• 한국시뮬레이션학회논문지
• /
• 제5권1호
• /
• pp.1-12
• /
• 1996

We propose a new approach to classifying a time series data into one of the autoregressive moving-average (ARMA) models. It is bases on two pattern recognition concepts for solving time series identification. The one is an extended sample autocorrelation function (ESACF). The other is a neural network-driven decision tree classifier(NNDTC) in which two pattern recognition techniques are tightly coupled : neural network and decision tree classfier. NNDTc consists of a set of nodes at which neural network-driven decision making is made whether the connecting subtrees should be pruned or not. Therefore, time series identification problem can be stated as solving a set of local decisions at nodes. The decision values of the nodes are provided by neural network functions attached to the corresponding nodes. Experimental results with a set of test data and real time series data show that the proposed approach can efficiently identify the time seires patterns with high precision compared to the previous approaches.

• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 1996년도 추계학술대회 학술발표 논문집
• /
• pp.294-297
• /
• 1996

This paper proposes the method to recognize of time series data based on the chaotic feature extraction. Features extract from time series data using the chaotic time series data analysis and the pattern recognition process is using a neural network classifier. In experiment, EEG(electroencephalograph) signals are extracted features by correlation dimension and Lyapunov experiments, and these features are classified by multilayer perceptron neural networks. Proposed chaotic feature extraction enhances recognition results from chaotic time series data.

• 대한원격탐사학회:학술대회논문집
• /
• 대한원격탐사학회 2003년도 Proceedings of ACRS 2003 ISRS
• /
• pp.952-954
• /
• 2003

In this paper, we present the different characteristics of NDVI fluctuation pattern between irrigated and non-irrigated area in Suphanburi province, in Central Thailand. For non-irrigated rice cultivation area, there is a strong correlation between NDVI fluctuation and peak rainfall, while there is a lower correlation with irrigated area. In this study, the 'peak detector' classifier was developed to identify the area of non-irrigated and irrigated cropping and its cropping intensity (number of crops per year). This classifier was created based on cropping characteristics such as number of crops, time or planting period of each crop and its relationship with the peak of rainfall. The classified result showed good accuracy in identification irrigated and nonirrigated rice cultivation areas.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2007년도 제38회 하계학술대회
• /
• pp.477-478
• /
• 2007

This paper proposes a new concept of support vector machine (SVM) based voltage stability classifier using time-series phasor data. The classifier, based on a linear SVM, can provide very effective signals for identification of long-term voltage stability. In addition, the SVM output is applicable as an voltage stability indicator when an amount of corrective controls are performed just to make the system reach around at the maximum deliverable point.

• Smart Structures and Systems
• /
• 제29권1호
• /
• pp.105-116
• /
• 2022

Stay cables play an essential role in cable-stayed bridges. Severe vibrations and/or harsh environment may result in cable failures. Therefore, an efficient structural health monitoring (SHM) solution for cable damage detection is necessary. This study proposes a data-driven method for immediately detecting cable damage from measured cable forces by recognizing pattern transition from the intact condition when damage occurs. In the proposed method, pattern recognition for cable damage detection is realized by time series classification (TSC) using a deep learning (DL) model, namely, the long short term memory fully convolutional network (LSTM-FCN). First, a TSC classifier is trained and validated using the cable forces (or cable force ratios) collected from intact stay cables, setting the segmented data series as input and the cable (or cable pair) ID as class labels. Subsequently, the classifier is tested using the data collected under possible damaged conditions. Finally, the cable or cable pair corresponding to the least classification accuracy is recommended as the most probable damaged cable or cable pair. A case study using measured cable forces from an in-service cable-stayed bridge shows that the cable with damage can be correctly identified using the proposed DL-TSC method. Compared with existing cable damage detection methods in the literature, the DL-TSC method requires minor data preprocessing and feature engineering and thus enables fast and convenient early detection in real applications.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2006년도 추계학술대회 논문집 전력기술부문
• /
• pp.36-39
• /
• 2006

This paper proposes a support vector machine (SVM) based power system voltage stability classifier using local measurement data. The excellent performance of the SVM in the classification related to time-series prediction matches the real-time data of PMU for monitoring power system dynamics. The methodology for fast monitoring of the system is initiated locally which aims to leave sufficient time to perform immediate corrective actions to stop system degradation by the effect of major disturbances. This paper briefly describes the mathematical background of SVM, and explains the procedure for fast classification of voltage stability using the SVM algorithm. To illustrate the effectiveness of the classifier, this paper includes numerical examples with a 11-bus test system.

