TABRIZ JOURNAL OF ELECTRICAL ENGINEERING
Year:2016 | Volume:46 | Issue:2 (76)|Papers Count:30

This paper describes a method named hierarchical unsupervised segmentation using lossy data compression for synthetic aperture radar (SAR) images in which super pixels are used instead of pixels. In this paper, using a combination of several features such as edge, texture, and intensity, the super pixels are merged where the merging procedure is done by two steps. Since, many super pixel-based methods suffer from long run-times, the two step-proposed algorithm beside improving the accuracy, reduces the running-time. The first step is merging super pixels which there are not discriminative and clear edges between them. The merging of super pixels is done in the second step, if the length of data codes is minimized under a distinct distortion. The proposed algorithm is performed on SAR images and the simulation results demonstrate that the algorithm has a good accuracy and high processing speed.

In this paper a transimpedance amplifier in 0.18mm CMOS technology for optical communication networks is proposed. In this design an RGC preamplifier is used as input. The RGC structure is applied to input in order to increase trans conductance and reduce input impedance. This structure neutralizes the effect of the photodiode parasitic capacitors at input of transimpedance amplifier due to low input impedance. The circuit uses capacitive degeneration and broadband matching network techniques and is designed to have bandwidth of 20 GHz, transimpedance gain of 60dBW and input-referred noise current with spectral density of 12 pA/sqrtHz in the amplifier frequency band. Also results including input referred noise and transimpedance gain was confirmed after layout implementation in cadence, where post layout simulation was performed considering parasitic capacitors.

This paper deals with the problem of finite-time stochastic stability and stabilization of networked control systems in the presence of random delay. At first, the network random delay is modeled as a Markov chain, and the resulting closed-loop system is a Markovian jump linear system. Since, due to network complexities, the accurate access to transition probabilities are hard or even impossible, some of the elements in the transition probability matrix are considered as unknown. Then, according to the definition of finite-time stochastic stability for discrete-time Markovian jump systems, a sufficient condition is proposed to guaranty the boundedness of the system states in the sense of mean-square, in a determined time interval. In the next step, the results are extended for the finite-time stochastic stabilization problem of this specific class of systems and the output feedback controller is designed such that the closed-loop system is stochastically finite-time stable. All the results are presented in the form of new linear matrix inequalities (LMIs). The main contribution of this paper is utilizing the static output feedback controller for finite-time stochastic stabilization of the closed-loop system, besides, new LMIs are employed to calculate the output feedback control law. Also, in order to verify the theoretical results and show the applicability of the proposed controller, simulations are performed for two systems.

Existence of design faults is a source of failure in the system. Detection and removal of faults in design phase prohibits increase of cost and time of system development. Formal methods are used for proving behavioral properties of systems. Formal methods can be used in crux design of computer games to ensure absence of deadlock, guarantee of answer's existence and to determine complexity of crux. Hierarchical colored Petri net is a formal method that can be used to evaluate cruxes of computer games such as path planning puzzle games. Unblock me game is modeled as a case study and automatic analysis of its two cruxes are studied in this paper. Modeling of game is done hierarchically. New approaches for eliminating state space explosion problem and decreasing running time of model is presented in this paper. Modeling is done with aim of automatic state space analysis of the system and required functions for proving behavioral properties are implemented. This paper is presented a method for modeling and proving behavioral properties of puzzle type computer games using hierarchical colored Petri net that and can be generalized to similar games.

Taguchi's method is an effective multi-variable optimization method based on orthogonal array concept which is offers a systematic and efficient optimization process. In this paper, an improved Taguchi's method for shape optimization of electromagnetic devices is developed and applied to torque profile optimization of Double Stator Switched Reluctance Motor (DSSRM) using Finite Element Analysis (FEA). The proposed method determines parameters which are dependent or have nonlinear effect on the system operation, and uses this information in the optimization process to increase the precision of Taguchi method. This research uses the response table efficiently to obtain the optimum parameter levels. Simulation results show the efficiency and accuracy of the proposed method for optimization of multi variable problems.

