TABRIZ JOURNAL OF ELECTRICAL ENGINEERING
Year:2020 | Volume:50 | Issue:1 (91)|Papers Count:40

Using a conservative stability analysis to design a time delay system controller usually cause a devaluation on the actuator performance. In addition, the results obtained considering deterministic stability criteria may not be reliable due to the existence of random delays. Whenever the feedback loop of a power system oscillation damping control system receives remote signals through communication networks, we face a POD system with random delays namely POD-RD system. In this paper, a new design approach to promote the performance of a POD-RD control system is proposed based on the delay scheduling control signal method. This is a two-step method for implementation. In the first step assuming the average value of the communication random delays, the controller's parameters and also delay parameter of the control signal are determined as initial values by minimizing sperctral abscissa of the closed loop system. In the next step, to include impact of the communication random delays, a repetitive procedure is proposed to determine the optimal value of the delay parameter of the control signal and also the controller's parameters. The objective of the optimization model is to minimize the spectral abscissa and the defined second order moment. The feasibility of the proposed method is evaluated by doing a number of simulations on the standard four-machine test system. The results reveal that the method can make a robust performance in the studied system.

In this paper, an adaptive continuous terminal sliding mode controller is proposed for a class of second-order time varying perturbed systems with unknown parameters. In the proposed method, the controller is a combination of continuous terminal sliding mode control and adaptive control. The convergence of the proposed method is based on a continuously-differentiable, homogeneous and strict Lyapunov function, and the unknown parameter of the system is estimated by an adaptive law. Zero-error convergence of the proposed method is attained with information about the output and its derivative; furthermore, in this method, perturbation is compensated and a continuous control signal is generated which reduces the chattering. The simulation results show proper performance of the proposed method.

This paper aims to present a novel and comprehensive model for actuator faults to address the problem of robust fault-tolerant controller design for networked control systems (NCSs) in the presence of phenomena such as random delays, model uncertainties, and actuator faults. For this purpose, firstly, the NCS has been appropriately modeled as discrete-time Markovian jump systems (MJSs) with partly-unknown transition probabilities. Then, the problem of mode-dependent static output feedback controller design has been studied not only as a convex optimization problem but also in the form of linear matrix inequalities (LMIs). Notably, the designed controller guarantees the stochastic stability of the closed-loop system in the presence of actuator faults and uncertainties. Finally, through numerical simulations, the theoretical results of this study are proved, and it has been shown that this method is more efficient and superior than other methods.

In this paper an ultra-low power two stage improved operational trans_conductance amplifier based on folded cascode is designed. The proposed operational trans_conductance amplifier operates in weak inversion region. The use of two folded branches in the signal amplification path for the first stage and the new feed_forward compensation path with low bias current on the second stage in this proposed amplifier increases the DC gain, unity gain frequency, slew rate and decreases the input referred noise. The simulation results in a TSMC 0. 18μ m CMOS technology it shows that the proposed operational trans_conductance amplifier has unity gain bandwidth of 117 KHz, and consumes 195 nW power from a 0. 6 V supply voltage with DC gain of 101. 4 dB.

The huge amount of data created constantly with increasing rate from different sources such as smart phones, social media, imaging technologies and etc. becomes difficult to be analyzed by conventional data analytic tools. For this reason a new field of research called Big Data Analytics is growing faster in the research and industrial communities. Clustering big datasets is one of the important challenges which attracts more and more attentions among researchers. In this paper first a method for non-automatic big data clustering (when the number of clusters is known) and then a two-stage method for big data automatic clustering (able in finding the number of clusters) based on grey wolf optimization algorithm are introduced. In the first stage the algorithm tries to find the number of clusters using a tree structure and in the second stage the main algorithm searches the solution space to find the position of centroids. The methodology is tested on 13 synthetics and 2 real big mobility datasets. The achieved results show its effectiveness in big data clustering.

This paper presents a fault tolerant approach for overactuated-oversensed systems subject to multiplicative and additive faults. The main idea is to use a reconciliatory unit in charge to discover online the possibly faulty physical sensors and exclude their measures from the generation of the ” virtual” sensors. In the proposed approach, the energetic based approach has been used to estimate the actuator multiplicative faults. In the next step, an unknown input observer has been coupled with the energetic based estimator to estimate sensor multiplicative and additive faults. Finally, the control allocation approach has been used to reach the desire output. Finally, a simulation example is reported in details to show the effectiveness of the scheme.

