• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.67 no.1
• /
• pp.114-123
• /
• 2018

This paper concerns a design and learning method of softmax function neural networks based on K-means clustering. The partial discharge data Information is preliminarily processed through simulation using an Epoxy Mica Coupling sensor and an internal Phase Resolved Partial Discharge Analysis algorithm. The obtained information is processed according to the characteristics of the pattern using a Motor Insulation Monitoring System program. At this time, the processed data are total 4 types that void discharge, corona discharge, surface discharge and slot discharge. The partial discharge data with high dimensional input variables are secondarily processed by principal component analysis method and reduced with keeping the characteristics of pattern as low dimensional input variables. And therefore, the pattern classifier processing speed exhibits improved effects. In addition, in the process of extracting the partial discharge data through the MIMS program, the magnitude of amplitude is divided into the maximum value and the average value, and two pattern characteristics are set and compared and analyzed. In the first half of the proposed partial discharge pattern classifier, the input and hidden layers are classified by using the K-means clustering method and the output of the hidden layer is obtained. In the latter part, the cross entropy error function is used for parameter learning between the hidden layer and the output layer. The final output layer is output as a normalized probability value between 0 and 1 using the softmax function. The advantage of using the softmax function is that it allows access and application of multiple class problems and stochastic interpretation. First of all, there is an advantage that one output value affects the remaining output value and its accompanying learning is accelerated. Also, to solve the overfitting problem, L2-normalization is applied. To prove the superiority of the proposed pattern classifier, we compare and analyze the classification rate with conventional radial basis function neural networks.

• School Mathematics
• /
• v.8 no.3
• /
• pp.265-290
• /
• 2006

The purpose of this study is to search an effective teaching & learning program by examining how much does Excel affect on problem solving of linear function. This study was based on qualitative case study. Teaching experiment was performed for seven periods with five students in 8th graders. Pre and posts tests were attempted to analyze the changes of student's ability on problem solving of linear function. The analysis of tests were performed in category with correct process-object perspective, near process-object perspective, incorrect process-object perspective. According to this study, the subjects showed an improvement on problem solving perspective of linear function. This meant that lessons using Excel had influenced on the problem solving of linear function. We noticed that exploring the learning environment with Excel could supplement paper-and-pencil environment. We believed that Excel with an intuitive dynamic and explorative skills can play a role in scaffolding to support problem solving of linear function.

• Research in Vestibular Science
• /
• v.17 no.4
• /
• pp.134-141
• /
• 2018

Objectives: We aimed to study the role of vestibular input on spatial memory performance in mice that had undergone bilateral surgical labyrinthectomy, semicircular canal (SCC) occlusion and 4G hypergravity exposure. Methods: Twelve to 16 weeks old ICR mice (n=30) were used for the experiment. The experimental group divided into 3 groups. One group had undergone bilateral chemical labyrinthectomy, and the other group had performed SCC occlusion surgery, and the last group was exposed to 4G hypergravity for 2 weeks. The movement of mice was recorded using camera in Y maze which had 3 radial arms (35 cm long, 7 cm high, 10 cm wide). We counted the number of visiting arms and analyzed the information of arm selection using program we developed before and after procedure. Results: The bilateral labyrinthectomy group which semicircular canal and otolithic function was impaired showed low behavioral performance and spacial memory. The semicircular canal occlusion with $CO_2$ laser group which only semicircular canal function was impaired showed no difference in performance activity and spatial memory. However the hypergravity exposure group in which only otolithic function impaired showed spatial memory function was affected but the behavioral performance was spared. The impairment of spatial memory recovered after a few days after exposure in hypergravity group. Conclusions: This spatial memory function was affected by bilateral vestibular loss. Space-related information processing seems to be determined by otolithic organ information rather than semicircular canals. Due to otolithic function impairment, spatial learning was impaired after exposure to gravity changes in animals and this impaired performance was compensated after normal gravity exposure.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.10a
• /
• pp.110-113
• /
• 2022

Nowadays, breaking through the conventional traffic signal control method based on mathematical optimization, artificial intelligence began to be used in the area. In response to this trend, many studies are ongoing to figure out how to utilize AI technology properly for traffic signal optimization. They just simply focus on which method will work well besides lots of machine learning techniques and abandon the reward function engineering. In many cases, the reward function consists of the average delay of the vehicles in the intersection. However, this may lead to AI's misunderstanding about the traffic signal control: what AI regards as a good situation may not be realistic. Even the reward function itself may not meet the service level. Therefore, this study analyzes the problems of previous reward functions and will suggest how to reward function can be enhanced.

