• 제어로봇시스템학회:학술대회논문집
• /
• 2000.10a
• /
• pp.494-494
• /
• 2000

This paper demonstrates that the largest Lyapunov exponent $\lambda$ of recurrent neural networks can be controlled by a gradient method. The method minimizes a square error $e_{\lambda}=(\lambda-\lambda^{obj})^2$ where $\lambda^{obj}$ is desired exponent. The $\lambda$ can be given as a function of the network parameters P such as connection weights and thresholds of neurons' activation. Then changes of parameters to minimize the error are given by calculating their gradients $\partial\lambda/\partialP$. In a previous paper, we derived a control method of $\lambda$via a direct calculation of $\partial\lambda/\partialP$ with a gradient collection through time. This method however is computationally expensive for large-scale recurrent networks and the control is unstable for recurrent networks with chaotic dynamics. Our new method proposed in this paper is based on a stochastic relation between the complexity $\lambda$ and parameters P of the networks configuration under a restriction. Then the new method allows us to approximate the gradient collection in a fashion without time evolution. This approximation requires only $O(N^2)$ run time while our previous method needs $O(N^{5}T)$ run time for networks with N neurons and T evolution. Simulation results show that the new method can realize a "stable" control for larege-scale networks with chaotic dynamics.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.5
• /
• pp.1912-1931
• /
• 2018

In future wireless network, user experience and energy efficiency will play more and more important roles in the communication systems compared to their roles at present. Quality of experience (QoE) and Energy Efficiency (EE) become the widely used metrics. In this paper, we study a combinatorial problem of QoE and EE and investigate a fair power allocation in heterogeneous networks. We first design a new metric, QoE-aware EE (QEE) to reflect the relationship of QoE and energy. Then, the concept of Utopia QEE is introduced, which is defined as the achievable maximum QEE in ideal conditions, for each user. Finally, we transform the power allocation process to an optimization of ratio of QEE and Utopia QEE and use invasive weed optimization (IWO) algorithm to solve the optimization problem. Numerical simulation results indicate that the proposed algorithm can get converged and efficiently improve the system energy efficiency and the QoE for each user.

• Proceedings of the KIEE Conference
• /
• 1999.07c
• /
• pp.1522-1522
• /
• 1999

This paper proposes the methods of short-term load forecasting using Kohonen neural networks and back-propagation neural networks. Historical load data is divided into 5 patterns for the each seasonal data using Kohonen neural networks and using these results, load forecasting neural network is used for next day hourly load forecasting. Normal days and holidays are forecasted. For load forecasting in summer, max-, and min-temperature data are included in neural networks for a better forecasting accuracy. To show the possibility of the proposed method, it was tested with hourly load data of Korea Electric Power Corporation. (1993-1997)

• Proceedings of the KIEE Conference
• /
• 2003.07a
• /
• pp.427-429
• /
• 2003

The reliability performance of distribution networks will be enhanced with the installation of application programs including feeder reconfiguration program for loss minimization and optimization of system capacity. Although the basic functions of remote monitoring and remote control provide no measurable cost benefit the suite of application programs developed and now being commissioned by KEPCO, the future will continue to offer this utility with substantial savings in system operational costs, improve system reliability and customer service.

• Journal of Institute of Control, Robotics and Systems
• /
• v.7 no.8
• /
• pp.664-674
• /
• 2001

In this paper, as a new category of fuzzy-neural networks architecture, we propose Fuzzy Polynomial Neural Networks (FPNN) and discuss a comprehensive design methodology related to its architecture. FPNN dwells on the ideas of fuzzy rule-based computing and neural networks. The FPNN architecture consists of layers with activation nodes based on fuzzy inference rules. Here each activation node is presented as Fuzzy Polynomial Neuron(FPN). The conclusion part of the rules, especially the regression polynomial, uses several types of high-order polynomials such as linear, quadratic and modified quadratic. As the premise part of the rules, both triangular and Gaussian-like membership functions are studied. It is worth stressing that the number of the layers and the nods in each layer of the FPNN are not predetermined, unlike in the case of the popular multilayer perceptron structure, but these are generated in a dynamic manner. With the aid of two representative time series process data, a detailed design procedure is discussed, and the stability is introduced as a measure of stability of the model for the comparative analysis of various architectures.

• Journal of Electrical Engineering and Technology
• /
• v.11 no.2
• /
• pp.274-284
• /
• 2016

The increasing penetration of distributed energy resources on the distribution networks have brought a number of technical impacts where voltage and thermal variations have been identified as the dominant effects. Active network management in distribution networks aims to integrate distributed energy resources with flexible network management so that distributed energy resources are organized to make better use of existing capacity and infrastructure. This paper propose active solutions which aims to solve the voltage and thermal issues in a distributed manner utilizing a collaborative approach. The proposed algorithms have been fully tested on a distribution network with distributed generation units.

• Journal of Information Processing Systems
• /
• v.13 no.3
• /
• pp.500-517
• /
• 2017

In this paper, an improved zone-based routing protocol for heterogeneous wireless sensor networks is proposed. The proposed protocol has fixed the sized zone according to the distance from the base station and used a dynamic clustering technique for advanced nodes to select a cluster head with maximum residual energy to transmit the data. In addition, we select an optimal route with minimum energy consumption for normal nodes and conserve energy by state transition throughout data transmission. Simulation results indicated that the proposed protocol performed better than the other algorithm by reducing energy consumption and providing a longer network lifetime and better throughput of data packets.

• Journal of Korea Multimedia Society
• /
• v.18 no.8
• /
• pp.909-914
• /
• 2015

In wireless sensor networks, sensor nodes have a short transmission range and data is transferred from source to destination node using the multi-hop transmission. Sensor nodes are powered by battery and the link qualities are different, and the routing protocol in the wireless sensor network is one of the important technical issues. Multipath routing was proposed to reduce the data congestion and increase data throughput. In the multipath routing, however, each path can be interfered by the other path, and it can aggravate network performance. In this paper, we propose the multipath routing scheme for multi-rate wireless sensor networks. The multipath routing selects transmission paths to minimize transmission delay and path interference.

• Journal of Institute of Control, Robotics and Systems
• /
• v.7 no.11
• /
• pp.962-966
• /
• 2001

The major problem at a branch point for point-to-multipoint available bit rate(ABR) services in asynchronous transfer mode (ATM) networks is how to consolidate backward resource management(BRM) cells from each branch for a multicast connection. In this paper, we propose an efficient feedback consolidation algorithm based on an adaptive dynamic threshold(ADT) to eliminate the consolidation noise and the reduce the consolidation delay. The main idea of the ADT algorithm lies in that each branch point estimates the ABR traffic condition of the network through the virtual queue estimation and the transmission threshold of the queue level in branch points is adaptively controlled according to the estimation. Simulation results show that the proposed ADT algorithm can achieve a faster response in congestion status and a higher link utilization compared with the previous works.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.11 no.10
• /
• pp.4680-4697
• /
• 2017

In future, since the user experience plays a more and more important role in the development of today's communication systems, quality of experience (QoE) becomes a widely used metric, which reflects the subjective experience of end users for wireless service. In addition, the energy efficiency is an increasingly important problem with the explosive growth in the amount of wireless terminals and nodes. Hence, a QoE-aware energy efficiency maximization based joint user access selection and power allocation approach is proposed to solve the problem. We transform the joint allocation process to an optimization of energy efficiency by establishing an energy efficiency model, and then the optimization problem is solved by chaotic clone immune algorithm (CCIA). Numerical simulation results indicate that the proposed algorithm can efficiently and reliably improve the QoE and ensure high energy efficiency of networks.