Cognitive radios are expected to opportunisticly communicate with covering a wide range of the frequency spectrum by continously sensing the spectrum and identifying available channels. In this paper, we introduced an inductorless and differential wideband LNA, which uses the complementary current-reuse structure and active and passive m G-enhancement technique, which leads to increase voltage gain, reduction in power consumption and noise, in the signal receiving path. Therefore, the proposed LNA is able to achieve 50 MHz – 1. 7 GHz bandwidth with 11 S less than-14. 26 dB, maximum voltage gain of 20. 2 dB, minimum NF of 3. 31 dB, while consuming only 1. 82 mA from a supply voltage of 1. 8 V to TSMC 0. 18 μ m CMOS technology. Also, simulations are performed using Spectre RF simulator.

A new SEPIC-based high step-up DC-DC converter is proposed in this paper. The proposed converter can work with high voltage gain in low duty cycle which causes to decrease the inductors current ripple, the conduction losses and the voltage stress across the semiconductors. Since the proposed converter is based on the SEPIC topology, it inherits the SEPIC’ s advantage of continuous input current which can be significantly important in renewable energy systems. The voltage gain of the proposed converter is more than the classic SEPIC converter and its structure and control system are simplistic. In this paper, steady state analysis of the converter in CCM and DCM mode is provided. The experimental results show the suitable performance of the proposed converter.

In this paper a method of blind recognition of the coding parameters for binary turbo product codes is proposed. The only knowledge is that the stream is encoded using binary product codes, while the coding parameters are unknown. The recognition processing includes two major procedures: code word length estimation and generator polynomial reconstruction. A hard decision method has been proposed in a previous literature. In this paper we propose the recognition approach in soft decisions situations with Binary-Phase-Shift-Key (BPSK) modulations and Additive-White-Gaussian-Noise (AWGN) channels. The code word length is estimated by maximizing the root information dispersion entropy function (RIDEF) for two dimension of the bit stream. And then we search for the code roots to reconstruct the primitive and generator polynomials. By utilizing the soft output of the channel, the recognition performance is improved and simulations show the efficiency of the proposed algorithm.

The semantic web technology and decision making based on the linked data is progressing every day. The linked data are managed as decentralized sources, and their quality is a serious concern. The assessment of the quality of linked data is a key to adopting them in different fields because each data set has been developed by a different group, using various methods and tools. The qualitative and quantitative diversity of such data is higher than those generated by official organizations and firms. In this paper, we first overview and evaluate the dimensions and measures for the quality assessment of data especially linked data. Then, we present a novel framework as a solution for improving linked data quality assessment and data fusion. The good quality data make good result in data fusion. Finally, we introduce six rules for handling data conflicts and a new metric for assessment of granularity level of data (GLQM) and adopt several tools to assess the quality of data using the proposed framework.

In this paper a time-to-digital converter, based on parallel vernier ring oscillator is presented. The delay elements have been applied in parallel form to two ring oscillators with different frequencies. The delay difference of the time-to-digital converter stages’ can be less than an inverter delay’ because of using parallel elements. simultaneous use of parallel delay elements and vernier method, The proposed time-to-digital converter can be achieved high resolution. The signal gating method is used to reduce power consumption. In this method, when the lag signal reaches the lead one, by determining the digital output code, the control signal is activated and stops the fluctuations in two ring oscillators, thus it prevents power dissipation. A typical 6 bit time-to-digital converter with the proposed method is simulated in 65nm standard CMOS technology. 1ps resolution, 382uW average power consumption, 2269um2 chip area and 600Ms/s sampling rate is obtained under 1V power supply.

As CMOS transistors are scaling down to improve performance, vulnerability of digital logic circuits to soft errors caused by energetic particles are increasing. Flip-flops (FF) are of the main elements of sequential logic circuits that are very susceptive to single event upset (SEU) and single event multiple-node upset (SEMU). In this paper, a new FF circuit robust against transient faults of SEU and SEMU caused by high-energy particle strikes is designed and evaluated. In comparison with previous work, the proposed circuit offers a low design cost, while, it also has a high degree of robustness against SEU/SEMU. This achievement is discussed and also evaluated by the simulations carried-out. Simulation results reveal that, the proposed circuit offers 20% improvement in power consumption and also 31% in delay as compared to the well-known and widely used MS-DICE FF. The effects of process, voltage, and temperature (PVT) variations on the performance of the proposed FF are also investigated and it is shown that this circuit has a reliable operation in the presence of PVT variations as well.

