• 한국안전학회지
• /
• 제31권3호
• /
• pp.82-88
• /
• 2016

Ballasted gravel will be damaged or worn by the repetitive train load. And these damages of ballast gravel could be increased by increasing vehicle speed. Therefore, various techniques for reducing the ballast pressure have been proposed, such as the attached pad type of sleeper bottom for ballasted track. In this study, spring stiffness test method were proposed to evaluate the performance of under sleeper pad for ballasted track. Standard ballast plate(SBP) was developed to simulate the ballast gravel and compared with the foreign test results. Experimental results showed a trend similar to the previous studies according to various loading plate type. specimen type(Type A, Type B) differences in spring stiffness according to hardness were not significant. Also, the FSP (Flat steel plate) - shaped jig is about 80% of the spring stiffness was greater than SBP. Therefore, to evaluate the actual spring stiffness of under sleeper pad for ballasted track, it was important to adopted the appropriate spring stiffness test method using the SBP to simulate actual field conditions.

• Journal of information and communication convergence engineering
• /
• 제13권2호
• /
• pp.123-131
• /
• 2015

Training neural networks is a complex task with great importance in the field of supervised learning. In the training process, a set of input-output patterns is repeated to an artificial neural network (ANN). From those patterns weights of all the interconnections between neurons are adjusted until the specified input yields the desired output. In this paper, a new hybrid algorithm is proposed for global optimization of connection weights in an ANN. Dynamic swarms are shown to converge rapidly during the initial stages of a global search, but around the global optimum, the search process becomes very slow. In contrast, the gradient descent method can achieve faster convergence speed around the global optimum, and at the same time, the convergence accuracy can be relatively high. Therefore, the proposed hybrid algorithm combines the dynamic particle swarm optimization (DPSO) algorithm with the backpropagation (BP) algorithm, also referred to as the DPSO-BP algorithm, to train the weights of an ANN. In this paper, we intend to show the superiority (time performance and quality of solution) of the proposed hybrid algorithm (DPSO-BP) over other more standard algorithms in neural network training. The algorithms are compared using two different datasets, and the results are simulated.

• 한국항해항만학회:학술대회논문집
• /
• 한국항해항만학회 2006년도 International Symposium on GPS/GNSS Vol.2
• /
• pp.101-106
• /
• 2006

The geoidal undulations are needed for determining the orthometric heights from the Global Positioning System GPS-derived ellipsoidal heights. There are several methods for geoidal undulation determination. The paper presents a method employing a simple architecture Two Layer Multiquadric-Biharmonic Artificial Neural Network (TLMB-ANN) to approximate an area of 4200 square kilometres quasigeoid surface with GPS-levelling data. Hardy’s Multiquadric-Biharmonic functions is used as the hidden layer neurons’ activation function and Levenberg-Marquardt algorithm is used to train the artificial neural network. In numerical examples five surfaces were compared: the gravimetric geometry hybrid quasigeoid, Support Vector Machine (SVM) model, Hybrid Fuzzy Neural Network (HFNN) model, Traditional Three Layer Artificial Neural Network (ANN) with tanh activation function and TLMB-ANN surface approximation. The effectiveness of TLMB-ANN surface approximation depends on the number of control points. If the number of well-distributed control points is sufficiently large, the results are similar with those obtained by gravity and geometry hybrid method. Importantly, TLMB-ANN surface approximation model possesses good extrapolation performance with high precision.

• 방송공학회논문지
• /
• 제23권2호
• /
• pp.186-196
• /
• 2018

본 논문에서는 객체를 추적하기 위해 합성곱 신경망 모델을 이용한 온라인 추적 기법을 제안한다. 오프라인에 모델을 학습시키기 위해서는 많은 수의 훈련 샘플이 필요하다. 이러한 문제를 해결하기 위해, 학습되지 않은 모델을 사용하고, 실험 영상으로부터 직접 훈련 샘플을 수집하여 모델을 갱신한다. 기존의 방법들은 많은 훈련 샘플을 획득하여 모델의 학습에 사용하였지만, 본 논문에서는 적은 수의 훈련 샘플만으로도 객체의 추적이 가능함을 증명한다. 또한 컬러 정보를 활용하여 새로운 손실 함수를 정의하였고 이로부터 잘못 수집된 훈련 샘플로 인해 모델이 잘못된 방향으로 학습되는 문제를 방지한다. 실험을 통해 4가지 비교 방법과 동등하거나 개선된 추적 성능을 보임을 증명하였다.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2007년도 Techno-Fair 및 추계학술대회 논문집 전기물성,응용부문
• /
• pp.208-209
• /
• 2007

The measurement system is composed of the industrial computers installed in the console and the measurement racks mounted on each car. It is utilized to accumulate the data by the communication card and the optical cable. The optical cable and power cable are coupled at the connector located in joint of train to make easy to disconnect car each other. The signal conditioner is designed to choose and to extend the channel for each sensor readily, In this study, the programs for measurement and analysis were also developed to understand the traction system characteristics of TTX. Using this measurement system, we studied that acceleration test, re-powering test and gradually powering test. From the test results, we saw the performances of the traction systems are normal.

