research results. And rupture strain may be well estimated by using the proposed method.

• 한국지능시스템학회논문지
• /
• 제7권3호
• /
• pp.34-42
• /
• 1997

본 논문은 새로은 퍼지 규칙의 생성을 위한 학습 알고리즘과 시계열 예측에의 응용을 다루고 있다. 데이터에서 IF-THEN문 형태의 퍼지 규칙을 생성시키는 과정에서 동일한 전건부(IF문)에 대해 상이한 후건부(THEN문)가 생겨 모순된 규칙을 형성시키는 경향이 있다. 수정된 중심값 방법(Modified Center Method)으로 명명된 새로운 알고리즘은 이와 같은 모순된 규칙의 형성을 효과적으로 해결하여, 시계열 예측을 수행하는데 그 오차를 줄일 수 있다. 알고리즘의 효과를 살표보기 위해 Mackey-Glass time series와 Gas Furnace data 분석에 적용하였다.

• 응용통계연구
• /
• 제16권2호
• /
• pp.293-303
• /
• 2003

개입효과가 포함된 시계열 자료에 대한 여러 시계열 모형에 의한 예측 방법들이 비교 분석된다. 개입이 있는 선형 ARIMA 모형, 비선형 ARCH 모형 및 개입이 있는 비선형 ARCH 모형 그리고 TONG 이 제안한 결합예측방법들이 소개되고, 실증분석으로 개입이 있다고 생각되는 한국건축허가면적 자료로부터 그 예측 수월성이 비교된다.

• 물과 미래
• /
• 제28권5호
• /
• pp.219-233
• /
• 1995

강우의 평균과 분산이 시 공간적으로 변하는 비정상 다변량 모형을 강우모형으로 선정하였다. 그리고 강우모형의 상태 및 매개변수의 추정을 위해 비정상 대변량 모형의 잔차항에 Kalman Filter 순환추정 알고리즘을 적용하여 강우예측모형 시스템을 구성하였다. 그후 반응시간이 짧은 도시지역에 설치된 T/M 강우관측소에 입력되는 매 시간(10분간격) 강우자료를 사용하여 호우개수방법에 의한 비정상(Non-stationary) 평균과 분산의 추정 그리고 호우속도 추정을 통한 정규잔차 공분산을 추정하여 다수의 지점들 및 선행시간들의 실시간 다변량 단기 강우예측 (On-line, Real-time, Multivariate Short-term, Rainfall Prediction)을 하였다. 강우예측시스템 모형에 의한 결과와 비정상 변량 모형에 의한 강우모의 결과가 잘 일치하였다. 그리고 예측정도를 측정하는 방법인 제곱 평균 제곱근 오차(RMSE)와 모형 효율성 계수(ME)를 분석한 결과, 강우 예측시간 즉 선행시간이 갈수록 제곱 평균 제곱근 오차가 커지고 모형 효율성 계수가 1로부터 점차 작아지는 것으로 보아 강우예측 정도가 떨어지는 것을 알 수 있었다. 또한 호우개수방법으로 구한 평균이 호우구조의 많은 부분을 차지하고 있음을 알 수 있었다.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2003년도 ICCAS
• /
• pp.2078-2081
• /
• 2003

This paper presents the studies on prediction models of tropospheric scintillation. The prediction scintillation models are Karasawa and ITU-R , which can be improved for different locations and circumstances. In this paper, the investigation of average time between variance ${\sigma}_n\;^2$ and the wet part of refractivity $N_{wet}$ under various conditions of meteorological parameters have been carried out at King Mongkut’s Institute of Technology Lankrabang , Bangkok , Thailand , in the range of Ku-band (12.260 GHz) on high elevation angle from Thaicom2 satellite. From the studies results shows that average period of time of 30 days are best suitable for find out the relation between average time variance ${\sigma}_n\;^2$ and the wet part of refractivity $N_{wet}$ according to Karasawa model, the average time variance is express as ${\sigma}_n\;^2=(0.003N_{wet}-0.1313)^2$ , the appropriation model for occurrence of scintillation has been analyzed and experimental results are carried out.

• Journal of Welding and Joining
• /
• 제31권5호
• /
• pp.15-19
• /
• 2013

In order to reduce a grand compute time in prediction of welding distortion and residual stress by 3D thermal elastic plastic analysis, idealization of welding that is methods to heat input simultaneously in all weld metal on the same welding direction is carried out on two weld joints(butt welding and fillet welding). Then, the accuracy of acquired results is investigated through the comparison of the high accuracy prediction results. The thermal conduction analysis results by idealization of welding, the temperature is raised accompany with beginning of heat input because all of weld metal is heated input at the same time. On the other side, the temperature witch predicted with high accuracy is raised at the moment heating source passes the measuring points. So, there is difference of time between idealization of welding and considering of moving heat source faithfully. However, temperature history by idealization of welding is well simulated a high accuracy prediction results.