Visual domain adaptation aims to learn robust models for the test data by knowledge transferring from a training data where the training and test sets are from different distributions. Existing approaches attempt to solve domain shift problem with either adaptation across domains or performing low-rank constraints. In this paper, we propose a two-phases unsupervised approach referred as image reconstruction Error minimization via Distribution Adaptation and low-rank constraint (EDA), which benefits from both the distribution adaptation and the low-rank constraints to tackle distribution mismatch across domains. In the first phase, our proposed approach projects the training and test data onto a common subspace in which the marginal and conditional distribution differences of domains are minimized. Moreover, EDA benefits from domain invariant clustering to discriminate between various classes of data. In the second phase, for preserving data structure in the shared subspace, EDA minimizes the data reconstruction error using low-rank and sparse constraints. Overall, EDA solves the domain mismatch problem in cubic time complexity. The proposed approach is evaluated on variety of visual benchmark datasets and its performance is compared with the other state-of-the-art domain adaptation methods. The average accuracy of EDA on 32 experiments is determined 68. 33% where outperforms other state-of-the-art domain adaptation methods with 4. 28% improvement.

Network coding is an effective approach for improving the performance of wireless networks, such that a new generation of routing protocols, known as network coding-aware protocols, recognize coding opportunities during their route discovery process, and try to establish routes containing appropriate coding opportunities. Most of the research works being done around this type of routing, only try to maximize the number of coding opportunities along data transmission route. However, merely considering coding opportunities, may decrease the optimality and the throughput of routes, since sometimes, by considering only the number of coding opportunities without paying attention to the quality of links along routes, the quality of detected routes will be decreased. In this paper, in order to resolve the deficiencies of previously related works, a new network coding-aware routing protocol based on fuzzy logic is proposed, such that the delay of links besides the satus of all nodes along different paths are evaluated through utilizing fuzzy logic, and the overall value of all paths are determined. Finally, paths wirh better status in terms of the quality of links and nodes, are selected for data transmission. Simulation results demonstrate that the proposed protocol has better performance than previous approaches, in terms of throughput, packet delivery ratio, and delay.

Botnets are known to be among the most popular malwares in cyber criminals for their practicality in carrying many cybercrimes as reported in the recent news. While many detection schemes have been developed, botnets remain the most powerful attack platform by constantly and continuously adopting new techniques and strategies. Thus, early identification and timely detection of botnets can take an effective step towards making perfect defense system. Most of existing botnet detection methods cannot detect botnets in real-time and in an early stage of their lifecycle before participating in a cyber-crime. In this work, we propose a novel approach to detect the BlackEnergy botnet traffic using Hidden Markov Model (HMM) within flow Intervals. In BlackEnergy, bots are controlled by attackers under a HTTP base command and control (C&C) infrastructure. First we analysis BlackEnergy’ s network traffic and extract its main features and network behavior patterns. Then we adapt the proposed HMM model with BlackEnergy botnet patterns and features. In addition to detecting the botnet communication traffic in both Attack and C&C stages, inferred HMM defines the stage of botnet lifecycle. Our proposed method detects botnet activity in small time intervals without having seen a complete network flow. Using existing datasets, we show experimentally that it is possible to identify the presence of botnets activity with high accuracy even in very small time windows.

Increasing harmonic contamination is one of the concerns in power systems that has been more considered. On the other hand, the high cost of compensation can eliminate the possibility of compensation across the network. Regarding the restructuring of power systems, it is possible to provide the compensation cost by contaminating subscribers. In this paper, a new method for controlling the network harmonic level is presented using harmonic pricing. Determining the fair harmonic contribution at the point of common coupling voltage is the base of pricing method. In addition, a new index has been proposed to determine the harmonic contribution of the customer and utility to the harmonic contamination of the point of common coupling. By using this index, an accurate impression of the customer and utility contribution in all of the dominant harmonic levels is obtained. The simulation results on the IEEE 14-bus standard network illustrate the effectiveness of the proposed method for pricing based on fair contribution of customer and network. Also, the results show that by justly determining the harmonic contribution, the customers can be encouraged to reduce or control their harmonic level. Otherwise, by aggregating the penalty fees, harmonic filters can be installed at the desired location to control the network harmonic control.

