• 대한산업공학회지
• /
• 제39권2호
• /
• pp.109-118
• /
• 2013

We propose a method for estimating coefficients of AR (autoregressive) model which named MLPAR (Maximum Likelihood of Pearson system for Auto-Regressive model). In the present method for estimating coefficients of AR model, there is an assumption that residual or error term of the model follows the normal distribution. In common cases, we can observe that the error of AR model does not follow the normal distribution. So the normal assumption will cause decreasing prediction accuracy of AR model. In the paper, we propose the MLPAR which does not assume the normal distribution of error term. The MLPAR estimates coefficients of auto-regressive model and distribution moments of residual by using pearson distribution system and maximum likelihood estimation. Comparing proposed method to auto-regressive model, results are shown to verify improved performance of the MLPAR in terms of prediction accuracy.

• Structural Engineering and Mechanics
• /
• 제15권3호
• /
• pp.299-314
• /
• 2003

This investigation presents an efficient method for identifying modal characteristics from the measured displacement, velocity and acceleration signals of multiple channels on structural systems. A Vector Backward Auto-Regressive model (VBAR) that describes the relationship between the output information in different time steps is used to establish a backward state equation. Generally, the accuracy of the identified dynamic characteristics can be improved by increasing the order of the Auto-Regressive model (AR) in cases of measurement of data under noisy circumstances. However, a higher-order AR model also induces more numerical modes, only some of which are the system modes. The proposed VBAR model provides a clear characteristic boundary to separate the system modes from the spurious modes. A numerical example of a lumped-mass model with three DOFs was established to verify the applicability and effectiveness of the proposed method. Finally, an offshore platform model was experimentally employed as an application case to confirm the proposed VBAR method can be applied to real-world structures.

• 말소리와 음성과학
• /
• 제2권4호
• /
• pp.145-151
• /
• 2010

In this paper, a robust feature compensation method is proposed for improving the performance of speech recognition. The proposed method is incorporated into the auto-regressive moving average (ARMA) based feature compensation. We employ variable weights for the ARMA filter according to the degree of speech activity, and pass the normalized cepstral sequence through the weighted ARMA filter. Additionally when normalizing the cepstral sequences in training, the cepstral means and variances are estimated from total training utterances. Experimental results show the proposed method significantly improves the speech recognition performance in the noisy and reverberant environments.

• 한국수학교육학회지시리즈D:수학교육연구
• /
• 제13권1호
• /
• pp.31-48
• /
• 2009

Utilizing the path analysis method, the study explores the relationships among the influential factors in mathematics modeling academic achievement. The following conclusions are drawn: 1. Achievement motivation, creative inclination, cognitive style, the mathematical cognitive structure and mathematics modeling self-monitoring ability, those have significant correlation with mathematics modeling academic achievement; 2. Mathematical cognitive structure and mathematics modeling self-monitoring ability have significant and regressive effect on mathematics modeling academic achievement, and two factors can explain 55.8% variations of mathematics modeling academic achievement; 3. Achievement motivation, creative inclination, cognitive style, mathematical cognitive structure have significant and regressive effect on mathematics modeling self-monitoring ability, and four factors can explain 70.1% variations of mathematics modeling self-monitoring ability; 4. Achievement motivation, creative inclination, and cognitive style have significant and regressive effect on mathematical cognitive structure, and three factors can explain 40.9% variations of mathematical cognitive structure.

• 전력전자학회논문지
• /
• 제27권4호
• /
• pp.316-324
• /
• 2022

This paper uses seasonal auto-regressive integrated moving average (S-ARIMA), which is efficient in seasonality between time-series models, to predict the degradation tendency for lithium-ion batteries and study a method for improving the predictive performance. The proposed method analyzes the degradation tendency and extracted factors through an electrical characteristic experiment of lithium-ion batteries, and verifies whether time-series data are suitable for the S-ARIMA model through several statistical analysis techniques. Finally, prediction of battery aging is performed through S-ARIMA, and performance of the model is verified through error comparison of predictions through mean absolute error.

