فصلنامه روشهای هوشمند در صنعت برق
پیاپی 57 (بهار 1403)

In this paper, a non-isolated high step-up soft-switching converter is proposed. The proposed converter is a boost converter combined with two voltage multiplier cells for boosting output voltage. Also, extend voltage gain of the proposed converter is achieved by using a coupled-inductor. Compare with other similar high step-up topologies with the same number of components, the proposed converter has a higher voltage gain and higher efficiency. An active clamp circuit is used so, the zero-voltage switching (ZVS) is achieved. Also, in the proposed converter, the voltage stresses on the switches are low. As the voltage stress decreases on the switch, Ron</sub> of the MOSFET is deceased and as a result conduction loss of the switch is decreased. So, the efficiency of this converter increased. In this paper, operational principle of the converter is described and the analytical, simulated results and prototype converters are validated using a 20V input and 400V </strong>output converter at 200W load.

Due to the high penetration of renewable energy resources and the direct impact on the power system, the issue of energy management has received more attention than researchers. Power-to-gas (P2G) system causes the surplus electricity generated from renewable energy resources in the network to be converted to gas and sold to the gas network, so energy management and profitability are a matter of particular importance, considering the two grids as a joint optimization of integrated energy systems. This paper presents a scenario-based stochastic mixed-integer linear programming (MILP) model to optimize integrated gas and electricity integrated systems considering natural gas distributed generation resources, P2G systems, energy storage systems, and electric vehicles. It aims to reduce the cost of purchasing energy and cut off the power of renewable energy resources. The 33-bus power distribution network and the 7-node natural gas network are considered for the analysis of the proposed model, and the proposed model is solved using the powerful Gurobi solver, considering various cases. The results of different cases show the performance of the proposed model.

Severe events such as floods, earthquakes and hurricanes cause disruption in the operation of distribution networks and lead to their islanding. In such cases, if the distribution networks have microgrids, these microgrids are able to separate from the main network and exchange energy with each other to reduce the operation and outage costs. Therefore, the energy management in a multi-microgrid network requires a decentralized operating framework to encourage microgrids to have transactions with each other by providing the necessary incentives. This paper developes a completely decentralized framework to improve the resilience of microgrids based on the organization of peer-to-peer energy transactions, taking into account the appropriate financial incentives for the participation of microgrids. The developed model protects the private data of each microgrid, such as load information and distributed generation resources, during market settlement. Using the developed decentralized model, microgrids can increase network resilience in the context of peer-to-peer energy exchanges, in addition to reducing their operating costs compared to the island mode. The proposed decentralized approach does not require a central controller and has a high convergence speed. Simulations are performed on a system with fourteen microgrids and the results are compared with the island approach to evaluate the performance of the proposed method. The simulations are performed in MATLAB R2020b environment using YALMIP toolbox. CPLEX 12.9 is also used to solve the optimization problem. The results show the efficiency of the proposed method in increasing the resilience and reducing the operating costs.

Reducing the number of voltage sources and the power electronics components while obtaining voltage boosting in the output voltage are the key parameters in the research area of the multilevel inverter design. A lesser number of components would ensure lesser cost while higher boosting ability increases its application potential. In this paper, a new H-bridge based single-source switched capa citor multilevel inverter structure is introduced. The proposed structure including two K-type units (KTU) can produce nineteen voltage levels with a voltage boosting of 1.5 times the input voltage. This converter consists of fourteen switches, two diodes, one voltage source and five capacitors with self-balancing capability. A comprehensive comparative comparison with the recent presented topologies have been carried out to investigate the performance of proposed structure. The main features of the proposed structure are utilizing single DC voltage source, self-balancing of the capacitors the capability of the input voltage, reducing the power electronics components in terms of voltage level count, and thus reducing the overall cost. The simulation results in the Matlab/simulink environment and the experimental laboratory results are provided to verify the satisfactory operation of the propo sed topology.

Design and optimization of high-power electric machines for the use of electric vehicles is one of the important issues today for the development of green technologies. Engines required for electric vehicles must have a power of more than 50 kW and must also produce torques of more than 200 Nm. The motors currently most commonly used in electric vehicle propulsion are permanent magnet synchronous machines. Due to the many problems of using permanent magnets in electric machines, the use of non-magnet electric machines such as switched reluctance machines have received much attention. Double stator switched reluctance machine is one of the newest types of these machines. Recently, another new type of electric machine called induction switched reluctance machine has been introduced for the use of electric vehicles. In this machine, the rotor conductors act like a magnetic shield by deflecting the magnetic flux, preventing magnetic field lines from passing through the rotor body. In this paper, a double-stator switched reluctance machine and an induction switched reluctance machine are considered and their properties are extracted by finite element method. The simulation results including torque profile, torque density and efficiency are presented and compared. Finally, the best topology for electric propulsion is proposed.

