• Proceedings of the KIEE Conference
• /
• 1987.11a
• /
• pp.45-48
• /
• 1987

In applications of step-motor, the requirements of small-size and strong-power gave birth to feed-back drive to the maximum efficiency and complete damping. In this paper, a new scheme that determines switching times from sensed values of position and speed is presented for simple bang-bang type drive circuit. The optimal values of each phase switching times are obtained using conjugate gradient method.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1996.10a
• /
• pp.198-207
• /
• 1996

Electro sensitive traffic system can't consider passenger car unit, so, it causes start up delay time and passenger waiting time. In this paper, it antecedently creates optimal traffic cycle of passenger car unit at the bottom traffic intersection. But, sometimes it can make mistakes due to changes in car weight, car speed, and control of feed-back data. Moreover, to prevent spillback, it can adapt control even though upper traffic intersection has a different saturation rate, road length, road slope and road width.

• 제어로봇시스템학회:학술대회논문집
• /
• 2001.10a
• /
• pp.74.3-74
• /
• 2001

This paper proposes a nonlinear adaptive controller based on back-stepping method for tracking reference substrate concentration by manipulating dilution rate in a continuous baker´s yeast cultivating process in stirred tank bioreactor. Control law is obtained from Lyapunov control function to ensure asymptotical stability of the system. The Haldane model for the specific growth rate depending on only substrate concentration is used in this paper. Due to the uncertainty of specific growth rate, it has been modified as a function including the unknown parameter with known bounded values. The substrate concentration in the bioreactor and feed line are measured. The deviation from the reference is observed when the external disturbance such as the change of the feed is introduced to the system ...

• International Journal of Korean Welding Society
• /
• v.3 no.2
• /
• pp.15-23
• /
• 2003

A genetic algorithm was applied to an arc welding process to determine near optimal settings of welding process parameters which produce good weld quality. This method searches for optimal settings of welding parameters through systematic experiments without a model between input and output variables. It has an advantage of being able to find optimal conditions with a fewer number of experiments than conventional full factorial design. A genetic algorithm was applied to optimization of weld bead geometry. In the optimization problem, the input variables were wire feed rate, welding voltage, and welding speed, root opening and the output variables were bead height, bead width, penetration and back bead width. The number of level for each input variable is 8, 16, 8 and 3, respectively. Therefore, according to the conventional full factorial design, in order to find the optimal welding conditions, 3,072 experiments must be performed. The genetic algorithm, however, found the near optimal welding conditions from less than 48 experiments.

• Proceedings of the IEEK Conference
• /
• 2006.06a
• /
• pp.837-838
• /
• 2006

This paper proposes a method of assembling parts using visual servoing in dynamic environment. We use SSD(Sum of Square Difference) based on adaptive template in order to detect a moving object in the case where the robot and the object both move. And the control input of the robot is obtained from the feed-back signal of the feature movement and the feed-forward signal of the camera movement in image plane.

• Proceedings of the Korean Society of Marine Engineers Conference
• /
• 2001.05a
• /
• pp.187-191
• /
• 2001

In the paper, an approach to the development of the selective catalytic reduction process of NOx is presented. The reduction process can be efficiently controlled using a conventional combination of feed-forward and feed-back control structures. The aim of this paper is to test and verify an approach to the SCR process which is based on an industrial pilot plant of combustion and nitric oxide formation. The systems are based on measurements of a NOx removal ratio and the fuel flow rate, and NH$_3$slip which are usually available as a part of de-NOx control system.

• Journal of the Korean Crystal Growth and Crystal Technology
• /
• v.16 no.3
• /
• pp.116-120
• /
• 2006

We have investigated plasma spray coated $Al_2O_3$ layers on Al-60 series substrates for development of wafer electrostatic chuck in semiconductor dry etching system. Samples were prepared without/with cooling bar on backside of samples, at various distances, and with different powder feed rates. There were many cracks and pores in the $Al_2O_3$ layers coated on Al-60 series substrates without cooling bar on the backside of samples. But the cracks and pores were almost disappeared in the $Al_2O_3$ layers on Al-60 series substrates coated with cooling bar on the back side of samples, 15 g/min. powder feed rate and various 60, 70, 80 mm working distances. Then the surface morphology was not changed with various working distances of 60, 70, 80 mm. When the powder feed rate was changed from 15 g/min to 20 g/min, the crack did not appear, but few pores appeared. Also the $Al_2O_3$ layer was coated with many small splats compared with $Al_2O_3$ layer coated with 15 g/min powder feed rate. The deposited rate of $Al_2O_3$ layer was higher when the process was done without cooling bar on the back side of sample than that with cooling bar on the back side of sample.

• Journal of Korean Tunnelling and Underground Space Association
• /
• v.2 no.3
• /
• pp.71-81
• /
• 2000

For the construction of NATM tunnel, it is required a design based on the accurate soil condition from soil investigation. However, in practice, it often designs tunnels without fully understanding the condition. Especially, when soft soil comes up, or ground water breaks out suddenly on the construction, it needs to secure the stability of tunnel by appropriate reinforcing construction according to the results of measurements on field superlatively reflecting the faced situation. This report reviews the mostsuitable stability of tunnel in the construction of soft soil of tunnel by numerical analysis using FDM after re-evaluated the soil properties through back analysis using the results of measurements to simulate abruptly occurred deformation. And applying steel pipe grouting row by row on the wall and the low part of tunnel and also applying the construction method of temporary invert after excavation of the upper part of tunnel, the excavation of soft soil tunnel secured the structural stability of tunnel has been completed.

• 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.

• Journal of Institute of Control, Robotics and Systems
• /
• v.21 no.6
• /
• pp.518-523
• /
• 2015

This paper proposes a gait phase classifier using a Recurrent Neural Network (RNN). Walking is a type of dynamic system, and as such it seems that the classifier made by using a general feed forward neural network structure is not appropriate. It is known that an RNN is suitable to model a dynamic system. Because the proposed RNN is simple, we use a back propagation algorithm to train the weights of the network. The input data of the RNN is the lower body's joint angles and angular velocities which are acquired by using the lower limb exoskeleton robot, ROBIN-H1. The classifier categorizes a gait cycle as two phases, swing and stance. In the experiment for performance verification, we compared the proposed method and general feed forward neural network based method and showed that the proposed method is superior.