In this paper, single-phase dynamic comparator was improved based on adding a clocked NMOS auxiliary latch at its output and two clocked NMOS transistors at its internal nodes. Using clocked NMOS latch and transistors increased the speed of comparator, without increasing the loading effect and kick-back noise on its previous and next stages. The advantage of the proposed comparator rather than the conventional comparator was higher speed with precence same power dissipations. To demonstrate the mention specifications, proposed and conventional comparators were analyzed and simulated at 2-GS/s sampling rate in 0. 18-μ m CMOS process. In this case, the outputs of proposed comparator were determined almost 18-ps (%9) faster than the outputs of conventional comparator. Also, to compare the performances of proposed and conventional comparators in flash ADCs, two 4-bits flash ADCs, using proposed and conventional comparators in their comparison chains blocks, were designed and simulated at 2-GS/s. In this case, the FOM of ADCs with the proposed and conventional comparators at nyquist rate input frequency achieved 0. 61-pJ/conv. step and 0. 72-pJ/conv. step, respectively. Corresponding, the ENOB of ADCs achieved 3. 74-Bit and 3. 45-Bit, respectively. In addition, the power dissipations of comparators array of ADCs with proposed and conventional comparators were 4. 23-mW and 4. 09-mW, respectively. Also, the power dissipations of two flash ADCs, without the power dissipations of their comparators array, were 11. 70-mW and 12. 03-mW, respectively.

Quantum dot cellular automata (QCA) as an important technology with minimal size, high speed, low latency and power consumption is suitable replacement for semiconductor transistor technology. The growing demand for observability and testability attracts more research on it. A full adder circuit is a basic unit in digital arithmetic and logic circuits. In this paper, a unique structure for testable full adder is presented in QCA. The implementation of the full adder circuit with the structure of the Built In-Self Test (BIST), its observational capabilities and its intermediate component controllability increase the reliability and the test capability of other designed circuits based on this. All of inputs in the proposed have testability and controllability capability and the middle nodes have observability feature. This design in contrast to its counterparts uses three-layer scheme and surpasses the best previous layer designs in terms of area, delay and complexity. The simulation results using QCADesigner software confirmed that the presented circuit works well and can be used as a high-performance design in QCA technology.

Domain adaptation can transfer knowledge from a training set (source domain) to a test set (target domain), promoting the performance of the model learned from the training set. In addition, sparse coding makes the learned model more succinct and easy to manipulate. However, the existence of the distribution mismatch across the source and target domains reduce the performance of model. In this paper, we propose an unsupervised domain adaptation model to minimize the prediction error of image classification. Sample reweighting is utilized to handle redundant and useless information of source data in the new representation. Moreover, the difference of the conditional distributions across the source and target domains is reduced along with the subspace alignment. Our proposed approach learns a sparse domain-invariant classifier in a latent subspace with preserving the structure of the input data. Extensive experiments demonstrate that our proposed approach shows 4. 49% improvement in classification accuracy on real-world datasets compared to state-of-the-art machine learning and domain adaptation methods.

Voltage unbalance reduces the power quality and causes problems in the customer equipment and in the grids operation, especially in grids including distributed generation sources. In microgrids, use of distributed generation sources controlled by the inverter, has made it possible to compensate voltage unbalance by distributed generation units. To compensate voltage unbalance, usually two control loops are used to control the positive and negative components in the dq rotary frame. Nevertheless, the control loops of the positive and negative sequences are not completely decoupled from each other and there is an oscillatory coupling between them. In this paper, the use of the decoupled double synchronous reference frame in order to eliminate oscillating couplings and for voltage unbalance compensation is proposed. The simulation results show that, using the proposed control method, the voltage unbalance at the point of common coupling is effectively reduced. Also, due to the decoupling of the positive sequence from the negative sequence control loop, the performance of the distributed generation unit in the power injection to the grid is not affected by the voltage unbalance control system.