The main objective of this paper is to employ predictive control strategy with the approach of active and reactive power control to improve power system stability in presence of wind farms based on Doubly Fed Induction Generator (DFIG). In this regard, the Super Capacitor Energy Storage System (SCESS) is used to control active power in the Grid Side Convertor (GSC) as well as balance the dc-link voltage and the Static Synchronous Series Compensator (SSSC) is employed to control power flow and reduce low frequency oscillations in the line leading to the infinite bus. The proposed strategy based on the predictive control is functional and can be simultaneously used to control the active and reactive power of the Rotor Side Convertor (RSC) as well as design the damping controller for SCESS and SSSC. In order to reduce computational complexity in selecting input paths, the Laguerre functions are used. Moreover, the exponential data weighting are used to reduce sampling time in the prediction horizon. Simulation results for the functional model predictive control is compared with the other two methods, model predictive control and PI controller and its superiority in improving power system stability is confirmed.

Clustering is one of the most widely used methods in data analysis. In the classical clustering methods it is assumed that all the features of the samples in a given data set make equal contribution when constructing the optimal clusters. However, in real world datasets, some of the features can exhibit higher relevance than others. Thus, the features with higher relevance are more important to form optimal result than those with lower relevance. To address this issue, in this paper we proposed a feature weighted fuzzy clustering method called RLWFCM. The proposed algorithm has three main advantages. The first advantage is that our method assigns different weights to each feature in each cluster. This is due to the fact that each feature may have different importance in different clusters. The second advantage is using a specific kernelized distance to reduce the noise-sensitivity of the algorithm. The third one is using an analytical method for adapting feature weights during training process. Moreover, in this paper the mathematical analysis to proof the convergence of the algorithm has been presented. Several experiments were performed on an artificial dataset and also five real world datasets and the results show that the proposed method outperformed the other feature weighted clustering methods.

Recent development of power electronics introduces the use of FACTS devices in power systems. The novel studies have shown that the presence of these devices, like UPFC, can enhance the interaction of system modes. In this paper, for the first time, the nonlinear interactions among different controllers of UPFC in a SMIB power system are analyzed, using the Modal Series method. To do so, a new and general method has proposed to obtain the Jacobian and Hessian matrices in closed form for power systems with UPFC. In addition, the effect of the damping coefficient and fault clearing time on interaction phenomenon examined. In order to analyzing the accuracy of the solutions obtained by different methods, a proximity measure is computed.

A data warehouse stores a large amount of data, which are usually used in decision support systems. Response time of these systems is too high because of their huge data. Since these systems are generally used by organization’s supervisors, reducing this response time is important. One of the major solutions for this problem is view materialization. Materialization of all views is impossible according to the constraint on memory space and the cost of maintenance these views. So, it is needed to select proper views to be materialized. Selection of these views is a kind of searching in a huge space that is considered as NP hard problem. Several methods are proposed to address this problem until now. Evolutionary algorithms are mostly used in solving MV problems. In this paper, Hybrid Cultural algorithm is used to select N top views among all views. Experiments show that this proposed algorithm has lower cost and higher speed than genetic algorithm, cuckoo search algorithm and Differential algorithm.

In this paper, simultaneous allocation of multi-distributed generation, capacitor bank and sectionaliser are utilized to improve efficiency of the distribution network. After placement of devices based on constant load model, technical performance of the system is evaluated under loads sensitive to voltage and frequency. Wind turbine and photovoltaic are renewable distributed generation units that along other devices are been studied. The proposed multi-objective functions are composed of reduction active and reactive power loss, improvement of voltage index and reliability indices. A hybrid optimization algorithm using Combination of firefly algorithm and particle swarm optimization is proposed to solve the multi-objective optimal allocation problem. Moreover, a method based on fuzzy mechanism is employed to extract one of the Pareto-optimal solutions as the best compromise one. The proposed method is implemented on a IEEE 69-bus distribution network and also actual 100-bus distribution system in Khoy-Iran. The results approve the effectiveness of the proposed method with simultaneous placement of different devices for improving technical performance of distribution network.

