Jungwon Yu

Measuring System for Impact Point of Arrow using Mamdani Fuzzy Inference System

The performance of arrow from a manufacturing process depends on arrow`s trajectory(archer`s paradox) and intensity of an impact points. Especially, when conducting a shooting experiment over and over in the same experiment condition, the intensity of impact point is an objective standard to judge the performance of the arrow. However, the analysis method for the impact point is not enough, a previous research of the arrow`s performance has been focused on a skill to optimize a manufacturing variables(feathers of an arrow, barb of an arrow, arrow`s shaft, weight, external diameter, spine). In this paper, We propose measurement system of arrow`s impact point with Mamdani fuzzy inference system and similarity of polygon for automation of impact point`s measurement. Measuring the impact point data of the arrow moving with a high speed(approximately 275km/h) by using line laser and photo diode array, then the measured data are mapped to arrow`s impact point with fuzzy inference and similarity of polygon.

A Clustering-Based Fault Detection Method for Steam Boiler Tube in Thermal Power Plant

System failures in thermal power plants (TPPs) can lead to serious losses because the equipment is operated under very high pressure and temperature. Therefore, it is indispensable for alarm systems to inform field workers in advance of any abnormal operating conditions in the equipment. In this paper, we propose a clustering-based fault detection method for steam boiler tubes in TPPs. For data clustering, k-means algorithm is employed and the number of clusters are systematically determined by slope statistic. In the clustering-based method, it is assumed that normal data samples are close to the centers of clusters and those of abnormal are far from the centers. After partitioning training samples collected from normal target systems, fault scores (FSs) are assigned to unseen samples according to the distances between the samples and their closest cluster centroids. Alarm signals are generated if the FSs exceed predefined threshold values. The validity of exponentially weighted moving average to reduce false alarms is also investigated. To verify the performance, the proposed method is applied to failure cases due to boiler tube leakage. The experiment results show that the proposed method can detect the abnormal conditions of the target system successfully.

Short-term hourly load forecasting using PSO-based AR model

Load forecasting is essential for effective and stable power system planning and operation. Decision making related to power system operation is influenced by futures electric load patterns. In this paper, particle swarm optimization (PSO) based autoregressive (AR) model is presented for short-term hourly load forecasting. First of all, among several potential input candidates, relevant inputs that have high correlation with prediction models output are selected. According to the number of selected inputs, the order of AR model is fixed. Finally, AR models parameters are optimized using PSO that is a global optimization algorithm. To verify the performance, the proposed method is applied to two kinds of real world hourly load dataset in South Korea. The proposed method shows good prediction accuracy.

A Study on Chaff Echo Detection using AdaBoost Algorithm and Radar Data

In pattern recognition field, data classification is an essential process for extracting meaningful information from data. Adaptive boosting algorithm, known as AdaBoost algorithm, is a kind of improved boosting algorithm for applying to real data analysis. It consists of weak classifiers, such as random guessing or random forest, which performance is slightly more than 50% and weights for combining the classifiers. And a strong classifier is created with the weak classifiers and the weights. In this paper, a research is performed using AdaBoost algorithm for detecting chaff echo which has similar characteristics to precipitation echo and interrupts weather forecasting. The entire process for implementing chaff echo classifier starts spatial and temporal clustering based on similarity with weather radar data. With them, learning data set is prepared that separated chaff echo and non-chaff echo, and the AdaBoost classifier is generated as a result. For verifying the classifier, actual chaff echo appearance case is applied, and it is confirmed that the classifier can distinguish chaff echo efficiently.

Chaff Echo Detecting and Removing Method using Naive Bayesian Network

Chaff is a kind of matter spreading atmosphere with the purpose of preventing aircraft from detecting by radar. The chaff is commonly composed of small aluminum pieces, metallized glass fiber, or other lightweight strips which consists of reflecting materials. The chaff usually appears on the radar images as narrow bands shape of highly reflective echoes. And the chaff echo has similar characteristics to precipitation echo, and it interrupts weather forecasting process and makes forecasting accuracy low. In this paper, the chaff echo recognizing and removing method is suggested using Bayesian network. After converting coordinates from spherical to Cartesian in UF (Universal Format) radar data file, the characteristics of echoes are extracted by spatial and temporal clustering. And using the data, as a result of spatial and temporal clustering, a classification process for analyzing is performed. Finally, the inference system using Bayesian network is applied. As a result of experiments with actual radar data in real chaff echo appearing case, it is confirmed that Bayesian network can distinguish between chaff echo and non-chaff echo.