In this paper, the leader-following group consensus of discrete-time fractional-order multi-agent systems with time-delay is studied. At First, two groups are considered. Then, the problem is extended to arbitrary number of groups. The agents are considered as discrete-time fractional order integrators with input time delay. The interaction between agents is described with a directed communication graph with fixed topology. The group consensus problem for the considered agents leads to asymptotic stability analysis of a discrete-time fractional order system with time-delay. Based on this idea, the necessary and sufficient condition to reach the leader-following group consensus in terms of the controller gains of agents is extracted. Moreover, the optimal value of the controller gains is calculated to minimize a special performance index. Numerical simulations show the performance of the proposed method.

Nowadays, speech enhancement has become one of the most important issues in signal processing. Noise reduction such that it does not disturb the original signal is an important challenge in speech enhancement. In this paper, we have proposed a new hybrid two-stage method for speech enhancement. In the proposed method, noisy speech signal is enhanced using perceptuallymotivated Bayesian approach in the first stage. Then, the signal is decomposed into sub-bands using wavelet packet decomposition. In the second stage, each sub-band signal is enhanced using NNESE method. Hyperparameters of NNESE is optimized using QPSO metaheuristic. Finally, all enhanced sub-band signals are combined together using wavelet packet reconstruction. Proposed method is better than others in terms of PESQ and Segmental SNR criteria over a subset of TIMIT speech database polluted by Noisex-92 noise database.

In this paper, a new method for designing autopilot for an air vehicle with polynomial nonlinear model is presented. Moreover the method is developed to fault tolerant control autopilot using combined projection and redesign methods. The nominal autopilot has been designed based on regulator controller for the polynomial nonlinear model and sum of square optimization theory. Previous theoretical methods have been extended from the regulator type to the tracking. Stability of the closed loop system is guaranteed by introducing a polynomial Lyapunov function that is derived using sum of squares optimization method. Then, the feasibility problem has been changed to the optimization problem. The proposed method has been used on an air vehicle supersonic with nonlinear polynomial model. The results showed proper operation in normal and faulty situations.

In this paper, using a hybrid control method including an active control method (circular chain) and an inactive control method (droop control), the contribution of each interlinking converter (IC) to the power transmitted between AC and DC subnetworks in islanded mode is determined. The values of real power in AC and DC sub-grids correspond to the frequency of AC sub-grid and the voltage of the DC sub-grid, respectively. Therefore, using this data and defining a droop relation, the value of initial power transmitted between AC and DC subnetworks is determined. Then, the active control system modifies the transmitted power considering the data received from the previous converter. The use of power parameter in the active control method decreases the bandwidth of transmission lines between the converters. On the other hand, two-stage control for determining the value of transmitted power retains the stability of the system in case of a disruption in transmission lines. However, the circular chain control system presented in this paper will continue to operate in case of disruption in the link between two adjacent ICs. The results of simulation in MATLAB confirm the credibility of the mentioned arguments.

Wide area power oscillation damping (WAPOD) is a very effective controller to damp out the inter-area electromechanical oscillations in the wide area power systems. This controller requires communication networks to receive the remote outputs. The use of communication networks causes delays in feedback loops. This delay reduces the controller's performance. Therefore, in order to improve the performance of the control system in the presence of delay, in this paper a WAPOD controller is designed using delay scheduling for the TCSC supplemental control. To select control structure, a method is proposed that allows optimization of stability in the range of delay. To design the controller, the the real part of the rightmost electromechanical mode is minimized. Case studies is based on a four-machine power system. Effectiveness of the proposed WAPOD controller is verified by simulation studies and compared with a conventional delayed design approach.

In image processing many algorithms get disturbed because of shadow. These problems can be avoided if the location of shadows is clear. This article presents a new optimized algorithm for extracting shadows from a single color aerial image. Most of similar works do this by using a suite color space. In this work two RGB and YCbCr color spaces are combined and a powerful ratio map has been created. Furthermore, in a new method, effect of sky blue color is determined that improves the ratio map. Candidate shadow and non-shadow regions are separated by applying Otsu’ s thresholding method. Because of ratio map performance it is not necessary to analyze region by region and leads to decrease computation cost. So it can be used in online application. The experimental results demonstrate the advantage of the proposed algorithm.