Soft lithography fabrication methods have been used enormously by MEMS researchers in recent years, due to their low cost and simplicity and batch fabrication feasibility. In this paper, a low cost and useful technique to fabricate Microstructures on a polymeric substrate by hot embossing is presented. The used master has been fabricated from SU-8 polymer on a glass substrate. The main parameters of hot embossing quality such as temperature and pressure are discussed and the optimum values are determined. The hydrophobic surface of PMMA substrate is modified to hydrophilic in order to be used in Microfluidic devises. As a good approach, Micro channels by minimum feature size of 15μm and depth of 50μm are implemented in this paper. Major advantages of this method are simplicity and low cost, because of not-complicated equipment and inexpensive materials for substrate.

The main advantage of employing stochastic programming compare to its deterministic counterpart in wind integrated thermal Unit Commitment (UC) problem is considering probable occurrence wind power and load scenarios by minimizing the expectation of operation cost. However, the major drawback of representing cost function by its expectation is ignoring the other important parameters of cost probability distribution function. One of the best methods to overcome this problem and at the same time maintaining the stochastic structure of the problem is the application of the concept of risk. In this context, in this paper the effect of wind resources ans load uncertainties on operator risk-taking in unit commitment problem will be analyzed by using the main risk measure namely Conditional Value at Risk (CVaR). The proposed model is Mixed Integer Linear Programming (MILP) which has been modified in the form of stochastic programming and containing risk measure terms. The wind power and load uncertainties has been considered in two ways including the consideration of wind penetration as a fraction of a basic value and also consideration of wind power and load prediction errors in the form of normal probability distribution with different variances.

The development of high power grid-connected photovoltaic (PV) systems and their adaptability with grid conditions is a crucial issue. In fault conditions, it can be considered with two sides, first being connected with the grid due to economical problems, second minimizing damages into the grid due to technical problems. The improvement of grid-connected PV systems performance under grid voltage sag (which is the most probable faults) is proposed in this paper. In point of common coupling in case of voltage sag, the active power will oscillating and the grid synchronization system would not work properly, which both are detrimental for unbalanced grid. In order to improve the performance of the exchanged power between the converter and the unbalanced grid, method of Positive-Negative-Sequence Control (PNSC: able to remove oscillation of active power) is suggested. This method is also effective to reduce the oscillations in DC voltage that can be detrimental for DC-link capacitor. In order to track the desired unbalanced reference current generated by PNSC, a proportional–resonant (PR) current controller has been designed. Furthermore, instead of Synchronous Reference Frame Phase Locked Loop (SRF-PLL) which suffers the distorted response in polluted grid, a Double Second-Order Generalized Integrator-Frequency Locked Loop (DSOGI-FLL) is used to synchronize inverter with the grid. Finally, to validate the proposed approach, the whole system is simulated via MatLab/Simulinkâ and simulation results have been compared with conventional method. Results verify the grid condition has improved in comparison with implementing the conventional method.

Without capacity mechanisms, generation investment in the deregulated power system tends towards boom and bust cycles. Due to lack of significant demand response to electricity price, the cycles may jeopardize supply adequacy or even leads to a black-out in the bust periods. In this paper, encouragement of investment in distributed generations, which is designed by the regulatory body is proposed to reduce the generation investment cycles. The traditional long-term model of generation investment dynamic is modified to model and consider the regulatory policies. Moreover, an optimization model of regulatory policies is also proposed in order to preserve the supply adequacy in power system. Particle swarm optimization algorithm is employed to solve the optimization problem. The proposed model is implemented in MATLAB software. The simulation results indicate the efficiency of the proposed model in improving the boom and bust cycles and preserving supply adequacy.

