• Journal of the Korean Society of Manufacturing Process Engineers
• /
• v.14 no.3
• /
• pp.65-73
• /
• 2015

This paper presents a method of predicting the machining parameters on the turning process of low carbon steel using a neural network with back propagation (BP) and particle swarm optimization (PSO). Cutting speed, feed rate, and depth of cut are used as input variables, while surface roughness and electric current consumption are used as output variables. The data from experiments are used to train the neural network that uses BP and PSO to update the weights in the neural network. After training, the neural network model is run using test data, and the results using BP and PSO are compared with each other.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2003.06a
• /
• pp.1533-1536
• /
• 2003

Dynamic model of the rubber-tired AGT light rail vehicle. various load and boundary conditions between vehicle and infrastructures(running track. guidance rail) were defined to analyze vehicular dynamic behaviors occurred by inclement weather(strong wind and earthquake). The dynamic analysis were performed by using a commercialized software RecurDyn. whose results for the magnitudes of wind and earthquake showed the resultant forces and accelerations between car body and bogie, or bogie guidance frame and infrastructures in the various track conditions(straight or curved tracks) and train velocities. Based on these results, speed limitation of tile vehicle were proposed to ensure system stability and passengers safety.

• MALSORI
• /
• no.67
• /
• pp.149-166
• /
• 2008

In spoken document retrievals (SDR), subword (typically phonemes) indexing term is used to avoid the out-of-vocabulary (OOV) problem. It makes the indexing and retrieval process independent from any vocabulary. It also requires a small corpus to train the acoustic model. However, subword indexing term approach has a major drawback. It shows higher word error rates than the large vocabulary continuous speech recognition (LVCSR) system. In this paper, we propose an probabilistic slot detection and n-gram based string matching method for phone based spoken document retrievals to overcome high error rates of phone recognizer. Experimental results have shown 9.25% relative improvement in the mean average precision (mAP) with 1.7 times speed up in comparison with the baseline system.

• Proceedings of the KSPS conference
• /
• 2007.05a
• /
• pp.311-314
• /
• 2007

Spoken dialog system development includes many laborious and inefficient tasks. Since there are many components such as speech recognizer, language understanding, dialog management and knowledge management in a spoken dialog system, a developer should take an effort to edit corpus and train each model separately. To reduce a cost for editting corpus and training each models, we need more systematic and efficent working environment. For the working environment, we propose DialogStudio as an spoken dialog system workbench.

• Transactions of the Korean Society of Machine Tool Engineers
• /
• v.10 no.2
• /
• pp.18-28
• /
• 2001

Super computers has been utilized to carry out vehicle dynamics in real time. This research propose an implicit integra-tion method for vehicle state variables. Newton chord method is empolyed to solve the equations of motion and con-straints. The equations of motion and constraints are formulated such that the Jacobian matrix for Newton chord method is needed to be computed only once for a dynamic analysis. Numerical experiments showed that the Jacobian matrix generat-ed at the initial time could have been utilized for the Newton chord iterations throughout simulations under various driving conditions. Convergence analysis of Newton chord method with the proposed Jacobian updating method is carried out. The proposed algorithm yielded accurate solutions for a prototype vehicle multibody model in realtime on a 400 MHz PC compatible.

• The Journal of the Acoustical Society of Korea
• /
• v.23 no.3E
• /
• pp.91-97
• /
• 2004

In this paper, we propose a classification system of TV program scenes based on audio information. The system classifies the video scene into six categories of commercials, basketball games, football games, news reports, weather forecasts and music videos. Two type of audio feature set are extracted from each audio frame-timbral features and coefficient domain features which result in 58-dimensional feature vector. In order to reduce the computational complexity of the system, 58-dimensional feature set is further optimized to yield l0-dimensional features through Sequential Forward Selection (SFS) method. This down-sized feature set is finally used to train and classify the given TV program scenes using κ -NN, Gaussian pattern matching algorithm. The classification result of 91.6% reported here shows the promising performance of the video scene classification based on the audio information. Finally, the system stability problem corresponding to different query length is investigated.

• 제어로봇시스템학회:학술대회논문집
• /
• 1992.10b
• /
• pp.558-563
• /
• 1992

In this paper an identification of nonlinear continuous systems by using neural network is considered. The nonlinear continuous system is identified by two steps. At first, a linear approximate model of the continuous system with nonlinearity is obtained by IIR filtering approach. Then the modeling error due to the nonlinearity is reduced by a neural network compensator. The teaching signals to train the neural network is gotten by smoothing the measurement data corrupted by noise. An illustrative example is given to demonstrate the effectiveness of the proposed approach.

• Journal of the Korean Society for Railway
• /
• v.8 no.2
• /
• pp.161-169
• /
• 2005

In this paper, we provide safety requirements and advices to guarantee the safety of an interface in a level crossing system which is an interface between the conventional facilities and the new ATP (Automatic Train Protection) system, as well as we accomplish a safety management for the facilities of a country that has a different standard with already standardized ATP system. The system model has been made based on a safety activity of the international standard, and then a tolerance of a risk by the safety activity through PHA (Preliminary Hazard Analysis) has been analyzed. finally we achieved HIA (Hazard Identification and Analysis) for the assumptions that have been produced from a operating scenario and a functional interface. Thus, the safety requirements for the interface has been provided from the safety plan of HIA, and we showed the safety activity to guarantee the system safety through HIA which was depend on the design.

• Journal of the Korean Society for Railway
• /
• v.9 no.3 s.34
• /
• pp.298-304
• /
• 2006

An active application of concrete track is being expected for the future constructions of Korean railroad. For the successful concrete track construction and design in earthwork areas, the roadbed behavior should be reasonably estimated using the proper analysis method. In this paper, behaviors of concrete track on the reinforced roadbed constructed with the standard stiffness and depth were estimated thorough numerical analyses and field measurements. A three dimensional finite difference method was employed to model the concrete tracks and subground. The settlement and vertical pressures caused by train load were estimated by the numerical method and compared with the field measurement results. The bearing characteristics of roadbed were presented and the proper method for the analysis of concrete track was proposed.

• Journal of the Korean Society for Railway
• /
• v.9 no.3 s.34
• /
• pp.305-312
• /
• 2006

Gravel ballasted tracks are used as a basic structure for the domestic railway tracks. However, such kind of tracks has few disadvantages with service life of the structure, such as rapid deterioration of the tracks. Due to this reason, there is a need to develop low maintenance track to improve the service life of the conventional line tracks. CMP paved tacks are one of the kind of concrete tracks those were manufactured by using the prepacked concrete techniques. The purpose to develop paved tracks is to reduce the maintenance cost. The most important controlling factors to design the paved tracks are surrounding environmental condition and repeated train loading. In this study, in order to investigate the deformation characteristics such as displacement, earth pressure, strain ratio, and crack along the repeated loading cycle, cyclic loading test through real scale model was carried out.