نشریه مهندسی برق و مهندسی کامپیوتر ایران
سال هفدهم شماره 3 (پیاپی 52، پاییز 1398)

In this paper, a novel structure and a control method to improve the performance of inductive heating circuits is proposed. In the presented structure, with the combination of the performance of a half-bridge resonant converter with voltage-boost capability, the reduction in output efficiency at low power and at high power is compensated to an acceptable level. The use of the low number of switches and diodes, the use of the high-quality capacitors with low capacities, good quality of the input current and also high power factor ensures the proper operation of the proposed converter. The switching of high frequency switches in the proposed structure is carried out as soft-switching where resulting in very low switching losses. In this converter, the design of the input filter in order to prevent the effects of electromagnetic interference has been prepared. Finally, to demonstrate the proposed structure operation, the simulation and experimental results are presented.

Controlled reactors as one of the flexible AC transmission systems play an important role in the availability and reliability of power systems.However, in the conventional reliability assessment of power systems, reactive power is considered only as a constraint for the network, and so far no precise model for assessing the reliability of reactors has been provided. In this paper, a new reliability model based on Markov process is proposed for a magnetically controlled reactor (MCR). In the modeling process, first the MCR structure is divided into two distinct parts, and then the extracted Markov models are combined based on frequency/duration technique.Since temperature changes play a significant role in changing the failure rate of electrical equipment, the effect of temperature changes in accordance with the MIL-217F standard has been considered in the proposed model and its impact on the probability of MCR operating modes has been evaluated. The simulation results have shown that in normal temperature conditions, the control system and at high temperatures, reactor windings can have the greatest impact on the availability of MCR. Comparison of reliability indices at different temperatures has shown that under different temperature conditions, different components will affect the availability of MCR. Therefore, in this condition, the measures needed to improve the reliability of the reactor can be different. This fact highlights the importance of considering the effect of operating temperature on reliability assessment as well as planning for preventive maintenance to improve the performance of reactive power sources.

In this paper, the modal analysis of a grid connected doubly-fed induction generator (DFIG) using small signal stability analysis is presented and effect of variation in system parameters such as mutual inductance, stator resistance, line reactance, shaft stiffness and wind speed on the eigenvalues, stability and damping ratio of different modes of system are studied. This analysis shows that which parameters variation can deviate the system from normal working conditions as well as which parameters variation can improve the system behavior. For evaluating the stability and different controllers design, the simulation results show the effects of parameters variations.

Planning for the coordinated expansion of generation units, taking into account the fuel needed, has always attracted the attention of power industry planners. Failure to coordinated expansion of electricity and gas networks will result in excessive investment and, in some cases, defects in power supply. Therefore, there is a need to a model coordinating the expansion of electricity and gas networks, while taking into account the technical constraints. In this study, centralized expansion of gas and electricity networks is modeled by considering N-1 security criterion in gas network. This modeling is from the perspective of a central investor who, considering the technical constraints, seeks to minimize the total cost of investment and operation of the electricity and gas networks. The results of the gas network investment problem indicate that there is a need to increase pipeline capacity in some areas. In the proposed case study, the investment cost of the gas network is $19 million, while the total cost of investment and operation of the gas network is $37.19 billion. On the other hand, in the electricity grid, new power plants need to be installed in the designated areas. The results also indicate that the capacity of the F-H transmission line should be increased. Moreover, considering the N-1 criterion for gas pipeline outages, the power grid would prefer to install about 3200 MW of new generation units all around the grid thereby boosting the power network against pipeline outages. However, 2400 MW of new generating units would be adequate when N-1 criterion is omitted.

In this paper, the impedance of low voltage grids in frequency range of 2 kHz - 150 kHz is estimated using rectangular pulse injections and signal processing techniques. The grid impedance is defined as division of voltage signal to current signal in frequency domain. In noisy condition, the accuracy of impedance estimation is directly dependent to energy of injected signal. The injection signal must has sufficient energy in the frequency range of estimation for an accurate impedance estimation. In the proposed method, several injection signals with different widths are selected with the Genetic algorithm. The grid responses to the injected signals are measured and then denoised for an accurate impedance estimation. When the measurement duration is low, the whole transient state of the grid is not measured; hence the impedance estimation is not accurate. Therefore, in this paper a method is proposed for determining the best measurement duration for impedance estimation using Time-Frequency distributions. The proposed method is applied on several simulated grids and the results show the ability and accuracy of the proposed method in grid impedance estimation.

In this paper, a new algorithm based on Kalman Filter (KF) is proposed for improving the reliability of power transformer differential protection. The three-phase currents of the transformer at the energization side are first estimated by KF. Three residual signals which are the differences between the measured (by current transformers) and estimated (by KF) are defined and considered as the decision criteria to discriminate the fault and inrush conditions. During the transformer energization, the residual signals are almost zero. However, in fault conditions, the residual signals exceed a determined threshold value and a trip command will be subsequently issued. The performance of the proposed method is verified using simulations in MATLAB and PSCAD software.

In this paper, a low voltage boost DC-DC converter has been presented. The circuit can be used for increasing the output voltage of miniature low voltage generators such as TEG, solar and piezoelectric. The converter is fully integrated and works with low voltages as low as 200 mV, and the output voltage can reach 1 V. Body biasing has been used to handle low input voltages. The output power density is 50 µW/mm2, and the converter uses 5 cross coupled stages with 76% efficiency. The maximum total efficiency of the converter for 6 µA load is 52%. The converter uses 0.2 mm2 of chip area using 90 nm technology.

Blind estimation of Physical layer transmission parameters, is one of the challenges for smart radios to adapt itself to network standards. These parameters could be transmission rate, modulation and coding scheme that is used for combating with channel errors. Therefore, Channel Coding Estimation, including code parameters, parity check matrix and generator matrix estimation, is one the interesting research topics in the context of software radios. Algebraic methods like Euclidean methods and Rank-based methods are usually performed on intercepted received sequence to estimate the code. Poor efficiency in a high error probability environment is the main drawback of this methods. Transform-based methods, like Walsh-Hadamard transform is one of the methods that could solve channel coding estimation problem. In this paper, new algorithm based on Walsh-Hadamard Transform is proposed that could reconstruct the parity check matrix of convolutional code with general k/n rate in a high error probability environments (BER>0.07), that has much better performance compared to other methods. This algorithm exploits algebraic properties of convolutional code in order to form k-n equation for estimation of k-n rows of the parity check matrix and then use Walsh-Hadamard transform to solve these equations. Simulation results verified excellent performance of the proposed algorithm in high error probability environments compared to other approaches.