Infrared target detection is a crucial step in many fields of infrared image and video processing. It plays an important role in military and surveillance applications. The aerial infrared target region usually has a small size and is embedded in a cluttered background. Therefore, very limited shape information and low signal to noise ratio (SNR) are the typical properties of aerial infrared targets. For the detection step, the paper idea is based on enhancing the dim small targets by applying Kim’s LoG filter and modifying New Top Hat transformation (NTH). Therefore, it improves the contrast ratio between the target and background region. Experimental results exhibit the ability of the new algorithm for application of aerial infrared target detection. The evaluated performances demonstrate that the suggested algorithm is more efficient compared with some widely used methods.

This paper presents a new approach for estimating the signal-to-noise ratio (SNR) of mixture signal, which is based on the computational auditory scene analysis (CASA). The ideal binary mask (IBM) which is generally the computational goal of CASA based systems is used to estimate the SNR of noisy speech signal. The proposed method is evaluated using IBM and some quasi-IBM masks. The method is simple and computationally efficient. Systematic evaluations show that the proposed method results in reasonable estimation of the input SNR level in a wide range of SNR values, and probable errors in estimating IBM do not affect so much the performance of the proposed system. Furthermore, simulation results show that the proposed system outperforms previous SNR estimation methods.

This paper presents a fault location technique for transmission lines with minimum current measurement units. The proposed method used proper current ratios to find fault locations based on both The venin theory in faulty power networks and calculation of short circuit currents in each branch. These current ratios are extracted based on lowest sensitivity to The venin impedance variations of the network. Proposed algorithm compares current ratios in offline calculations with associated values collected by measurements with a lookup table mechanism. A similarity measure algorithm can find location of the fault including line number and the distance. A multi-objective optimization problem using Bees Algorithm (BA) is used to search monitoring points with minimum possible current measurements. Accuracy of the proposed algorithm is evaluated in a widely used multi-machine network of Western Systems Coordinating Council (WSCC).

In this paper, an improved reduced-order sliding mode controller with PI controller is discussed for positive output elementary super-lift Luo (POESLL) converter operated in continuous conduction mode. Because of successive switching operation, the dynamic of this converter is nonlinear severely. The proposed controller improves the dynamic and static performance of the converter. The main advantage of the proposed controller compared with other controllers is the direct control and regulation of the output voltage. The performance of this controller is verified for various operating points in MATLAB software. In addition, a laboratory setup is made. Experimental results show the controller has an excellent performance against variations of the operating point.

In this paper, a new form of signal flow graph technique is used for modeling a bidirectional buck-boost converter and extracting the needed transfer functions. The signal flow graph method has a more simple process in comparison with other modeling techniques. The modeled converter is employed in energy management system of an electric vehicle. The energy system of the electric vehicle is composed of two energy sources, ultra-capacitor and battery pack. The proposed hybrid source has the advantages like: high energy density transmission and high instantaneous power output on the ultra-capacitor. Here, by using the mentioned converter, the energy management between the two sources is performed. Furthermore, the needed configuration for energy management system is presented and a proper controller based on obtained model for energy management in hybrid sources is proposed. Finally, the simulation results are brought to verify the precision of the proposed modeling technique and effectiveness of it in energy management system of an electric vehicle.

In recent years, reduced switch count converters have garnered particular attention because they have some benefits such as low cost low weight and small size. So far, different structures of power converters have been introduced to reduce the number of switches that can be classified in two categories, single and multiple outputs. In this paper, three structures of reduced switches count converters are suggested to improve the power quality as inter line unified power quality conditioner (IUPQC). The switch count reduction methods reduce the complexity of the proposed structures, the total price of the system and switching losses. Control and modulation schemes are provided to guarantee the compensation of disturbances arising from network and load and obtain the unity power factor, too. The simulation results verify suitable effects of proposed control and modulation schemes for proposed converters.

