Papers by Keyword: Extreme Value

Paper TitlePage

Authors: Hai Xia Guo
Abstract: In this paper, we proposed two improved methods for impulse noise of image. Firstly we propose an improved multiple extreme value examine for pixels for pixels. Secondly we propose a variable threshold ,a point-set composed with prior pixels was proposed to determine the threshold value. The approach makes noise well-detected and can reduce lost and false noise relatively. Experiments confirm the validity and effectiveness of the method , and has the less computation.
Authors: Costas P. Providakis
Abstract: This paper presents the use of statistically rigorous algorithms combined with electromechanical (E/M) impedance approach for health monitoring of engineering structures. In particular, a statistical pattern recognition procedure is developed, based on frequency domain data of electromechanical impedance, to establish a decision boundary for damage identification. In order to diagnose damage with statistical confidence, health monitoring is cast in the context of outlier detection framework. Inappropriate modeling of tail distribution of outliers imposes potentially misleading behavior associated with damage. The present paper attempts to address the problem of establishing decision boundaries based on extreme value statistics so that the extreme values of outliers associated with tail distribution can be properly modeled. The validity of the proposed method is demonstrated using finite element method (FEM) simulated data while a comparison is performed for the extreme value analysis results contrasted with the standard approach where it is assumed that the damage-sensitive features are normally distributed.
Authors: Wan Sharil Ahmad Termizi, Mohd Shahir Liew
Abstract: Prediction of extreme environmental loadings is crucial in the design of offshore platforms. Combinations of independent 100-year loadings are usually used assuming that it will bring the maximum value that is suitable for the design. It is rather impossible for such independent combinations to occur at the same time. Addressing this issue requires approaches in two methods. The first would be to find the probabilities of joint effect of these parameters, while the second would be to forecast the extreme value of each parameter. Joint densities and extreme value analysis have become available due to significant advancements in fluid dynamics and computer science. By using the combination of these two techniques, the actual loading of wind and wave can be obtained, leading to optimum and robust design.
Authors: Mojtaba Mahdavi, Mohamad Mahdavi
Abstract: —This paper explores a new method to find most reliable lifetime distribution function of systems, via an applied heuristic model. In most operating systems which complicity of system obligates high degree of reliability, this model will be use to propose a more appropriate reliability function between lifetime distribution based and another that is based on relevant Extreme Value distribution. A numeric study will be reviewed to clear the concepts of this paper.
Authors: V.S. Deeva, S.М. Slobodyan, V.S. Teterin
Abstract: Retaining structure of homogeneous fluid and granular stream is one of the main criteria for technological process assuring the high quality outcome in many industries, including mechanical engineering and oil & gas industry. For example, in oil and gas industry during the pipeline transportation of oils there is a strong trend for cluster aggregation, and particle coarsening and entanglement. Dehomogenization of particle stream results in reverse dynamics of the stream. The importance of prevention and minimization of small particles coalescence by separating the oil stream leads to the need of improving the properties of the dispersers to boost their efficiency. Our paper investigates the operating principle of the disperser for separating particles (separator), which is designed by the authors. We have considered a particle stream of dispersed structure. We have obtained the conformity with the stability of the disperser operation. To yield the results we use the extremum problems for differential equations. This approach provides strong evidence that there are optimum parameters of the dispersers, which result in better stability of the particle stream.
Authors: Yu Sheng Quan, Dai Juan Wang, Guang Chen, Zong Cheng Zhang
Abstract: Power transformer is the important equipment in power system, once the occurrence of faults may bring serious harm to power grid, and even cause great losses. This paper proposes a transformer winding insulation defect detection method. The method only uses the impulse voltage test. It can obtain the diagnosis signal source which contains a lot of harmonic. This paper can judge the insulation defects in the location according to compare the extreme value distribution of every winding voltage and currents correlation coefficient .and can determine the coordinate values. The method does not need large or complex additional equipment, can realize the detection and diagnosis of transformer winding insulation defect location. It is image, intuitive, convenient and practical. This paper studies the insulation defects and the severity of transformer winding, puts forward the method for changing winding current and voltage detection winding insulation defects. And it uses a large number of experiments to verify the insulated and reliability of transformer winding according to different position, different types of insulation defects.
Showing 1 to 6 of 6 Paper Titles