• 한국지능시스템학회논문지
• /
• 제14권2호
• /
• pp.130-135
• /
• 2004

본 논문에서는 데이터 마이닝을 이용한 단기 전력 부하 예측 시스템의 새로운 설계 기법을 제안한다. 제안된 단기 부하 예측시스템은 Takagj-Sugeno (T-S) 퍼지 모델 기반 예측기와 분류기로 구성된다. 또한, 제안된 T-S 퍼지 모델 기반 분류기는 전반부 가우시안 집합과 후반부 선형화된 베이지안 분류기로 구성된다 분류기의 파라미터들은 주어진 훈련 집합의 통계적 수치로 쉽게 얻어진다. 제안된 T-S 퍼지 모델 기반 예측기는 한 가지 입력에 대한 선형 시계열 예측기의 볼록 조합 형태를 가진다. 후반부 파라미터 추정 문제는 실제 전력 부하와 예측 전력 부하의 놈(norm)을 최소화하는 볼록 최적화 문제로 간주한다. 그 문제는 선형 행렬 부등식으로 설정됨으로써 후반부 파라미터는 추정된다. 전반부 파라미터 추정문제는 선형 시계열 예측기들이 모여진 전체 T-S 퍼지 시스템의 출력과 실제 전력 부하 사이의 에러를 최소화하는 문제이다. 이 문제는 경사치 하향 기법이 적용하여 해결되었다 제안된 기법의 유용성을 검증하기 위해 본 논문은 하루 후 24시간 전력 부하 예측과 하루 후 최고 전력부하를 예측 실험을 제공한다.

• 한국음향학회지
• /
• 제16권2호
• /
• pp.58-65
• /
• 1997

본 논문에서는 시계열 image data를 안정되고 높은 정확도로 분류할 수 있는 자동분류법을 제안하였다. 제안한 방법은 대상 영역에 관한 분류도가 기존재하던 가, 아니면 최소한 시계열 image data 중 어느 한 image data가 분류되어 있다고 하는 전제조건에 그 기초를 두고 있다. 분류도는 training area를 선정하기 위라여 사용하는 기준주제도로 사용되어진다. 제안한 방법은 1)기준주제도를 사용한 training data의 추출, 2)taining data의 균질성에 의거한 변화화소의 검출, 3)검출된 변화화소에 대한 clustering, 4)training data의 재구성, 5)maximum likelihood classifier와 같은 판별법에 의한 분류 등 5개의 단계로 구성된다. 제안한 방법의 성능을 정량적으로 평가하기 위하여 4개의 시계열 Landsat TM image data를 제안한 방법과 숙련된 operator가 필요한 기존의 방법으로 각각 분류하여 비교 검토하였다. 그 결과, 기존의 방법으로는 숙련된 operator가 필요하고, 분류도를 얻기까지 수일이 소요되는 데 반하여, 제안한 방법으로는 숙련된 operator 없이, 신뢰성 있는 분류도를 수 시간 내에 자동으로 얻을 수 있었다.

• Journal of Positioning, Navigation, and Timing
• /
• 제12권1호
• /
• pp.1-9
• /
• 2023

The problem of classifying a non-line-of-sight (NLOS) signal in a multipath channel is important to improve global navigation satellite system (GNSS) positioning accuracy in urban areas. Conventional deep learning-based NLOS signal classifiers use GNSS satellite measurements such as the carrier-to-noise-density ratio (CN_0), pseudorange, and elevation angle as inputs. However, there is a computational inefficiency with use of these measurements and the NLOS signal features expressed by the measurements are limited. In this paper, we propose a Convolutional Neural Network (CNN)-based NLOS signal classifier that receives successive Auto-correlation function (ACF) outputs according to a time-series, which is the most primitive output of GNSS signal processing. We compared the proposed classifier to other DL-based NLOS signal classifiers such as a multi-layer perceptron (MLP) and Gated Recurrent Unit (GRU) to show the superiority of the proposed classifier. The results show the proposed classifier does not require the navigation data extraction stage to classify the NLOS signals, and it has been verified that it has the best detection performance among all compared classifiers, with an accuracy of up to 97%.

• 한국생산제조학회지
• /
• 제7권1호
• /
• pp.69-74
• /
• 1998

In order to monitor the tool wear in milling operation, cutting force is measured as the tool wear increased. The digital signal processing methods are used to detect the tool wear . As AR parameter extract the feature of tool wear , it can be used as input parameter of pattern classifier. The FFT monitor the tool wear exactly , but it can not do real time signal processing. The band energy method can be used to real time monitoring of tool wear ,but int can degrade the exact monitoring.