Jae-Hoon Cho

Parameter Optimization of Extreme Learning Machine Using Bacterial Foraging Algorithm

Recently, Extreme learning machine(ELM), a novel learning algorithm having much faster than the traditional gradient-based learning algorithm, was proposed for single-hid den-layer feedforward neural networks (SLFNs). Usually, the initial input weights and hidden biases of ELM are randomly chosen, and then the output weights are analytically determined by using Moore-Penrose (MP) generalized inverse. However, ELM may need higher number of hidden neurons due to the random determination of the input weights and hidden biases. In this paper, an optimization method based on the bacterial foraging (BF) algorithm is proposed to adjust the input weights and hidden biases. Experimental result shows that this method can achieve better performance for problems having higher dimension than others.

A Variable Step Size Incremental Conductance MPPT of a Photovoltaic System Using DC-DC Converter with Direct Control Scheme

This paper presents a novel maximum power point tracking for a photovoltaic power (PV) system with a direct control plan. Maximum power point tracking (MPPT) must usually be integrated with photovoltaic (PV) power systems so that the photovoltaic arrays are able to deliver maximum available power. The maximum available power is tracked using specialized algorithms such as Perturb and Observe (P&O) and incremental Conductance (indCond) methods. The proposed method has the direct control of the MPPT algorithm to change the duty cycle of a dc-dc converter. The main difference of the proposed system to existing MPPT systems includes elimination of the proportional.integral control loop and investigation of the effect of simplifying the control circuit. The proposed method thus has not only faster dynamic performance but also high tracking accuracy. Without a conventional controller, this method can control the dc-dc converter. A simulation model and the direct control of MPPT algorithm for the PV power system are developed by Matlab/Simulink, SimPowerSystems and Matlab/Stateflow.

Neuro-Fuzzy Rule Generation for Backing up Navigation of Car-like Mobile Robots

An automatic neuro-fuzzy rule generation scheme is proposed for backing up navigation of car-like mobile robots. The proposed method is based on the Conditional Fuzzy C-Means (CFCM) and Fuzzy Equalization (FE) methods. The CFCM is adopted to render clusters, which can represent the homogeneous properties of the given input and output fuzzy data, and also the FE method is used to systematically construct the fuzzy membership functions for the ANFIS. From these, a compact size of fuzzy rules can be automatically obtained, which satisfy the given goal. The proposed method has been applied to a truck, and also to a truck-trailer backing up navigation problem, and good results have been achieved in comparison to previous work.

Renewable Source and Hybrid System Modeling for Smart Grid

Recently, smart grid for solving energy problems have been receiving growing attention. Also, renewable energy sources such as photovoltaic and fuel cell as future energy for realizing smart grid have been widely studied. On the other hand, hybrid structures have been proposed since the output power of these renewable energy sources is usually dependent on weather conditions. This paper proposes a hybrid system involving a proper photovoltaic in the hybrid system, Polymer Elecrolyte Membrane Fuel Cell with water electrolyzer and ultracapacitor. The results of simulation and output of the proposed model are established and analysed by Matlab/Simulink and SimPowerSystems.

A Study on the Comparison of the Stand-alone operation Characteristics and Matlab/Simulink Dynamic Model of PEMFC and PEMFC/SCB

The available power generated from the FC power plant may not be sufficient to meet sustained load demand, peak demand or transient events. An supercapacitor bank(SCB) can supply a large burst of power, but it cannot store a significant amount of energy. The combined use of FC and SCB has the potential for the better energy efficiency, reducing the cost of FC technology and improved dynamic response. In this paper, A single PEMFC and PEMFC operated in parallel with a SC bank are presented, A new dynamic model of PEMFC system, the converter and controller has been developed for stand-alone applications. The simulation results are presented using Matlab/Simulink, and SimPowerSystems environments. It is confirmed that the results show a good performance and stable DC-link voltage for proposed dynamic and mathematical models developed for the combined FC/SCB.

Power Control modeling and Simulation of Hybrid Power System for Building Microgrid Connected Application

In this paper, we propose to study the possibility of using a photovoltaic system combined with a high speed micro-turbine. This hybrid system can work as stand-alone system or grid connected system as it will be a part of a micro-grid. Initially, we propose Matlab/Simulink dynamic models of photovoltaic, micro turbine systems and supercapacitor. Then, we carry out a simulation comparison of the two systems, this is, with supercapacitor and without supercapacitor bank. We show that supercapacitor bank as short-term storage device was necessary to reduce the fast fluctuation of power in the case of sensitive loads. It is verified the simulation results of Matlab/Simulink based hybrid power system represent the effectiveness of the suggested hybrid power system.

Feature Extraction and Classification of High Dimensional Biomedical Spectral Data

In this paper, we propose the biomedical spectral pattern classification techniques by the fusion scheme based on the SpPCA and MLP in extended feature space. A conventional PCA technique for the dimension reduction has the problem that it can`t find an optimal transformation matrix if the property of input data is nonlinear. To overcome this drawback, we extract features by the SpPCA technique in extended space which use the local patterns rather than whole patterns. In the classification step, individual classifier based on MLP calculates the similarity of each class for local features. Finally, biomedical spectral patterns is classified by the fusion scheme to effectively combine the individual information. As the simulation results to verify the effectiveness, the proposed method showed more improved classification results than conventional methods.

Gait Recognition and Person Identification for Surveillance Robots

The surveillance robot has been an important component in the field of service robot industry. In the surveillance robot technology, one of the most important technology is to identify a person. In this paper, we propose a gait recognition method based on contourlet and fuzzy LDA (Linear Discriminant Analysis) for surveillance robots. After decomposing a gait image into directional subband images by contourlet, features are obtained in each subband by the fuzzy LDA. The final gait recognition is performed by a fusion technique that effectively combines similarities calculated respectively in each local subband. To show the effectiveness of the proposed algorithm, various experiments are performed for CBNU and NLPR DB datasets. From these, we obtained better classification rates in comparison with the result produced by previous methods.

Feature Selection by Genetic Algorithm and Information Theory

In the pattern classification problem, feature selection is an important technique to improve performance of the classifiers. Particularly, in the case of classifying with a large number of features or variables, the accuracy of the classifier can be improved by using the relevant feature subset to remove the irrelevant, redundant, or noisy data. In this paper we propose a feature selection method using genetic algorithm and information theory. Experimental results show that this method can achieve better performance for pattern recognition problems than conventional ones.

Comparative Study for Optimal Sizing of a Hybrid PV/Wind/Diesel/Battery Storage System by Metaheuristic Optimization Methods

In a hybrid energy system, different energy sources (photovoltaic (PV), wind, diesel, etc.) as well as energy storage devices are connected together to supply the electrical load. Since the produced power of PV and wind turbine (WT) is dependent on the variation of the resources (sun and wind) and the load demand fluctuates, the main attribute of such hybrid systems is the ability of satisfying the load at any time and storing the excess energy for the later use in deficit conditions. This paper presents a methodology to size and to optimize a stand-alone hybrid PV/Wind/Diesel/Battery bank minimizing the Total annual cost and Loss of Power Supply Probability (LPSP) using a GA and PSO based optimization algorithm respectively. The effectiveness of the proposed method was verified by Matlab software.In this paper, first the mathematical model of various parts of hybrid system is presented. Then, the proposed algorithm is used. Finally, simulation results (number of PV panels, number of wind turbines, number of battery storages, system total cost,power diagram of hybrid power system components) for solar-wind -diesel systems is presented.The simulation results of the proposed approach show that the use of PSO can be more efficient than GA in the size optimization of hybrid energy system.