The unit commitment problem is a complex nonlinear optimization problem with large dimensions and linear and nonlinear constraints for which no approach has been yet able to extract the global optimum solution. In this paper, a new heuristic approach is represented which comprises two major stages: 1- Intelligent generation of initial population, 2- The bat search algorithm integrated with repair procedures. In this approach, the initial solutions are generated using a priority list and a new repair procedure for the violations related to minimum up/down time constraint. The solutions generated in this stage are infect initial solutions for the bat intelligent algorithm which performs a local search. To establish spinning reserve and minimum up/down time constraints, new procedures have been designed, which play important roles in the efficiency of the search and decreasing computational times. The proposed approach has been implemented on practical test systems of IEEE 54-unit, the Taiwan power system (Taipower), and the standard 100-unit test system. The results reveal the approach’s success especially in achieving the lowest production cost in the literature for the first two systems.

Regarding the vast extension of information technology, communication systems and electronic instrumentation, the employment of phasor measurement units (PMUs) has been proposed for power system observability and wide-area protection. Nevertheless, the construction of communication channels and installation of PMU equipment are highly expensive and require optimizations. In this paper, first the constraints are proposed for the optimal PMU placement problem considering fault state observability of power system in presence of zero bus injection, and afterwards the mathematical formulation is presented. For more realistic modeling of power network, line outages have been considered. Finally, the proposed method is implemented on standard IEEE 9, 14, and 30 bus networks, and the enhanced results are represented.

The most important security requirements for the microgrid connected to the network are islanding detection and protection for anti-islanding. In this paper, a new method is presented for islanding detection in a standard microgrid that equipped with wind turbine with double feed induction generator (DFIG), combined heat and power (CHP) and photovoltaic cells (PV) systems. In This method, all control systems of DFIG, CHP and PV cells are modeled. The PV control system is worked for maximum power point tracking (MPPT) of PV system. In this paper, a new method for islanding detection method based on disturbance current injection into the system through the quadrate axis (q-axis) of the DFIG grid side converter is presented. Frequency variation of the wind turbine stator voltage is selected as an index for islanding detection. In this method, wind turbine stator voltage that is index of islanding detection, is fed to a logic circuit and then the frequency of system is evaluated and if frequency violate from threshold (59.9-60.1) Hz, this condition is detected as islanding condition. The proposed method is able to detect islanding condition from transient condition. This capability leads to significant reduction of non detection zone (NDZ) areas. Simulation results using PSCAD/EMTDC software show the effectiveness of the proposed method in islanding detection in a standard microgrid.

This paper proposes a new traveling wave-based fault location method for power transmission grids. The proposed method involves an online and an offline stage. In the offline stage, the transmission network is represented by a weighted graph. The shortest paths between any two sensors are determined. The branches of the graph that are not on at least one of the shortest path, are divided into virtual lines associated with the location of the sensors. In the online stage, according to first arrival time at different sensors, the points belonging to the shortest paths obtained during the offline stage with the possibility of having a fault are identified Then using an evaluation index, the exact point of fault and the fault inception time are determined. The proposed method is independent of the fault type and can be utilized in transmission network with any size. Simulation studies carried out using EMTP software and MATLAB, verify high accuracy of the proposed method.

In this paper, two new hybrid controllers and a new control approach are presented to control the injected voltage of the DVR. In the initial control stages, the improved fuzzy type-2 principles are used to estimate the nonlinearities online. Thus, using this type of controller, the applied disturbances and the nonlinearities are modeled and compensated. Moreover, the fuzzy type-2 controller is designed in a manner so that it adapts its structure to the environments conditions which decreases the calculation burden. In the second phase, to decrease the calculation burden furthermore, a new controller replaces the fuzzy controller. This controller, by combining principles of variable structure control with conventional linear control, apart from shaping the system dynamics, will result in a soft and smooth operation. Moreover, in the new approach, the load current is used to control the load voltage, leading to reduction of burden on protective devices. The system stability is also confirmed using lyapunov’s stability criterion in compensating all symmetrical and unsymmetrical faults. Simulation results confirm the applicability of this controller using this approach.

