• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제11권10호
• /
• pp.4887-4907
• /
• 2017

Accurate traffic flow forecasting is critical to the development and implementation of city intelligent transportation systems. Therefore, it is one of the most important components in the research of urban traffic scheduling. However, traffic flow forecasting involves a rather complex nonlinear data pattern, particularly during workday peak periods, and a lot of research has shown that traffic flow data reveals a seasonal trend. This paper proposes a new traffic flow forecasting model that combines seasonal relevance vector regression with the hybrid chaotic simulated annealing method (SRVRCSA). Additionally, a numerical example of traffic flow data from The Transportation Data Research Laboratory is used to elucidate the forecasting performance of the proposed SRVRCSA model. The forecasting results indicate that the proposed model yields more accurate forecasting results than the seasonal auto regressive integrated moving average (SARIMA), the double seasonal Holt-Winters exponential smoothing (DSHWES), and the relevance vector regression with hybrid Chaotic Simulated Annealing method (RVRCSA) models. The forecasting performance of RVRCSA with different kernel functions is also studied.

• 한국경영정보학회:학술대회논문집
• /
• 한국경영정보학회 2007년도 International Conference
• /
• pp.608-613
• /
• 2007

Prediction of corporate bankruptcies has long been an important topic and has been studied extensively in the finance and management literature because it is an essential basis for the risk management of financial institutions. Recently, support vector machines (SVMs) are becoming popular as a tool for bankruptcy prediction because they use a risk function consisting of the empirical error and a regularized term which is derived from the structural risk minimization principle. In addition, they don't require huge training samples and have little possibility of overfitting. However. in order to Use SVM, a user should determine several factors such as the parameters ofa kernel function, appropriate feature subset, and proper instance subset by heuristics, which hinders accurate prediction results when using SVM In this study, we propose a novel hybrid SVM classifier with simultaneous optimization of feature subsets, instance subsets, and kernel parameters. This study introduces genetic algorithms (GAs) to optimize the feature selection, instance selection, and kernel parameters simultaneously. Our study applies the proposed model to the real-world case for bankruptcy prediction. Experimental results show that the prediction accuracy of conventional SVM may be improved significantly by using our model.

• 한국소음진동공학회논문집
• /
• 제14권2호
• /
• pp.119-127
• /
• 2004

Butterfly valves are popularly used in service in the industrial and water works pipeline systems with large diameter because of its lightweight, simple structure and the rapidity of its manipulation. Sometimes cavitation can occur. resulting in noise, vibration and rapid deterioration of the valve trim, and do not allow further operation. Thus, the monitoring of cavitation is of economic interest and is very importance in industry. This paper proposes a condition monitoring scheme using statistical feature evaluation and support vector machine (SVM) to detect the cavitation conditions of butterfly valve which used as a flow control valve at the pumping stations. The stationary features of vibration signals are extracted from statistical moments. The SVMs are trained, and then classify normal and cavitation conditions of control valves. The SVMs with the reorganized feature vectors can distinguish the class of the untrained and untested data. The classification validity of this method is examined by various signals that are acquired from butterfly valves in the pumping stations and compared the classification success rate with those of self-organizing feature map neural network.

• 한국정밀공학회지
• /
• 제15권1호
• /
• pp.78-86
• /
• 1998

This paper presents a numerical method of the minimum-time trajectory planning for a robot manipulator amid obstacles. Each joint displacement is represented by the linear combination of the finite-term quintic B-splines which are the known functions of the path parameter. The time is represented by the linear function of the same path parameter. Since the geometric path is not fixed and the time is linear to the path parameter, the coefficients of the splines and the time-scale factor span a finite-dimensional vector space, a point in which uniquely represents the manipulator motion. The displacement, the velocity and the acceleration conditions at the starting and the goal positions are transformed into the linear equality constraints on the coefficients of the splines, which reduce the dimension of the vector space. The optimization is performed in the reduced vector space using nonlinear programming. The total moving time is the main performance index which should be minimized. The constraints on the actuator forces and that of the obstacle-avoidance, together with sufficiently large weighting coefficients, are included in the augmented performance index. In the numerical implementation, the minimum-time motion is obtained for a planar 3-1ink manipulator amid several rectangular obstacles without simplifying any dynamic or geometric models.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1997년도 추계학술대회 논문집 학회본부
• /
• pp.600-602
• /
• 1997

In this study, a hybrid neural net consisting of an Adaptive LVQ(ALVQ) algorithm and MLP is proposed to perform speaker identification task. ALVQ is a new learning procedure using adaptively feature vector sequence instead of only one feature vector in training codebooks initialized by LBG algorithm and the optimization criterion of this method is consistent with the speaker classification decision rule. ALVQ aims at providing a compressed, geometrically consistent data representation. It is fit to cover irregular data distributions and computes the distance of the input vector sequence from its nodes. On the other hand, MLP aim at a data representation to fit to discriminate patterns belonging to different classes. It has been shown that MLP nets can approximate Bayesian "optimal" classifiers with high precision, and their output values can be related a-posteriori class probabilities. The different characteristics of these neural models make it possible to devise hybrid neural net systems, consisting of classification modules based on these two different philosophies. The proposed method is compared with LBG algorithm, LVQ algorithm and MLP for performance.