The trust inference problem in a social network is defined as anticipating the trust level that a user can have to another user who is not directly connected to him on the trust network. This research aims to propose a method for trust inference in a trust network. Previous research studies are mainly limited to one type of trust network, and they cannot be used for different trust networks with different values of trust. In this research, soft computing and neural network model are used to predict trust values. To train the neural network system, genetic algorithm is exploited. One of the main advantages of the proposed method is that, unlike previous methods, it is not limited to one type of trust network, and it can also be used for trust networks with different values of trust. In the proposed method, at first four proposed features are extracted from the trust network, and afterward, the proposed neural network system is trained using these features as well as the genetic algorithm. The proposed method is implemented on the standard trust network and is compared with other similar methods. Experimental results indicate that the proposed method is able to produce more accurate results in comparison with previous methods.

In recent years, design algorithms have been introduced based on X parameters. One of these algorithms determines the load impedance that maximizes the output power. This algorithm uses an iterative process. The weighting factor in this algorithm determines the convergence rate of this algorithm. By choosing this weighting factor unsuitably, the algorithm may be slow down or become unstable. In this paper, a reduction weighted method and an adaptive weighing method are proposed to accelerate the convergence and prevent instability. By using these methods in different amplifiers, their performance is investigated compared with constant weights. According to the results, the two proposed methods have a higher convergence rate than the constant weight algorithm. These methods are compared with each other. Finally, the adaptive weight algorithm with the lowest repeat steps was selected as the optimal method. Both of the proposed methods are stable in all cases.

An important parameter in the deposition of thin-film Indium-Tin-Oxide layers by magnetron sputtering is the spacing between the anode and cathode electrodes, by which the conditions of plasma deposition and the properties of the resulting films are controlled. In this paper, thin ITO films are deposited on glass substrates using the DC-magnetron sputtering technique at RT for different electrode spacing. The thickness of the layers has been measured in the range 110-370 nm. The XRD studies confirm that crystalline structure of ITO is cubic bixbyite. The sheet resistance of the samples prepared at the electrode spacing of 5, 7, 9, and 11 cm, are 17. 7, 16, 1723, and 5207 Ω /□ , respectively. The transmittance of the films in the spectral range of 400-800 nm is 75-85%. The lowest sheet resistance of 16 Ω /□ is obtained at the spacing of 7 cm, having a transparency of 85% and a thickness of 230nm, which confirms that it is appropriate for use as the transparent contact in amorphous silicon solar cells. For such samples the average grain size of the crystallites and the root-mean-square of the roughness of the surface are 50± 5 nm and ≈ 1. 97 nm, respectively. The process simulations are also achieved using xpdp1, X'Pert and MATLAB software.

This paper proposes a digital adaptive calibration to compensate passive Δ Σ modulator error based on output-referred error model of passive integrator. The passive integrator transfer function, which leads to gain and phase errors, degenerates the performance of modulator due to signal attenuation in modulator loop. To identify the adaptive filter factors, a 1-bit pseudo-random noise test signal is applied to the input of the passive Δ Σ modulator. The error factors are identified by making correlation between output and test signals. The simulations show the effectiveness of the intrinsic error compensation of the modulator.

Inappropriate design of transmission line towers in areas with high lightning density increases the outage rate of the line due to direct lightning strikes. In this paper, The lightning activities of areas covered by western regional electricity company during five recent years were collected from weather stations. Based on collected datasets, the Isokeraunic level and the lightning density of different areas have been computed. This information were further used to investigate and predict the outage rate of double circuit 230 kV transmission lines with one shield wire (existing plan) due to direct lightning strikes. Due to high density of lightning activities in some areas, the calculation results show high outage rate in these areas, which has been confirmed by actual experiences. In order to overcome this problem, a tower design with two shield wires was proposed for new developments. The outage rate of the transmission line with the two shield wire design with extracted lightning densities were compared with the existing tower installed. The results show high degree of improvement in outage rates. In order to modify the existing transmission line outage rates, installation of surge arresters and their capabilities were studied. Due to statistic nature of lightning, Monte Carlo method along with EMTP simulations were used in investigations. Simulations results show placement of surge arresters on two upper phases can reduce the shielding failure outage rates in the transmission line, significantly. The pros and cons of surge arrester installations in transmission line to compensate inappropriate tower design are discussed. Results of this study can be used for further transmission line developments in these areas.