This paper presents a new method for distribution system planning considering load uncertainty that the obtained network is robust to the load forecasting uncertainty. At first a method for optimal network planning ignoring load uncertainty has been presented and then by mixing this method with Monte Carlo Method, a novel method has been presented for planning distribution system considering load uncertainty. Supposing that the location and size of HV substations are obtained in another work, the optimal route of MV feeders are obtained through the new Imperialist Competitive Algorithm (ICA) developed for the optimal expansion planning of distribution network. In order to check the radial structure of the network obtained in any iteration of ICA, a novel mathematical algorithm is employed. For the scenario with considering load uncertainty, optimal network and the histogram figures for some electrical parameters of the optimal network are obtained. According to the results, the obtained optimal network is robust to the load uncertainty.

Growing demand to access frequency spectrum, cognitive radio is introduced so that the secondary users (SU) are allowed to use unused spectrum of primary user. If the primary user (PU) uses the full frequency spectrum, the existing unused spectrum in space-domain can create ideal opportunistic interference alignment considering cognitive networks. Since the PU knows perfect channel knowledge, it can perform water-filling algorithm and singular value decomposition, then secondary users transmit along the spatial direction with the design pre-coder. In this paper, we improve the efficiency of the system with a transmit space-time coding technique introduced by Alamouti. In this regard, the SU transmit and receive antennas are reduced and after performing interference alignment algorithm, but the interference does not becomes exactly zero. In fact, to remove this interference, SU transmits Alamouti code. In this paper, we consider the case of two secondary users opportunistically exploiting the same frequency band utilized by a licensed user. Our goal is to design low-complexity precoders to realize interference cancellation for each secondary user. The main idea is to design precoders such that the two users transmit over two orthogonal spaces. As a result, the decoders can project the received signals to each of the orthogonal spaces and decode the information of each user without any interference from the other user.

This paper proposes a statistical method for content-aware image shrinking. This method divides image into equal quads by an initial uniform grid and computes an important factor for each of quads guided by importance map. The importance map is generated automatically using a novel combination of image edge density and Harel saliency measurement. The method then computes scaling factor for each of quads by using of statistical measures such as the mean and the variance of quads importance factors. To preserve the important areas as far as possible and to destroy the unimportant areas more, this method uses statistical functions to give the larger value to important quads. To prevent bending of straight horizontal lines, in the vertical direction, the algorithm equally resizes quads that are placed in the same row. Also, to prevent bending of straight vertical lines, the algorithm equally squeezes and stretches quads that are placed in the same column. Experimental results of different types of images show that our method can achieve better objective and subjective quality with preserving prominent objects.

In this paper, a novel routing protocol based on the quality of service (QoS) in wireless sensor networks is proposed. In the proposed method, cluster heads are determined by the sink node and routing operation is done according to the QoS and energy efficiency criteria. Multi-criteria decision making algorithm along with the preferences table is used to uniformly distribute load and traffic the network. The method involves priority of QoS measures such as reliability, energy and delay to determine fitness of each route in a hierarchical manner. The proposed protocol is implemented in MATLAB simulation environment and its effectiveness is evaluated and compared with BERR protocol. Simulation results show that the proposed algorithm outperforms the BERR protocol by about 50% in energy consumption and 150% in the network life time in large scale networks. Since the proposed method tries to selects the shortest routes and uses the retransmission of lost packets, it can improve average of delay for about 30% in large scale networks and shows high reliability.

To calculate shielding failure of a lightning protection system, several analytical methods such as Electro-Geometric Model and Rolling Sphere Method have been proposed. These methods are used to determine protected and unprotected areas, but they can’t show the frequency of shielding failure to determine more risky areas. In this paper, a new numerical model is proposed by using Leader Progression Model and Artificial Bee Colony Algorithm to identify the most vulnerable area on complex structures. The progression of lightning downward leader, the inception and propagation of upward leaders are considered step by step until final jump. Structures, air terminals, downward and upward leaders are modeled by different shapes of charges, and the electric field in an environment is computed by charge simulation method. Lightning is assumed as artificial bee, and areas with high shielding failure are the best food sources with high nectar amount. So the result produced by ABC algorithm is the most lightning risky area. Finally, lightning performance of an asymmetric structure protection system is evaluated, and based on simulation results the best position of air terminals is determined. The proposed method can be used for protection system design of any complex structure.