• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.63 no.11
• /
• pp.1608-1614
• /
• 2014

Power demand forecasting is an important factor of the peak management. This paper deals with the 15 minutes ahead load forecasting problem in a DC urban railway system. Since supplied power lines to trains are connected with parallel, the load characteristics are too complex and highly non-linear. The main idea of the proposed method for the 15 minutes ahead prediction is to use the daily load similarity accounting for the load nonlinearity. An Euclidean norm with weighted factors including loads of the neighbor substation is used for the similar load selection. The prediction value is determinated by the sum of the similar load and the correction value. The correction has applied the neural network model. The feasibility of the proposed method is exemplified through some simulations applied to the actual load data of Incheon subway system.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.27 no.6B
• /
• pp.586-592
• /
• 2002

In this paper, new channel scheduling algorithms which reduce transmission delay caused by wireless network are proposed in AMC/TDM wireless data communication. The concept of the Proposed algorithms is based on the proportional fairness, M-LWDF, and performance-guaranteeing algorithm proposed by Xin Liu. The proposed algorithm can be applied to QoS guaranteeing services as well as best-effort services. Simulation results show that new algorithm reduced transmission delay upto 11.5% in case of proportional fairness algorithm and also decreased transmission delay unto 9% in case of M-LWBF algorithm.

• Electrical & Electronic Materials
• /
• v.10 no.8
• /
• pp.783-793
• /
• 1997

A VLSI implementation of a self-learning neural network integrated circuits using a linearity synapse transistor is investigated. The thickness dependence of oxide current density stress current transient current and channel current has been measured in oxides with thicknesses between 41 and 112 $\AA$, which have the channel width $\times$ length 10 $\times$1${\mu}{\textrm}{m}$, 10 $\times$ 0.3${\mu}{\textrm}{m}$ respectively. The transient current will affect data retention in synapse transistors and the stress current is used to estimate to fundamental limitations on oxide thicknesses. The synapse transistor has represented the neural states and the manipulation which gaves unipolar weights. The weight value of synapse transistor was caused by the bias conditions. Excitatory state and inhitory state according to weighted values affected the drain source current.

• Proceedings of the KIEE Conference
• /
• 1998.07b
• /
• pp.566-568
• /
• 1998

In this paper, we suggest a optimal design method of Fuzzy-Neural Networks model for complex and nonlinear systems. FNNs have the stucture of fusion of both fuzzy inference with linguistic variables and Neural Networks. The network structure uses the simpified inference as fuzzy inference system and the BP algorithm as learning procedure. And we use a clustering algorithm to find initial parameters of membership function. The parameters such as membership functions, learning rates and momentum coefficients are easily adjusted using the genetic algorithms. Also, the performance index with weighted value is introduced to achieve a meaningful balance between approximation and generalization abilities of the model. To evaluate the performance index, we use the time series data for gas furnace and the sewage treatment process.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.32 no.3
• /
• pp.1-15
• /
• 2007

In this paper, the return link of interactive GEO satellite network providing multimedia services is considered. First, we classify data by delay characteristics, and analyze the numbers of expected lost packets and expected delay packets for each data class of each terminal. Next we mathematically formulate optimal rate scheduling model to minimize the weighted sum of the numbers of expected lost packets and expected delay packets considering priority of each data class. We also suggest a dynamic rate scheduling scheme based on Lagrangean relaxation technique and subgradient technique to solve the proposed model in a fast time. Extensive experiments show that the proposed scheme provides encouraging results.

• 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.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1993.06a
• /
• pp.1118-1120
• /
• 1993

Fuzzy logic places a considerable burden on an inference engine for applications such as control or approximate reasoning. Various neural network architectures have been proposed to deal with the computational task, and yet, maintain flexibility in the desired traits of the final system. Recently, we introduced a trainable network architecture whose nodes implement weighted Yager additive hybrid operators for fuzzy logic inference in an approximate reasoning setting. In this paper we examine the utility of such networks for control situations. We show that they are capable of learning control functions which are piece-wise monotonic in each of the variables. The learning ability is demonstrated through an example.

• Proceedings of the Korean Institute of Electrical and Electronic Material Engineers Conference
• /
• 2000.07a
• /
• pp.59-62
• /
• 2000

A VLSI implementation of a self-learning neural network integrated circuits using a linearity synapse transistor is investigated. The thickness dependence of oxide current density, stress current, transient current and channel current has been measured in oxides with thicknesses between 41 and 112 $\AA$, which have the channel width$\times$length 10$\times$1${\mu}{\textrm}{m}$ respectively. The transient current will affect data retention in synapse transistors and the stress current is used to estimate to fundamental limitations on oxide thicknesses. The synapse transistor has represented the neural states and the manipulation which gave unipolar weights. The weight value of synapse transistor was caused by the bias conditions. Excitatory state and inhitory state according to weighted values affected the drain source current.

• Journal of Korea Multimedia Society
• /
• v.22 no.6
• /
• pp.665-675
• /
• 2019

In this paper, we compare and analyze the classification performance of deep learning algorithm Convolutional Neural Network(CNN) ac cording to ensemble generation and combining techniques. We used several CNN models(VGG16, VGG19, DenseNet121, DenseNet169, DenseNet201, ResNet18, ResNet34, ResNet50, ResNet101, ResNet152, GoogLeNet) to create 10 ensemble generation combinations and applied 6 combine techniques(average, weighted average, maximum, minimum, median, product) to the optimal combination. Experimental results, DenseNet169-VGG16-GoogLeNet combination in ensemble generation, and the product rule in ensemble combination showed the best performance. Based on this, it was concluded that ensemble in different models of high benchmarking scores is another way to get good results.

• Smart Media Journal
• /
• v.12 no.7
• /
• pp.52-58
• /
• 2023

Alzheimer's disease is one type of dementia, the symptoms can be treated by detecting the disease at its early stages. Recently, many computer-aided diagnosis using magnetic resonance image(MRI) have shown a good results in the classification of AD. Taken these MRI images and feed to Free surfer software to extra the features. In consideration, using T1-weighted images and classifying using the convolution neural network (CNN) model are proposed. In this paper, taking the subjects from ADNI of subcortical and cortical features of 190 subjects. Consider the study to reduce the complexity of the model by using the single layer in the Res-Net, VGG, and Alex Net. Multi-class classification is used to classify four different stages, CN, EMCI, LMCI, AD. The following experiment shows for respective classification Res-Net, VGG, and Alex Net with the best accuracy with VGG at 96%, Res-Net, GoogLeNet and Alex Net at 91%, 93% and 89% respectively.