Floating photovoltaic (FPV) systems are a new approach to the use of water-based photovoltaic (PV) systems. This system creates a new opportunity to increase the production capacity of solar PV systems, especially in the northern regions of Iran, where the price of land is high. To enhance the efficiency of inverters connected to the network of FPV systems in the northern regions of Iran, we have combined the structure of a two-stage and a multi-string inverter in this paper. On the other hand, the perturb and observe (P&O) method is one of the most common methods for maximum power point tracking (MPPT) with a variety of disadvantages including algorithm fluctuations during sudden changes in radiation. Since these sudden changes during radiation occur abundantly in the northern regions of Iran due to cloudy weather, a modified P&O algorithm is proposed by adding a current change parameter to overcome this problem. In fact, the ZETA converter and the proposed algorithm are used in inverter and track the maximum power point and in the second stage, DC to AC conversion occurs. To evaluate the efficiency improvement, the proposed inverter is compared with a single-stage centralized inverter. This study also considered the effect of wind and water temperature on the production capacity of the FPV system. System simulation is performed using Matlab/Simulink software. The simulation results show that the proposed two-stage multi-string inverter produce an average of 18.88 kWh, which is an increase compared to the centralized single-stage inverter.

In this paper, a novel general architecture for single-loop Sigma-Delta Modulator is presented by combination low-distortion and noise-coupled techniques for high-resolution low-power applications. The low-distortion technique in the proposed architecture makes its signal transfer function equal to one. In addition, the noise-coupled technique increases the order of quantization noise shaping at the modulator output. The purpose of using these techniques in design of the architecture is to increase the order of the modulator without needing to additional operational amplifiers during its circuit implementation to finally achieve a low-power modulator compared to similar ones. To reduce the required amplifiers, a second order infinite impulse response (IIR) filter was used instead of an integrator in the modulator loop. To evaluate the performance of the proposed structure, its implementation and simulation for speech recognition application, i.e., digital hearing aids, were performed in 180nm CMOS (complementary metal-oxide semiconductor) technology. For a third-order structure with a sampling rate of 64 and an input sine signal of -6dBFS and a sampling frequency of 2.56MHz, the signal to noise and distortion (SNDR) is 81.9dB and the dynamic range (DR) is 88dB. The power consumption of the modulator is 126.9 μW and its bandwidth is 20 KHz. The results of circuit and system level simulations prove its performance.

One of the main reasons of death in the world, mostly affecting seniors, is brain stroke. Almost 85% of all brain strokes are ischemic due to internal bleeding in a part of the brain. Due the high mortality rate, quick diagnosic and treatment of ischemic and hemorrhagic strokes are of utmost importance. In this paper, to realize microwave brain imaging system, a circular array-based of modified bowtie antennas located around the multilayer head phantom with a spherical target with radius of 1 cm as intracranial hemorrhage target aresimulated in CST simulator. To obtain satisfied radiation characteristics in the desired band (from 0.5-5 GHz) an appropriate matching medium is designed. First, in the processing section, a confocal image-reconstructing method based using delay and sum (DAS) and delay, multiply and sum (DMAS) beam-forming algorithms is used. The reconstructed images generated shows the usefulness of the proposed confocal method in detecting the spherical target in the range of 1 cm. The main purpose of this paper is stroke classification using deep learning approaches. For this, an image classification algorithm is developed to estimate the stroke type from reconstructed images. By using the proposed deep learning method, the reconstructed images are classified into different categories of cerebrovascular diseases using a multiclass linear support vector machine (SVM) trained with convol uti onal neural networks (CNN) features extracted from the images. The simulated results show the suitability of the proposed image reconstruction method for precisely localizing bleeding targets, with 89% accuracy in 9 seconds. In addition, the proposed deep-learning approach shows good performance in terms of classification, since the system does not confuse between different classes.

As a promising vision toward obtaining high reliability and better energy management, nowadays power grid is transferring to the smart grid (SG). This process is changing continuously and needs advanced methods to process big data produced by different segments. Artificial intelligence methods can offer data-driven services by extracting valuable information which is produced by meter devices and sensors in smart grids. To this end, machine learning (ML), deep learning (DL), reinforcement learning (RL), and deep reinforcement learning (DRL) can be applied. These methods are able to process huge amounts of data and propose an appropriate solution to solve power industry complex problems. In this paper, the state-of-the-art approaches based on artificial intelligence used by smart power grids for applications and data sources are investigated. Also, the role of big data in smart power grids, and its features such life cycle, and efficient services such as forecast, predictive maintenance, and fault detection are discussed.

High blood pressure is one of the risk factors for coronary heart disease, which causes severe damage to the body. A timely diagnosis of blood pressure disease can protect a person from the complications of this disease. A noninvasive method for measuring blood pressure is oscillometric. Accordingly, the blood pressure is estimated by measuring the oscillations created by the opposition of the arterial pressure and the pressure of the cuff wrapped around the arm. In this research, the main goal is to create a software platform for simulating the behavior of veins and cuffs, which can be used to check the performance of different blood pressure measurement algorithms by the Oscillometric method. In this regard, all components including the cuff, and brachial artery, how to extract oscillations from the blood pressure curve, and estimate systolic and diastolic pressures will be modeled. By modeling in MATLAB, the blood pressure measurement can be evaluated without the need for a clinical condition. The output of blood pressure parameters can be obtained by entering the main characteristics of arterial pressure as input. The output of modeling with real samples of 50 measured cases and the accuracy of estimating systolic and diastolic pressures according to two algorithms of maximum oscillation and The maximum/minimum slope were checked considering the actual values. The results of comparing the modeling performance with the measured values indicate that the maximum oscillation algorithm has a better performance than the maximum/minimum slope algorithm. The mean error value in the maximum oscillation algorithm for maximum amplitude pressure, systole, and diastole is 0.64 ± 1.9, 0.82 ± 1.6, and 5.1 ± 6.8, respectively.