• Corrosion Science and Technology
• /
• 제19권6호
• /
• pp.302-309
• /
• 2020

This study aimed to predict the pitting corrosion characteristics of AL-6XN super-austenitic steel using multiple linear regression. The variables used in the model are degree of sensitization, temperature, and pH. Experiments were designed and cyclic polarization curve tests were conducted accordingly. The data obtained from the cyclic polarization curve tests were used as training data for the multiple linear regression model. The significance of each factor in the response (critical pitting potential, repassivation potential) was analyzed. The multiple linear regression model was validated using experimental conditions that were not included in the training data. As a result, the degree of sensitization showed a greater effect than the other variables. Multiple linear regression showed poor performance for prediction of repassivation potential. On the other hand, the model showed a considerable degree of predictive performance for critical pitting potential. The coefficient of determination (R2) was 0.7745. The possibility for pitting potential prediction was confirmed using multiple linear regression.

• Computers and Concrete
• /
• 제16권3호
• /
• pp.343-356
• /
• 2015

1078 sets of mixtures in total that include fly ash, slag, and/or silica fume have been collected for prediction on concrete properties. A new database platform (Compos) has been developed, by which the stepwise multiple linear regression (SMLR) and BP artificial neural networks (BP ANNs) programs have been applied respectively to identify correlations between the concrete properties (strength, workability, and durability) and the dosage and/or quality of raw materials'. The results showed obvious nonlinear relations so that forecasting by using nonlinear method has clearly higher accuracy than using linear method. The forecasting accuracy rises along with the increasing of age and the prediction on cubic compressive strength have the best results, because the minimum average relative error (MARE) for 60-day cubic compressive strength was less than 8%. The precision for forecasting of concrete workability takes the second place in which the MARE is less than 15%. Forecasting on concrete durability has the lowest accuracy as its MARE has even reached 30%. These conclusions have been certified in a ready-mixed concrete plant that the synthesized MARE of 7-day/28-day strength and initial slump is less than 8%. The parameters of BP ANNs and its conformation have been discussed as well in this study.

• 한국전자파학회논문지
• /
• 제23권8호
• /
• pp.1014-1017
• /
• 2012

본 논문에서는 고해상도 SAR(Synthetic Aperture Radar) 영상 획득을 위해 방위각 방향 해상도를 향상시키기 위한 방법을 소개하였다. SAR 영상 획득을 위해 ${\omega}-k$(omega-k) 알고리즘을 이용하였으며, 2차원 주파수 영역에서 방위각 방향으로 AR(Auto-Regressive) 방법을 이용한 외삽을 이용하여 해상도를 향상시켰다. AR 방법은 선형 예측(linear prediction) 모델을 기반으로 한 외삽 기법이다. AR 방법을 이용한 외삽 기법 중에서 Burg 알고리즘을 이용하여 예측 차수(prediction order)와 표적의 거리에 따른 성능 비교 결과를 보여 준다.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2004년도 ICCAS
• /
• pp.631-634
• /
• 2004

In this paper, the Bayesian recurrent neural network (BRNN) is proposed to predict time series data. Among the various traditional prediction methodologies, a neural network method is considered to be more effective in case of non-linear and non-stationary time series data. A neural network predictor requests proper learning strategy to adjust the network weights, and one need to prepare for non-linear and non-stationary evolution of network weights. The Bayesian neural network in this paper estimates not the single set of weights but the probability distributions of weights. In other words, we sets the weight vector as a state vector of state space method, and estimates its probability distributions in accordance with the Bayesian inference. This approach makes it possible to obtain more exact estimation of the weights. Moreover, in the aspect of network architecture, it is known that the recurrent feedback structure is superior to the feedforward structure for the problem of time series prediction. Therefore, the recurrent network with Bayesian inference, what we call BRNN, is expected to show higher performance than the normal neural network. To verify the performance of the proposed method, the time series data are numerically generated and a neural network predictor is applied on it. As a result, BRNN is proved to show better prediction result than common feedforward Bayesian neural network.