• 한국지능정보시스템학회:학술대회논문집
• /
• 한국지능정보시스템학회 2006년도 춘계학술대회
• /
• pp.167-175
• /
• 2006

Building accurate corporate bankruptcy prediction models has been one of the most important research issues in finance. Recently, support vector machines (SVMs) are popularly applied to bankruptcy prediction because of its many strong points. However, in order to use SVM, a modeler should determine several factors by heuristics, which hinders from obtaining accurate prediction results by using SVM. As a result, some researchers have tried to optimize these factors, especially the feature subset and kernel parameters of SVM But, there have been no studies that have attempted to determine appropriate instance subset of SVM, although it may improve the performance by eliminating distorted cases. Thus in the study, we propose the simultaneous optimization of the instance selection as well as the parameters of a kernel function of SVM by using genetic algorithms (GAs). Experimental results show that our model outperforms not only conventional SVM, but also prior approaches for optimizing SVM.

• 한국컴퓨터정보학회논문지
• /
• 제15권7호
• /
• pp.11-17
• /
• 2010

본 논문에서는 클래스 불균형 학습을 위한 이차 최적화 문제의 해를 구하는 개선된 SMO 학습 알고리즘을 제안한다. 클래스에 서로 다른 정규화 값이 부여되는 지지벡터기계의 최적화 문제의 구현에 SMO 알고리즘이 적합하며, 제안된 알고리즘은 서로 다른 클래스에서 선택된 두 라그랑지 변수의 현재 해를 구하는 학습 단계를 반복한다. 제안된 학습 알고리즘은 UCI 벤치마킹 문제에서 테스트되어 클래스 불균형 분포를 반영하는 g-mean 평가를 이용한 일반화 성능이 SMO 알고리즘과 비교되었다. 실험 결과에서 제안된 알고리즘은 SMO에 비해 적은 클래스 데이터의 예측율을 높이고 학습시간을 단축시킬 수 있다.

• Nuclear Engineering and Technology
• /
• 제50권8호
• /
• pp.1306-1313
• /
• 2018

The randomness and incipient nature of certain faults in reactor systems warrant a robust and dynamic detection mechanism. Existing models and methods for fault diagnosis using different mathematical/statistical inferences lack incipient and novel faults detection capability. To this end, we propose a fault diagnosis method that utilizes the flexibility of data-driven Support Vector Machine (SVM) for component-level fault diagnosis. The technique integrates separately-built, separately-trained, specialized SVM modules capable of component-level fault diagnosis into a coherent intelligent system, with each SVM module monitoring sub-units of the reactor coolant system. To evaluate the model, marginal faults selected from the failure mode and effect analysis (FMEA) are simulated in the steam generator and pressure boundary of the Chinese CNP300 PWR (Qinshan I NPP) reactor coolant system, using a best-estimate thermal-hydraulic code, RELAP5/SCDAP Mod4.0. Multiclass SVM model is trained with component level parameters that represent the steady state and selected faults in the components. For optimization purposes, we considered and compared the performances of different multiclass models in MATLAB, using different coding matrices, as well as different kernel functions on the representative data derived from the simulation of Qinshan I NPP. An optimum predictive model - the Error Correcting Output Code (ECOC) with TenaryComplete coding matrix - was obtained from experiments, and utilized to diagnose the incipient faults. Some of the important diagnostic results and heuristic model evaluation methods are presented in this paper.

• 산업경영시스템학회지
• /
• 제43권4호
• /
• pp.208-216
• /
• 2020

With the spread of smart manufacturing, one of the key topics of the 4th industrial revolution, manufacturing systems are moving beyond automation to smartization using artificial intelligence. In particular, in the existing automatic machining, a number of machining defects and non-processing occur due to tool damage or severe wear, resulting in a decrease in productivity and an increase in quality defect rates. Therefore, it is important to measure and predict tool life. In this paper, ν-ASVR (ν-Asymmetric Support Vector Regression), which considers the asymmetry of ⲉ-tube and the asymmetry of penalties for data out of ⲉ-tube, was proposed and applied to the tool wear prediction problem. In the case of tool wear, if the predicted value of the tool wear amount is smaller than the actual value (under-estimation), product failure may occur due to tool damage or wear. Therefore, it can be said that ν-ASVR is suitable because it is necessary to overestimate. It is shown that even when adjusting the asymmetry of ⲉ-tube and the asymmetry of penalties for data out of ⲉ-tube, the ratio of the number of data belonging to ⲉ-tube can be adjusted with ν. Experiments are performed to compare the accuracy of various kernel functions such as linear, polynomial. RBF (radialbasis function), sigmoid, The best result isthe use of the RBF kernel in all cases

• Smart Structures and Systems
• /
• 제30권3호
• /
• pp.303-315
• /
• 2022

The degradation of wheel tread may result in serious hazards in the railway operation system. Therefore, timely wheel defect diagnosis of in-service trains to avoid tragic events is of particular importance. The focus of this study is to develop a novel wheel defect detection approach based on the relevance vector machine (RVM) which enables online detection of potentially defective wheels with trackside monitoring data acquired under different running-speed conditions. With the dynamic strain responses collected by a trackside monitoring system, the cumulative Fourier amplitudes (CFA) characterizing the effect of individual wheels are extracted to formulate multiple probabilistic regression models (MPRMs) in terms of multi-kernel RVM, which accommodate both variables of vibration frequency and running speed. Compared with the general single-kernel RVM-based model, the proposed multi-kernel MPRM approach bears better local and global representation ability and generalization performance, which are prerequisite for reliable wheel defect detection by means of data acquired under different running-speed conditions. After formulating the MPRMs, we adopt a Bayesian null hypothesis indicator for wheel defect identification and quantification, and the proposed method is demonstrated by utilizing real-world monitoring data acquired by an FBG-based trackside monitoring system deployed on a high-speed trial railway. The results testify the validity of the proposed method for wheel defect detection under different running-speed conditions.