With the apperance of cloud computing, users receive computing resources according to pay as you go of cloud service provider. An optimized scheduling approach for mapping all the tasks to the resources is an essential problem due to the limitations and dynamics of resources for requests which vary during the time. This solution may lead to improvement of system's efficiency. There are different methods for cloud computing scheduling with different parameters such as response time, makespan, waiting time, energy consumption, cost, utilization rate, and load balancing. But many of these methods are not suitable for improving scheduling performance in a condition that users requests change during the time. So in this thesis a scheduling method based on reinforcement learning is proposed. Adopting with environment conditions and responding to unsteady requests, reinforcement learning can cause a long-term increase in system's performance. The results show that this proposed method can not only reduce the response time and makespan but also increase resource efficiency as a minor goal. Our proposed illustrates improvements in response time for 49. 52%, 46. 03%, 43. 99%, 43. 53% and 38. 68% over Random, Mix, FIFO, Greedy and Q-sch algorithms, respectively.

This article reports on the design of a 8-channel neural recording amplifier and stimulation back-end in TSMC 0. 18μ m technology which can do the recording and stimulating simultaneously. The design of proposed neural amplifier is based on indirect negative feedback and provides tunable lower cutoff frequency, and digitally-programmable upper cutoff frequency and voltage gain. Moreover, the proposed circuit employs attenuators in the same feedback loop in order to further reduce the silicon area consumed by the capacitors and at the same time to increase the input impedance of the circuit. The 8-channel designed circuit consumes 0. 27mm2 of chip area and operated with a supply voltage of 1. 8V, power consumption of each channel is 27μ W with the THD of-50dB@1kHz and output voltage swing of 0. 95Vpp. The designed stimulation bak-end circuit, in addition to traditional rectangular pulse shapes, can generates biphasic stimulation pulses with exponential shapes, whose time constants are digitally programmable. A class-B second generation current conveyor is designed to be used for delivering stimulation current pulses of up to ± 96μ A to the target tissue and a charge pump block is in charge of the generating supply voltages of ± 3. 3V. The circuit consumes 0. 043mm2 of silicon area for each channel (excluding charge pumps). Simulation results indicate that the stimuli generator meets expected requirements with the maximum power consumption of 1. 2mW when connected to electrode-tissue impedance of as high as 25kΩ .

In this paper design and simulation of a new tunable acceleration MEMS switch is proposed. Using the proposed switch, it is possible to measure the accelerations in the range between milli-g and 90g. To tune the desired acceleration switching, two actuators are used. These actuators are electrostatic comb drive and piezoelectric actuator. The electrostatic comb drive actuator operates in high linear range and the piezoelectric actuator can measure very low accelerations without pull in phenomenon. In the structure, the stopper is used to avoid pull in phenomenon due to the electrostatic actuation. In the piezoelectric actuator section, the governing equations for the deflection due to piezoelectric actuation is extracted. Based on design, the resolution is about 0. 15g. The pull in voltage of electrostatic comb drive actuator is 50 volt and maximum voltage tuning of piezoelectric actuator is 85 volt. To verify, the proposed structure is first calculated using matlab software and then simulated using intellisuite software.

Due to importance of harmonic analysis in interconnected network. In this paper, a method is proposed that could be determined the main responsible of the harmonic pollution of the desired bus voltage or the current desired line in interconnected networks. In this paper, the defects of using the "vector projection" method are presented to solve the above problem; so, a method is proposed based on reducing the amplitude of the harmonic voltage and current due to the reduction of the harmonic amplitude. Both methods are evaluated in a 6-buses IEEE grid using Digsilent and MATLAB software at a specific harmonic frequency. The results of the evaluation of the two methods indicate that the method of "vector projection" is not necessarily the correct answer if the number of harmonic sources is more than two sources; consequently. However, the results of the proposed method are highly rational in the interconnected grids.