• KEPCO Journal on Electric Power and Energy
• /
• 제5권3호
• /
• pp.201-207
• /
• 2019

An anchorage system is essential for most reinforced concrete structures to connect building components. Therefore, the prediction of strength of the anchor is very important issue for safety of the structures themselves as well as structural components. The prediction models in existing design codes are, however, not applicable for large anchors because they are based on the small size anchors with diameters under 50 mm. In this paper, new prediction models for strength of a single anchor, especially the tensile strength of a single anchor, is developed from the experimental results with consideration of size effect. Size effect in the existing models such as ACI or CCD method is based on the linear fracture mechanics which is very conservative way to consider the size effect. Therefore, new models are developed based on the nonlinear fracture mechanics rather than the linear fracture mechanics for more reasonable prediction. New models are proposed by the regression analysis of the experimental results and it can predict the tensile strength of both small and large anchors.

• 대한전자공학회논문지SP
• /
• 제45권4호
• /
• pp.84-91
• /
• 2008

본 논문에서는 초음파 센서를 이용하여 고령자를 위한 앰비언트 디스플레이 시스템을 제안하고, 시스템의 신뢰도를 높이기 위해서 선형 예측 알고리즘을 적용하였다. 본 논문에서는 시스템의 사용자를 고령자로 제안하여 일반인에 비해 느린 움직임으로 가정하였고 얻어진 데이터가 모두 극점인 데이터의 특성상 AR(Autoregressive) 모델을 사용하여 Yule-Walker 방식의 선형 예측 알고리즘을 적용하였다. 선형 예측 알고리즘을 적용하기 위해서는 적절한 참조 데이터와 차수의 결정이 요구된다. 본 논문에서는 데이터의 특성과 평균 에러, 계산량을 고려하여 50개의 참조데이터를 이용한 16차의 시스템을 통해서 앰비언트 디스플레이 시스템의 신뢰도를 평균 74.39%, 최대 97.97%정도 높일 수 있음을 확인하였다.

• 품질경영학회지
• /
• 제26권4호
• /
• pp.143-150
• /
• 1998

We propose an adaptive monitoring a, pp.oach for serially correlated data. This algorithm uses the adaptive linear prediction lattice filter (ALPLF) which makes it compute process parameters in real time and recursively update their estimates. It involves computation of the forward and backward prediction errors. CUSUM control charts are a, pp.ied to prediction errors simulaneously in both directions as an omnibus method for detecting changes in process parameters. Results of computer simulations demonstrate that the proposed adaptive monitoring a, pp.oach has great potentials for real-time industrial a, pp.ications, which vary frequently in their control environment.

• 한국항공운항학회지
• /
• 제24권4호
• /
• pp.44-52
• /
• 2016

The departure flow management is the planning tool to optimize the schedule of the departure aircraft and allows them to join smoothly into the overhead traffic flow. To that end, the arrival time prediction to the merge point for the cruising aircraft is necessary to determined. This paper proposes a trajectory prediction model for the cruising aircraft based on the machine learning approach. The proposed method includes the trajectory vectored from the procedural route and is applied to the historical data to evaluate the prediction performances.

• 한국멀티미디어학회논문지
• /
• 제6권1호
• /
• pp.114-120
• /
• 2003

블라인드 채널 추정은 매우 중요한 문제이다 기존의 방법들은 대부분 고차통계를 이용한 방법들이었으나 최근에 안테나 어레이를 통과한 수신신호나 수신신호를 오버샘플링한 신호의 2차통계를 이용한 방법들에 관한 많은 연구가 진행되고 있다. 본 논문에서는 다채널 선형예측 방법을 이용한 블라인드 적응 채널 추정 방법을 제안하고 컴퓨터 모의실험을 통하여 제안한 방법과 기존의 방법의 성능을 비교 분석한다. 본 논문에서 제안한 방법은 기존의 방법들보다 우수한 성능을 보였으며 채널의 정확한 차수를 모르는 경우에도 우수한 성능을 보임을 확인하였다.

• 대한의용생체공학회:학술대회논문집
• /
• 대한의용생체공학회 1997년도 추계학술대회
• /
• pp.228-231
• /
• 1997

In this paper, we designed an ART(Adaptive Resonance Theory) network based on LPC(Linear Prediction Coefficient) for classification of PVB (Premature Ventricular Beat: PVC, LBBB, RBBB). The procedure of proposed system consists of the error calculation, feature generation and processing of the ART network. The error is calculated after processing by linear prediction algorithm and the features of ART network or classification are obtained from the binary ata determined by threshold method. In conclusion, ART network has good performance in classification of PVB.