Applied Mechanics and Materials Vols. 530-531

Abstract: The analysis of electromagnetic interferences (EMI) has been a heated problem in the field of Electromagnetic Compatibility (EMC). As the demand of efficiency and effectiveness is getting higher, the traditional methods have become the short board in analysis process. These methods havent provided a solution to analyze the relation among multiple EMI signals, and the data clustering and mining are currently done manually. To address this problem, in this paper we propose a one-stop method based on the wavelet packet decomposition (WPD) and self-organized feature map (SOM), aiming to provide a systematical and solution to extract and analyze multiple EMI signals. Experimental results are also provided to demonstrate the validity and efficiency of the proposed method.

Abstract: This paper presents a feature extraction method based on LMD and MSE for reciprocating compressor according to the strong nonstationarity, nonlinearity and features coupling characteristics of vibration signal. The vibration signal was decomposed into a set of PFs, and then multiscale entropy of the first several PFs were calculated as feature vectors with different scale factors. Based on the maximum of average Euclidean distances, the feature vectors which have the best divisibility were selected. The feature vectors of reciprocating compressor at different bearing clearance states were extracted using this method, and superiority of this method is verified by comparing with the results of sample entropy.

Abstract: A thermal radiation monitoring system was established to measure the thermal radiation distribution of the crude oil tank fires, and the effect of open area, horizontal and vertical distance on the radiant heat flux was analyzed. The experimental results are compared with the calculated values by the point source model. It shows that the point source model have a high accuracy to calculate the radiant heat flux of the crude oil tank fire when the distance between the target point and the flame center is larger than the diameter of the tank.

Abstract: Due to the high ground fault resistance and the complexity of power distribution network structure (such as too many nodes, branches and too long lines), adopting common traveling wave method and ac injection method can not effectively locate the single-phase grounding fault in the distribution network system.To solve above problems and determine the position of the point of failure prisely, this paper adopted the dc location method of injecting the dc signal from the point of failure under the power outage offline. This paper introduces the single phase dc method and the method of three phase dc, and the simulation shows that the dc location method is effective and feasible.

Abstract: The chaotic system is sensitive to initial value,and this characteristic can be applied in the interference signal detection.There are periodic state,critical state and chaotic state in a chaotic system.When the system is in the critical state,a small perturbation of system parameter may lead to a qualitative change of the systems state.This paper introduces a method to detecting interference signals based on the chaotic system and a new idea of using it into the detection of distribution network disturbance when distributed generation is connected to the grid.In this paper we choose the duffing oscillator as a chaotic system to study,use MATLAB to carry on the simulation,and research the changes of the system when the interference signal to be detected is added to the system.By means of observing the phase locus chart and time domain chart the interference signal will be detected.

Abstract: Based on the credibility theory, this paper is devoted to the fuzzy programming problem. The expected-value model of fuzzy programming problem is provided under credibility theory. For solving the fuzzy programming problem efficiently, Latin Hypercube Sampling, fuzzy simulation, Support Vector Machine and Artificial Bee Colony algorithm are integrated to build a hybrid intelligent algorithm. The proposed method has excellent consistency and efficiency in solving fuzzy programming problem, and is particularly useful for expensive systems.

Abstract: In this paper, a modified limited memory BFGS method for solving large-scale unconstrained optimization problems is proposed. A remarkable feature of the proposed method is that it possesses a global convergence property even without convexity assumption on the objective function. The implementations of the algorithm on CUTE test problems are reported, which suggest that a slight improvement has been achieved.

Abstract: The techniques and the processes to divide the image into several parts which have different features and to pick up foreground are called image segmentation. In this work, we propose a new approach for gray scale image segmentation based on level set method. At first, every pixel on the image is divided into either similar-property class or dissimilar-property class based on the variance of a small area centered at the pixel. Then, the velocity of curve evolution for these two classes is defined respectively. It is determined by a value called the dissimilarity of the area. Experimental results show that this approach can obtain good segmentation results of artificial images and real medical images fast and accurately.

Abstract: The existence of specular highlights is a great obstacle of shape-from-Shading. For a single gray-scale image with only intensity information, the existed highlights detection methods based on chroma or polarization analysis cannot directly be applied on it. So, a method using surface shape is provided, it makes full use of the imaging process. Through the surface normal estimation, compute the maximum a posteriori probability of each reflection composition under simulated annealing and detect the highlights areas. Finally, remove the highlights based on the assumption of curvature continuity. Experiments reveal that this method is effective on synthetic and real-world images, improves the accuracy of surface recovery for image combined specular highlights.

Abstract: To find the need of images from Internet is a heavy work, image annotation techniques can be used to help us quickly find the required images, the traditional image annotation method can't between the image and text classification, cannot achieve most web users demand, this paper proposes an image semantic annotation technology automatically, based on the SVM technology. The first we map image visual feature to one or more images of rough concept by using annotation, then we preprocess web page text information, lastly, we selected high similarity of keywords as image annotation. From the experimental effect it can effectively improve the image tagging precision.