Seong-Jun Kim

PRINCIPAL COMPONENTS BASED SUPPORT VECTOR REGRESSION MODEL FOR ON-LINE INSTRUMENT CALIBRATION MONITORING IN NPPS

In nuclear power plants (NPPs), periodic sensor calibrations are required to assure that sensors are operating correctly. By checking the sensor’s operating status at every fuel outage, faulty sensors may remain undetected for periods of up to 24 months. Moreover, typically, only a few faulty sensors are found to be calibrated. For the safe operation of NPP and the reduction of unnecessary calibration, on-line instrument calibration monitoring is needed. In this study, principal componentbased auto-associative support vector regression (PCSVR) using response surface methodology (RSM) is proposed for the sensor signal validation of NPPs. This paper describes the design of a PCSVR-based sensor validation system for a power generation system. RSM is employed to determine the optimal values of SVR hyperparameters and is compared to the genetic algorithm (GA). The proposed PCSVR model is confirmed with the actual plant data of Kori Nuclear Power Plant Unit 3 and is compared with the Auto-Associative support vector regression (AASVR) and the auto-associative neural network (AANN) model. The auto-sensitivity of AASVR is improved by around six times by using a PCA, resulting in good detection of sensor drift. Compared to AANN, accuracy and cross-sensitivity are better while the auto-sensitivity is almost the same. Meanwhile, the proposed RSM for the optimization of the PCSVR algorithm performs even better in terms of accuracy, auto-sensitivity, and averaged maximum error, except in averaged RMS error, and this method is much more time efficient compared to the conventional GA method.

A Clustering Approach to Wind Power Prediction based on Support Vector Regression

A sustainable production of electricity is essential for low carbon green growth in South Korea. The generation of wind power as renewable energy has been rapidly growing around the world. Undoubtedly wind energy is unlimited in potential. However, due to its own intermittency and volatility, there are difficulties in the effective harvesting of wind energy and the integration of wind power into the current electric power grid. To cope with this, many works have been done for wind speed and power forecasting. It is reported that, compared with physical persistent models, statistical techniques and computational methods are more useful for short-term forecasting of wind power. Among them, support vector regression (SVR) has much attention in the literature. This paper proposes an SVR based wind speed forecasting. To improve the forecasting accuracy, a fuzzy clustering is adopted in the process of SVR modeling. An illustrative example is also given by using real-world wind farm dataset. According to the experimental results, it is shown that the proposed method provides better forecasts of wind power.

A Prediction Method of the Gas Pipeline Failure Using In-line Inspection and Corrosion Defect Clustering

Corrosion has a significant influence upon the reliability assessment and the maintenance planning of gas pipeline. Corrosion defects occurred on the underground pipeline can be obtained by conducting periodic in-line inspection (ILI). However, little study has been done for practical use of ILI data. This paper deals with remaining lifetime prediction of the gas pipeline in the presence of corrosion defects. Because a pipeline parameter includes uncertainty in its operation, a probabilistic approach is adopted in this paper. A pipeline fails when its operating pressure is larger than the pipe failure pressure. In order to estimate the failure probability, this paper uses First Order Reliability Method (FORM) which is popular in the field of structural engineering. A well-known Battelle code is chosen as the computational model for the pipe failure pressure. This paper develops a Matlab GUI for illustrating failure probability predictions Our result indicates that clustering of corrosion defects is helpful for improving a prediction accuracy and preventing an unnecessary maintenance.

A Short-Term Wind Speed Forecasting Through Support Vector Regression Regularized by Particle Swarm Optimization

A sustainability of electricity supply has emerged as a critical issue for low carbon green growth in South Korea. Wind power is the fastest growing source of renewable energy. However, due to its own intermittency and volatility, the power supply generated from wind energy has variability in nature. Hence, accurate forecasting of wind speed and power plays a key role in the effective harvesting of wind energy and the integration of wind power into the current electric power grid. This paper presents a short-term wind speed prediction method based on support vector regression. Moreover, particle swarm optimization is adopted to find an optimum setting of hyper-parameters in support vector regression. An illustration is given by real-world data and the effect of model regularization by particle swarm optimization is discussed as well.

An Application of Fuzzy Logic with Desirability Functions to Multi-response Optimization in the Taguchi Method

Although it is widely used to find an optimum setting of manufacturing process parameters in a variety of engineering fields, the Taguchi method has a difficulty in dealing with multi-response situations in which several response variables should be considered at the same time. For example, electrode wear, surface roughness, and material removal rate are important process response variables in an electrical discharge machining (EDM) process. A simultaneous optimization should be accomplished. Many researches from various disciplines have been conducted for such multi-response optimizations. One of them is a fuzzy logic approach presented by Lin et al. [1]. They showed that two response characteristics are converted into a single performance index based upon fuzzy logic. However, it is pointed out that information regarding relative importance of response variables is not considered in that method. In order to overcome this problem, a desirability function can be adopted, which frequently appears in the statistical literature. In this paper. we propose a novel approach for the multi-response optimization by incorporating fuzzy logic into desirability function. The present method is illustrated by an EDM data of Lin and Lin [2].

A Fuzzy Inference based Reliability Method for Underground Gas Pipelines in the Presence of Corrosion Defects

Remaining lifetime prediction of the underground gas pipeline plays a key role in maintenance planning and public safety. One of main causes in the pipeline failure is metal corrosion. This paper deals with estimating the pipeline reliability in the presence of corrosion defects. Because a pipeline has uncertainty and variability in its operation, probabilistic approximation approaches such as first order second moment (FOSM), first order reliability method (FORM), second order reliability method (SORM), and Monte Carlo simulation (MCS) are widely employed for pipeline reliability predictions. This paper presents a fuzzy inference based reliability method (FIRM). Compared with existing methods, a distinction of our method is to incorporate a fuzzy inference into quantifying degrees of variability in corrosion defects. As metal corrosion depends on the service environment, this feature makes it easier to obtain practical predictions. Numerical experiments are conducted by using a field dataset. The result indicates that the proposed method works well and, in particular, it provides more adviso ryestimations of the remaining lifetime of the gas pipeline.

Analysis of defective patterns on wafers in semiconductor manufacturing: A bibliographical review

The existing works on automatic detection and/or classification of clusters of defective dies on wafers is reviewed. The literature is classified into three major categories, namely, spatial randomness test, automatic cluster detection only, and automatic detection and classification of clusters. Future research directions are also discussed.