• International Journal of Control, Automation, and Systems
• /
• v.6 no.4
• /
• pp.526-533
• /
• 2008

We propose a wavelet neural network (WNN) control method for active queue management (AQM) in an end-to-end TCP network, which is trained by adaptive learning rates (ALRs). In the TCP network, AQM is important to regulate the queue length by passing or dropping the packets at the intermediate routers. RED, PI, and PID algorithms have been used for AQM. But these algorithms show weaknesses in the detection and control of congestion under dynamically changing network situations. In our method, the WNN controller using ALRs is designed to overcome these problems. It adaptively controls the dropping probability of the packets and is trained by gradient-descent algorithm. We apply Lyapunov theorem to verify the stability of the WNN controller using ALRs. Simulations are carried out to demonstrate the effectiveness of the proposed method.

• Nuclear Engineering and Technology
• /
• v.30 no.3
• /
• pp.222-234
• /
• 1998

The Westinghouse OT$\Delta$T DNB protection logic heavily restricts the operation region by applying the same logic for a full range of operating pressure in order to maintain its simplicity. In this work, a fuzzy neural network method is used to estimate the DNB protection limit using the measured average temperature and pressure of a reactor core. Fuzzy system parameters are optimized by a hybrid learning method. This algorithm uses a gradient descent algorithm to optimize the antecedent parameters and a least-squares algorithm to solve the consequent parameters. The proposed method is applied to Yonggwang 3&4 nuclear power plants and the proposed method has 5.99 percent larger thermal margin than the conventional OT$\Delta$T trip logic. This simple algorithm provides a good information for the nuclear power plant operation and diagnosis by estimating the DNB protection limit each time step.

• Korean Journal of Air-Conditioning and Refrigeration Engineering
• /
• v.16 no.2
• /
• pp.135-141
• /
• 2004

The present paper concerns a smoke movement in a tunnel fire with a vertical shaft. The model tunnel measured 13.4m long, 0.4m wide and 0.4m high. The cross section is 1： 20 of a full scale tunnel. Ethanol was used as a fuel. The fire size in model tests varied from 1.35 kW to 13.37 kW, which corresponds to full scale fires of 2.41 to 23.91 MW. Smoke front velocity and temperatrue were decreased due to the vertical shaft install. Temperature was reduced maximum about 2$0^{\circ}C$ at ceiling and about 23$^{\circ}C$ at vertical position. CO concentration was reduced as the vent width widened. When vent width was more than 15 cm, CO concentration was not reached 100 ppm. Descent degree of the smoke layer was confirmed through the visualization.

• Proceedings of the Acoustical Society of Korea Conference
• /
• 1994.06a
• /
• pp.997-1002
• /
• 1994

Predictive neural network models are powerful speech recognition models based on a nonlinear pattern prediction. Those models can effectively normalize the temporal and spatial variability of speech signals. But those models suffer from poor discrimination between acoustically similar words. In this paper, we propose a discriminative training algorithm for predictive neural network models based on a generalized probabilistic descent (GPD) algorithm and minimum classification error formulation (MCEF). The Evaluation of our training algorithm on ten Korean digits shows its effectiveness by 40% reduction of recognition error.

• Nuclear Engineering and Technology
• /
• v.49 no.8
• /
• pp.1696-1710
• /
• 2017

An improved Reactor Regulating System (RRS) logic architecture, which is combined with genetic algorithm (GA), is implemented in this work. It is devised to provide an optimal solution to the current RRS. The current system works desirably and has contributed to safe and stable nuclear power plant operation. However, during the ascent and descent section of the reactor power, the RRS output reveals a relatively high steady-state error, and the output also carries a considerable level of overshoot. In an attempt to consolidate conservatism and minimize the error, this work proposes to apply GA to RRS and suggests reconfiguring the system. Prior to the use of GA, reverse engineering is implemented to build a Simulink-based RRS model. Reengineering is followed to produce a newly configured RRS to generate an output that has a reduced steady-state error and diminished overshoot level. A full-scope APR1400 simulator is used to examine the dynamic behaviors of RRS and to build the RRS Simulink model.

• Proceedings of the Korean Nuclear Society Conference
• /
• 1997.10a
• /
• pp.299-304
• /
• 1997

A neuro-fuzzy control algorithm is applied for xenon spatial oscillations in a pressurized water reactor. The consequent and antecedent parameters of the fuzzy rules are tuned by the gradient descent mettled. The reactor model used for computer simulations is a two-point xenon oscillation model. The reactor core is axially divided into two regions and each region has one input and one output and is coupled with the other region. The interaction between the regions of the reactor core is treated by a decoupling scheme. This proposed control of mettled exhibits very fast responses to a step or a ramp change of target axial offset without any residual flux oscillations.

• Journal of Power Electronics
• /
• v.16 no.6
• /
• pp.2150-2161
• /
• 2016

Mechanical and electrical parameter uncertainties cause dynamic and static estimation errors of the rotor speed and position, resulting in performance deterioration of sensorless control systems. This paper applies an extended nonlinear observer to interior permanent magnet synchronous motors (IPMSM) for the simultaneous estimation of the rotor speed and position. Two compensation methods are proposed to improve the observer performance against parameter uncertainties: an on-line rotational inertia adjustment approach that employs the gradient descent algorithm to suppress dynamic estimation errors, and an equivalent flux error compensation approach to eliminate static estimation errors caused by inaccurate electrical parameters. The effectiveness of the proposed control strategy is demonstrated by experimental tests.

• International Journal of Control, Automation, and Systems
• /
• v.4 no.5
• /
• pp.660-667
• /
• 2006

This paper presents a novel region-based snake method for vessel segmentation. According to geometric shape analysis of the vessel structure with different scale, an efficient statistical estimation of vessel branches is introduced into the energy objective function, which applies not only the vessel intensity information, but also geometric information of line-like structure in the image. The defined energy function is minimized using the gradient descent method and a new region-based speed function is obtained, which is more accurate to the vessel structure and not sensitive to the initial condition. The narrow band algorithm in the level set framework implements the proposed method, the solution of which is steady. The segmentation experiments are shown on several images. Compared with other geometric active contour models, the proposed method is more efficient and robust.

• Proceedings of the IEEK Conference
• /
• 2002.07c
• /
• pp.1626-1629
• /
• 2002

This paper describes novel methods to construct fuzzy inference rules with gradient descent. The present methods have a constructive mechanism of the rule unit that is applicable in two parameters: the central value and the width of the membership function in the antecedent part. The first approach is to create the rule unit at the nearest position from the input space, for the central value of the membership function in the antecedent part. The second is to create the rule unit which has the minimum width, for the width of the membership function in the antecedent part. Experimental results are presented in order to show that the proposed methods are effective in difference on the inference error and the number of learning iterations.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.31B no.8
• /
• pp.211-218
• /
• 1994

In this paper we develope a hybrid learning rule to improve the robustness of multi-layer Perceptions. In most neural networks the activation of a neuron is deternined by a nonlinear transformation of the weighted sum of inputs to the neurons. Investigating the behaviour of activations of hidden layer neurons a new learning algorithm is developed for improved robustness for multi-layer Perceptrons. Unlike other methods which reduce the network complexity by putting restrictions on synaptic weights our method based on error-backpropagation increases the complexity of the underlying proplem by imposing it saturation requirement on hidden layer neurons. We also found that the additional gradient-descent term for the requirement corresponds to the Hebbian rule and our algorithm incorporates the Hebbian learning rule into the error back-propagation rule. Computer simulation demonstrates fast learning convergence as well as improved robustness for classification and hetero-association of patterns.