• 전력전자학회논문지
• /
• 제4권6호
• /
• pp.491-499
• /
• 1999

본 논문은 새로운 제어기법의 고성능 전동차 추진용 IGBT VVVF 인버터에 관한 것이다. 현차 적용 전에 견인력 제어성의 우수함과 안정성을 검증하기 위해, 현차 조건에서 발생될 수 있는 경우에 대한 모의 주행시험을 실시하고 그 결과를 보인다. 모의 주행시험은 160톤의 차량을 등가화한 관성부하장치를 이용하여 정상상태는 물론 전압변동 등 각종 과도상태에 대해서도 실시하였다. 본 논문에서는 4병렬 접속된 견인전동기의 토오크 제어성 향상을 위한 벡터제어기법 적용과 공간벡터 변조기법에 동기방식을 적용하여 500Hz이하의 저주파 스위칭으로써 6스텝까지 연속 과변조 제어가 가능한 최적의 PWM방법을 제시한다.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2007년도 제38회 하계학술대회
• /
• pp.1183-1184
• /
• 2007

Light Rail Transit (LRT) is optimized vehicle system for complex urban circumstance. LRT systems have many merits such as improved accuracy and safety. There are many LRT systems such as monorail, tram, automated guideway transit, linear induction motor propulsion and so on. These systems have been operated in Japan and other advanced countries. In Korea, local government has many projects to apply the advanced LRT system. But there are no standards regulation, performance test regulation and construction regulation for monorail system, linear induction motor system and tram in Korea. We expect that standardization brings economical construction and safety. The linear induction motor system has been usually applied to subway in Japan and ART(Advanced Rapid Transit) in Canada. In Korea, the linear induction motor system has been adopted for Yong-In LRT and currently under construction. This paper presents signalling system and TCMS(train control and monitoring system) of linear induction motor system.

• 한국소음진동공학회논문집
• /
• 제12권7호
• /
• pp.542-549
• /
• 2002

Computer simulation is essential to design the suspension elements of railway vehicle. By computer simulation, engineers can assess the feasibility of given design variables and chance them to get a bettor design. Even though commercial simulation codes are used, the computational time and cost remains non-trivial. Therefore, malty researchers have used a mesa model made by sampling data through simulation. In this paper, four mesa-models for each index group such as ride comfort, derailment Quotient, unloading radio and stability index, are constructed by use of neural network. After these meta models are constructed, multi-objective optimization are achieved by using the differential evolution. This paper shows that the optimization of design variables using the neural network model is very efficient to solve the complex optimization Problem.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제9권10호
• /
• pp.4126-4142
• /
• 2015

Finger vein recognition is a biometric technology using finger veins to authenticate a person, and due to its high degree of uniqueness, liveness, and safety, it is widely used. The traditional Symmetric Local Graph Structure (SLGS) method only considers the relationship between the image pixels as a dominating set, and uses the relevant theories to tap image features. In order to better extract finger vein features, taking into account location information and direction information between the pixels of the image, this paper presents a novel finger vein feature extraction method, Multi-Orientation Weighted Symmetric Local Graph Structure (MOW-SLGS), which assigns weight to each edge according to the positional relationship between the edge and the target pixel. In addition, we use the Extreme Learning Machine (ELM) classifier to train and classify the vein feature extracted by the MOW-SLGS method. Experiments show that the proposed method has better performance than traditional methods.

• 열처리공학회지
• /
• 제22권4호
• /
• pp.203-209
• /
• 2009

Dual mass flywheel is the newly developed flywheel system which reduces the noise and vibration and make a better and comfortable ride of cars by adding inertia mass and damping device. However, verification of performance for this system should be carried out since this system is under developing status in our country. Especially, the durability for each part of this system should be guaranteed. Durable properties of driver plate which is the key component of dual mass flywheel were first investigated both in the raw (SCM435 in JIS) and heat-treated material. In addition, fatigue life analysis of driver plate was preformed in the real condition and the results were verified by comparison with the results of rig test.