• Journal of Communications and Networks
• /
• v.12 no.5
• /
• pp.459-465
• /
• 2010

A new blind equalization algorithm that is based on maximizing the probability that the constant modulus errors concentrate near zero is proposed. The cost function of the proposed algorithm is to maximize the probability that the equalizer output power is equal to the constant modulus of the transmitted symbols. Two blind information-theoretic learning (ITL) algorithms based on constant modulus error signals are also introduced: One for minimizing the Euclidean probability density function distance and the other for minimizing the constant modulus error entropy. The relations between the algorithms and their characteristics are investigated, and their performance is compared and analyzed through simulations in multi-path channel environments. The proposed algorithm has a lower computational complexity and a faster convergence speed than the other ITL algorithms that are based on a constant modulus error. The error samples of the proposed blind algorithm exhibit more concentrated density functions and superior error rate performance in severe multi-path channel environments when compared with the other algorithms.

• Proceedings of the Korean Society of Machine Tool Engineers Conference
• /
• 1997.04a
• /
• pp.68-75
• /
• 1997

This paper presents a new approach to the design of neural-fuzzy controller for the speed and azimuth control of a track vehicle. The proposed control scheme uses a Gaussian function as a unit function in the frzzy-neural network, and back propagaton algorithm to train the fuzzy-neural network controller in the framework of the specialized learning architecture. It is proposed a learning controller consisting of two neural network-fuzzy based on independent reasoning and a connection net with fixed weights to simply the neural networks-fuzzy. The performance of the proposed controller is shown by performing the computer simulation for trajectory tracking of the speed and azimuth of a track vehicle driven by two independent wheels.

• Journal of the Korean Society of Industry Convergence
• /
• v.17 no.2
• /
• pp.69-76
• /
• 2014

This paper proposes a new approach to the design of cruise control system of a mobile robot with two drive wheel. The proposed control scheme uses a Gaussian function as a unit function in the fuzzy neural network, and back propagation algorithm to train the fuzzy neural network controller in the framework of the specialized learning architecture. It is proposed a learning controller consisting of two neural network-fuzzy based on independent reasoning and a connection net with fixed weights to simply the neural networks-fuzzy. The performance of the proposed controller is shown by performing the computer simulation for trajectory tracking of the speed and azimuth of a mobile robot driven by two independent wheels.

• Journal of Intelligence and Information Systems
• /
• v.3 no.1
• /
• pp.69-81
• /
• 1997

In this paper, first we propose an architecture of fuzzy neural networks with triangular fuzzy weights. The proposed fuzzy neural network can handle fuzzy input vectors. In both cases, outputs from the fuzzy network are fuzzy vectors. The input-output relation of each unit of the fuzzy neural network is defined by the extention principle of Zadeh. Also we define a cost function for the level sets(i. e., $\alpha$-cuts)of fuzzy outputs and fuzzy targets. Then we derive a learning algorithm from the cost function for adjusting three parameters of each triangular fuzzy weight. Finally, we illustrate our a, pp.oach by computer simulation examples.

• Proceedings of the KIEE Conference
• /
• 2005.07d
• /
• pp.2681-2683
• /
• 2005

In this paper, we present the structure of fuzzy neural network(FNN) based on wavelet function, and apply this network structure to the identification of nonlinear systems. For adjusting the shape of membership function and the connection weights, the parameter learning method based on the gradient descent scheme is adopted. And an approach that uses adaptive learning rates is driven via a Lyapunov stability analysis to guarantee the fast convergence. Finally, to verify the efficiency of our network structure. we compare the Identification performance of proposed wavelet based fuzzy neural network(WFNN) with those of the FNN, the wavelet fuzzy model(WFM) and the wavelet neural network(WNN) through the computer simulation.

• Journal of Institute of Control, Robotics and Systems
• /
• v.5 no.2
• /
• pp.189-199
• /
• 1999

This paper presents an optimization algorithm for a stable Self Dynamic Neural Network(SDNN) using genetic algorithm. Optimized SDNN is applied to a problem of controlling nonlinear dynamical systems. SDNN is dynamic mapping and is better suited for dynamical systems than static forward neural network. The real-time implementation is very important, and thus the neuro controller also needs to be designed such that it converges with a relatively small number of training cycles. SDW has considerably fewer weights than DNN. Since there is no interlink among the hidden layer. The object of proposed algorithm is that the number of self dynamic neuron node and the gradient of activation functions are simultaneously optimized by genetic algorithms. To guarantee convergence, an analytic method based on the Lyapunov function is used to find a stable learning for the SDNN. The ability and effectiveness of identifying and controlling a nonlinear dynamic system using the proposed optimized SDNN considering stability is demonstrated by case studies.