Majlesi Journal of Electrical Engineering
Volume:9 Issue: 4, Dec 2015

Looking in to the modern technology, we see new inventions and innovations which are unique. As an example, over the past years, new technology has been created in computer networks of which metropolitan area network (MAN) has had a noticeable advance in urban communication. Therefore the aim of this article is to provide and compile information regarding MAN technology and MPLS. Utilizing Ethernet in urban networks needs the developing potential and strength that only exists in MPLS and IP levels. Layer two and three joint networks are solutions that join simplicity and economy of Ethernet at IP/MPLS scale. Different transfer technologies have been used in MAN so metropolitan services should be given on a mixture of optic and data switches, MPLS have been developed for this work and using generalized-MPLS makes MPLS deployments possible even on none-IP-aware switches. In the beginning theories about using MAN technology to produce urban networks are explained. Then a set of problems regarding the management of resources and provisioning barriers are introduced following the solutions by means of IP/MPLS network which have been implemented and validated in eSNP simulator.

In this paper, two adaptive H∞ control schemes based on a genetic wavelet kernel support vector machine (SVM) and a hybrid genetic wavelet kernel SVM is presented for nonlinear uncertain systems. In these methods, wavelet kernel SVM is employed to establish the adaptive controller and an on-line learning rule for the weighting vector and bias is derived. The H∞ control technique is combined with adaptive control algorithm and wavelet support vector machine to achieve the desired attenuation on the tracking error caused by wavelet-SVM approximation error and external disturbances. The most important feature of the proposed control strategy is its inherent robustness and its ability to handle the nonlinear behaviour of the system. The results of simulation show this SVM online algorithm controller is very effective and the SVM controller can achieve a satisfactory performance.

Switched systems are an important class of hybrid systems. In recent years, such systems have drawn considerable attentions in control field. A switched fuzzy system is a switching system, for which all subsystems are fuzzy systems. This paper investigates the robust state estimation problem for a class of uncertain switched fuzzy systems with time-varying delays. By using appropriate switched Lyapunov functional approach, average dwell time scheme and filtering theory, delay dependent sufficient conditions for the solvability of this problem are stablished in terms of linear matrix inequalities (LMIs). An illustrative example is provided to show the effectiveness of the proposed theoretical results.

The main objectives of distribution network reconfiguration are real power loss reduction and relieving the overload. As well, integration of distributed generation (DG) units and fixed/switched capacitor banks are effective options for operation cost reduction, reducing system losses, improving voltage profile and increasing voltage stability index in the distribution systems. Therefore, simultaneous network reconfiguration, optimal placement of DG units and fixed/switched capacitor banks is proposed in this paper. The dedicated genetic algorithm is proposed for simultaneous network reconfiguration and allocation of DG units and fixed/switched capacitor banks in distribution systems considering different load levels. The proposed approach is able to find good network configurations and determining the optimal site and size of DG units and fixed/switched capacitor banks in distribution systems, simultaneously. The considered objective function encompasses the total cost of power loss, the investment and operation costs of DG units, the installation cost of fixed/switched capacitor banks and cost of purchased active power demand from upstream network. The proposed method is applied to 69-bus distribution system. The simulation results and comparison of different scenarios reveals that the simultaneous network reconfiguration and allocation of DG units and fixed/switched capacitor banks, results in significant advantages in the total cost.

Supporting multiple paths for concurrent data transmission is one of the most important features of the SCTP_CMT protocol. Generally, the priority of paths or best main path are not considered in this protocol; only the priority of data and communication currents are taken into account. Thus, choosing the best main path is very important. In this study, a method is introduced for the selection of path with the highest quality of service based on the features of all the lines. Prolonged data transmission time caused by failure to select optimal main path of the SCTP_CMT in dissimilar lines was reduced through the particle swarm optimization algorithm. The results of simulation in OMNET++ and MATLAB environment showed that the data transmission speed to select the main path in the proposed algorithm was decreased.

In this article we have introduced new techniques for generating compound images, or mosaics, which include reconstruction of main images using homomorphous square patches. These patches are extracted from another image called Key Image and are sorted next to each other in order to produce the image. To perform that, criterions like color level average, variance and histograms are used due to their important role in describing the façade of an image. PCA is a valuable statistical method in pattern recognition among the data; that can be used an images as well. These methods basically help us separate and analyze image parameters. PCA implementation results will be used as filters for the main image leading to extraction of constructor lines and different parts of image construction texture. These extracted parameters from the main and key image can be a local or global reference to select parts in order to produce the mosaic. The results of experiments has been shown that our method base on PCA is better than other methods.