• 한국지능시스템학회논문지
• /
• 제19권5호
• /
• pp.713-719
• /
• 2009

본 논문은 비선형 특성을 내재한 물 수요예측을 위하여 기존의 시계열 자기회귀 알고리즘과 다층신경망 학습방법을 결합한 단기 물 수요 예측 알고리즘을 개발하였다. 제시된 방법을 검증하기 위한 사례연구로 2007년도와 2008년도 전북지역의 광역상수도 A정수장에서 취득된 데이터를 활용하여 알고리즘 구축 및 제안 방법의 정확도를 분석하였다. 실험 결과 다중회귀모델은 MAPE가 5.1%, AR모델은 3.8%, 제안된 방법인 AR+MLP 모델은 3.6%로 나타나 성능이 우수한 것으로 나타났다. 따라서 제안된 방법을 사용할 경우 정수장에서 단기 물 수요예측에 유용하게 활용할 수 있음을 보였다.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1994년도 Proceedings of the Korea Automatic Control Conference, 9th (KACC) ; Taejeon, Korea; 17-20 Oct. 1994
• /
• pp.134-139
• /
• 1994

In this paper the comparison between the neural networks and traditional approaches as system identification method are considered. Two model structures of neural networks are the state space model and the input output model neural networks. The traditional methods are the AutoRegressive eXogeneous Input model and the Nonlinear AutoRegressive eXogeneous Input model. The examples considered do not represent any physical system, no a priori knowledge concerning their structure has been used in the identification process. Testing inputs for comparison are the sinusoidal, ramp and the noise ramp.

• 대한산업공학회지
• /
• 제32권1호
• /
• pp.9-17
• /
• 2006

This paper deals with a discrete simulation optimization method for designing a complex probabilistic discrete event simulation. The developed algorithm uses the configuration algorithm that can change decision variables and the stopping algorithm that can end simulation in order to satisfy the given objective value during single run. It tries to estimate an auto-regressive model for evaluating correctly the objective function obtained by a small amount of output data. We apply the proposed algorithm to M/M/s model, (s, S) inventory model, and known-function problem. The proposed algorithm can't always guarantee the optimal solution but the method gives an approximate feasible solution in a relatively short time period. We, therefore, show the proposed algorithm can be used as an initial feasible solution of existing optimization methods that need multiple simulation run to search an optimal solution.

• 한국정보통신학회논문지
• /
• 제23권2호
• /
• pp.117-126
• /
• 2019

부정맥 분류를 위한 기존 연구들은 분류의 정확성을 높이기 위해 신경망, 퍼지, 시계열 주파수 분석, 비선형 분석법 등이 연구되어 왔다. 이러한 방법들은 분류율를 향상시키기 위해 정확한 특징점과 많은 양의 신호를 처리해야 하기 때문에 데이터의 가공 및 연산이 복잡하며, 다양한 부정맥을 분류하는데 어려움이 있다. 본 연구에서는 AR(Auto Regressive) 모델링 기반의 특징점 추출과 SVM(Support Vector Machine)을 통한 조기수축 부정맥 분류 방법을 제안한다. 이를 위해 잡음을 제거한 ECG 신호에서 R파를 검출하고 QRS와 RR 간격의 특정 파형 구간을 모델링하였다. 이후 최적 세그먼트 길이(n1, n2), 최적 차수( p1, p2)의 4가지 AR 모델링 변수를 추출하고 SVM을 통해 Normal, PVC, PAC를 분류하였다. 연구의 타당성을 입증하기 위해 MIT-BIH 부정맥 데이터베이스를 대상으로 한 R파의 평균 검출 성능은 99.77%, Normal, PVC, PAC 부정맥은 각각 99.23%, 97.28, 96.62의 평균 분류율을 나타내었다.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1996년도 한국자동제어학술회의논문집(국내학술편); 포항공과대학교, 포항; 24-26 Oct. 1996
• /
• pp.1213-1216
• /
• 1996

This paper describes a precise temperature control method for refrigerating and air conditioning systems. The control technique is based on the optimal servo control design method and the control algorithm is implemented on a personal computer. To control the precise temperature, two actuators such as an inverter for the compressor speed control and a stepping motor for regulating the expansion valve are used. The superheat and evaporator temperatures are chosen as the system output. So a multivariable system which has two inputs and two outputs to be controlled. The complicative model is identified by using an ARX(Auto Regressive eXogenous) model and the controller is designed by using the Matlab software.