Identifying chaff echoes in weather radar data using tree-initialized fuzzy rule-based classifier

In order to produce reliable weather forecasts, it is essential to discriminate non-meteorological targets from rain clouds in weather radar data. Identification of chaff echoes, which is one of the main noise sources, is uncertain and imprecise for skilled weather experts because characteristics of them are similar to those of precipitation echoes. This paper uses tree-initialized fuzzy classifier (FC) to identify chaff echoes. Fuzzy models have been widely applied to the domain of uncertainty and vagueness. Classification and regression tree is used to generate an initial crisp model (a set of crisp rules). The number of the rules, corresponding to complexity of the model, is systematically determined by performance criterion. Finally, after transforming the crisp model to the fuzzy one straightforwardly, parameters of the FCs are optimized by genetic algorithms. FCs have more flexible decision boundaries than binary decision trees with rectangular partitioning. In order to evaluate identification performance, the FCs, and comparison methods are applied to many cases where both chaff and non-chaff echoes occurred simultaneously. The results of experiments show that the FCs achieve the best identification performance.

Locally-Weighted Polynomial Neural Network for Daily Short-Term Peak Load Forecasting

Electric load forecasting is essential for effective power system planning and operation. Complex and nonlinear relationships exist between the electric loads and their exogenous factors. In addition, time-series load data has non-stationary characteristics, such as trend, seasonality and anomalous day effects, making it difficult to predict the future loads. This paper proposes a locally-weighted polynomial neural network (LWPNN), which is a combination of a polynomial neural network (PNN) and locally-weighted regression (LWR) for daily short-term peak load forecasting. Model over-fitting problems can be prevented effectively because PNN has an automatic structure identification mechanism for nonlinear system modeling. LWR applied to optimize the regression coefficients of LWPNN only uses the locally-weighted learning data points located in the neighborhood of the current query point instead of using all data points. LWPNN is very effective and suitable for predicting an electric load series with nonlinear and non-stationary characteristics. To confirm the effectiveness, the proposed LWPNN, standard PNN, support vector regression and artificial neural network are applied to a real world daily peak load dataset in Korea. The proposed LWPNN shows significantly good prediction accuracy compared to the other methods.

Automatic structure identification of TSK fuzzy model for stock index forecasting

Among financial time-series analysis tasks, stock index forecasting has been considered as one of the challenging and difficult tasks. Since an accurate stock index prediction enhances stock market returns, it is highly promising research area and has attracted particular attention. In this paper, to predict stock index with complex and nonlinear characteristics, an automatic structure identification (SI) method of TSK fuzzy model is proposed. Typically, SI procedures of fuzzy models consist of relevant input selection, fuzzy rule generation and parameter search space determination. In this study, mutual information is employed to select relevant input variables and fuzzy c-means (FCM) clustering algorithm is used to generate fuzzy if-then rules. In FCM clustering, the number of clusters should be fixed in advance. This paper uses performance criterion to determine the optimal number of clusters in FCM clustering. After deciding the optimal cluster number, fuzzy if-then rules are extracted and parameter search space boundaries are fixed. Finally, premise and consequent parameters are optimized by cooperative random learning particle swarm optimization proposed by Zhao et al. To confirm the effectiveness, the proposed TSK fuzzy modeling method and comparison methods are applied to Korea Composite Stock Price Index dataset. The experimental results show that the TSK fuzzy models with the proposed SI method outperform comparison methods.

A Study of Positioning Error Compensation Using Optical-Sensor and Three-Frame

This paper studies about the developed measuring instrument’s degree of precision. Through this research, we could convert archer’s paradox phenomenon to numerical data and make it use as a performance evaluation element. The instrument uses three frames which composed of line-laser and photodiode sensors for reconstructing full-shape of a flying arrow. Moreover, in order to measure arrows’ position precisely which flies about 300km/h, Artificial Neural Network is used for calibration of the measuring instrument. Two grid-plates for calibrating the measuring instrument are installed in first and third frames among three frames. After that, calibration process is performed using converted coordinate data from arrow’s position by placing an arrow at holes in the grid-plate. From the suggested arrow position measurement system and the experiment data using the instrument, the automated system for quality control of arrows or performance experiment could be established.

Genetic-Based k-Nearest Neighbor for Chaff Echo Detection

This paper proposes genetic-based k-nearest neighbor method for chaff echo identification. Weather radar provides various data: location, velocity, direction, and range of typhoon or precipitation, precipitation intensity, altitude and location of thunderstorm and rainfall. Above this data, topography echo, anomalous echo, second echo and chaff echo are observed from weather radar, and they are disrupt weather forecasting. They are called non-weather echo. In order to improve weather forecasting, we propose genetic-basedk-nearest neighbor for chaff echo identification. Experimental result shows that chaff echoes are well removed, so performance weather forecasting will also be improved.