Electric arc furnaces are nonlinear and time variant loads that cause high fluctuations of current, voltage and current harmonics, voltage flicker, frequency variations, and so on. On the other hand, these furnaces are the huge energy consumer in power networks. Therefore, it is necessary to present a precise and actual model for analyzing the performance of these loads so that power quality studies, design of reactive power compensators and analysis of the effects of these loads on electric arc furnaces can be done correctly. In this paper, a new model for electric arc furnaces is proposed based on actual measurements of Mobarakeh steel company’ s arc furnaces. This model considers the effects of all elements of arc furnaces, such as the effects of voltage and current transformers, flexible cables, and furnace transformers for modeling. In this way, this model is capable of generating various power quality parameters and can be very useful in the accurate design of reactive power compensators for arc furnaces. The proposed model is compared with one of the most commonly used models, called Cassie-Mayer. The results of the comparison show the accuracy and validity of the proposed model.

Mobile Edge computing is a new paradigm that allows users to combat the constraints of mobile devices For overcoming the long delay of cloud computing which decreases the quality of service, the users send their data to the server which is located at the edge of the network instead of sending a part of the program to a cloud server which is located in place far from the users. In this paper, a system which consists of a cell and several users is considered in which the network users to accomplish their computations request service from cloud and edge servers that are participating with each other. It is assumed that the users access the radio spectrum by a Non-orthogonal Multiple Access (NOMA) method and the queue theory is used for modeling the problem from the users and server sides. The main goal is to minimize the total users' energy consumption, the delay in users' receiving service, and the whole cost of using servers. The mathematical model of this problem would result in a multi-objective constraint non-convex optimization problem. SCA method is used to achieve the globally optimum solution. Through the simulation results, it is shown that by the applied assumptions in the proposed model, the amount of energy, delay and total cost of network is decreased around 50%.

Test case prioritization has been often used to alleviate the costs associated with software regression testing. Current techniques have attempted to estimate the fault exposing potential of test cases using code coverage information and rank them using a heuristic. However, studies show that coverage does not strongly correlate with the effectiveness and fault exposing potential of test cases. Relying on the results of studies that demonstrated the effectiveness of code metrics in fault prediction, we speculate that code metric information can be leveraged to design a new effective technique for test case prioritization. Based on our hypothesis, in this paper, a new prioritization technique is proposed that works based on data fusion on code complexity metrics. The novelty of our technique lies in its original viewpoint to estimate fault exposing potential of test cases in prioritization. To evaluate the proposed technique, we have conducted experiments on faulty versions of seven Java benchmarks. In the experiments, we often observed at least 70% performance in prioritization measured in terms of average percentage of fault detection, which validates our hypothesis.

This paper presents a mixed integer linear program (MILP) model as well as a solution method based on the Benders decomposition algorithm for optimal design of sustainable Energy Hub (EH). A modeling framework to consider environmental (Env) and social (Soc) impacts of the EH's components is incorporated to achieve sustainable EH. First, the life cycles of different components are analyzed in order to determine Env and Soc impacts. Then, the EH design model is developed by using scenarios-based stochastic programming integrating Conditional Value-at-Risk (CVaR) in its objective function as a risk criterion to deal with uncertain nature of parameters. Benders decomposition (BD) algorithm is used to decompose the original problem in order to address heavy computational burden of the problem, especially when a large number of scenarios is used to properly represent uncertainties. The results shows effectiveness of the proposed BD to handle large problem sizes compared to the CPLEX solver and indicates that taking the external cost into account resulted in higher renewable Distributed Energy Resources (DERs) in the optimal configuration, which have lower negative Env and Soc impacts. Also, strength of stochastic programming in handling data uncertainty and controlling risk level is investigated.

Recently, researchers proposed heuristic frameworks which are based on the Wisdom of Crowds in order to evaluate and select the basic results. In these methods, basic results are evaluated by diversity, independency and decentralization metrics. Then, the evaluated results are selected by thresholding, and combined by a consensus function. This paper aims to propose a method for automatic evaluation of the optimized threshold values based on the basic features of the input data in WOCCE. Also, Uniformity, a metric which is based on APMM, is introduced for calculating the diversity of two basic clustering results. Furthermore, Weighted Evidence Accumulation Clustering (WEAC), a new method for considering independency as a weight in the process of combining the basic results, is introduced in this paper. The experimental results indicate that the proposed method has higher efficiency in comparison with the results of other cluster ensemble methods.