مجله مهندسی برق و الکترونیک ایران
سال هجدهم شماره 2 (تابستان 1400)

Phasor measurement units (PMU) have an important role for reliable operation in the power systems. Because of the high cost of PMU and also the lack of suitable communication infrastructure to transfer the measured information, the problem of optimal PMU placement (OPP) for incomplete observability has been proposed. In this paper, a new formulation was proposed to optimize the OPP problem considering the limited number of PMU in the power system. In the proposed formulation, to achieve the higher degrees of precision in state estimation, the fitness function was defined to maximize the number of observable buses and to minimize the depth of unobservability, simultaneously. Also, a comprehensive formulation of the OPP problem was introduced to have a minimum depth of unobservability and three different scenarios were designed to study the effect of considering the depth of unobservability. To simulate the results, IEEE 14, 24, 30, 39, and 57 bus test systems were used. The calculated results showed that the proposed formulation has a proper performance e.g. in IEEE 39 bus test system which had 27 observable bus in the best solution with a maximum depth of degree as 7, could reach to the 26 observable bus with a maximum depth of degree as 4 by changing the PMU placement.

In this paper, the multi-objective PMU placement is presented. Two objectives are fulfilled simultaneously which are the minimization of the equipment installation cost (i.e. minimizing the number of installed PMUs) along with the maximization of the system measure redundancy index which is equal to the maximization of system’s reliability. Moreover, the limitation on the number of PMU measurement channels is investigated for fulfilling the total system observability and increasing the measurement redundancy. In a power system, transformers tap is employed as an equipment for voltage regulation, however, the variation signal of this parameter is not mostly sent to the central control which leads to a severe disturbance in state estimation program. Consequently, transformers tap observability should be considered in PMU placement problem. Accordingly, a virtual bus is added to the grid in order to model the variant transformer tap. Due to the advantages of meta-heuristic methods, the multi-objective Gray wolf algorithm based on dominancy and Pareto criteria concepts is applied.

The use of scattered generation sources, capacitors and reactor banks is rised due to the development of distribution networks as well as rising demand for electricity. Determining the place of installation and capacity are two important factors that affect network power losses and improve network performance. Distributed power plants and reactors, if connected to the grid in an appropriate location, will have various effects, including loss reduction, improved voltage profiles, and increased network reliability. On the other hand, their lack of proper location on the network increases the losses and increases the cost of production and transmission. In this paper, the Zanjan Regional Power Network has been studied and identified with network protection approach. By reviewing the sensitivity of the test points for introducing the DC reactor, a DC reactor is introduced, and by applying the reactor to the points, a sample of the network is investigated and simulation and experimental results are given.

In this paper, the results of partial discharge measurements on more than 25 hydroelectric power plant generators have been investigated. All generators’ insulation defects have been considered and a statistical report is made of types of defects and the share of each defect. statistical parameters for different insulation defects are calculated. Using 3-phase partial discharge measurements that performed simultaneously on each machine, enabled us to apply 3PARD (3-Phase Amplitude Relation Diagram) method for differentiating multiple defects and reducing noises. Effect of PD measurement in High and low frequency ranges for different PD types has been investigated and a correction factor for correlating High frequency measurement data with corresponding value in low frequencies has been proposed. Different Types of PRPD (Phase Resolved Partial Discharge) patterns of insulation defects investigated. Considering the effect of load and temperature conditions on these patterns, with several insulating defects noted, the effect of load and temperature changes on these patterns has been analyzed. Considering the effects of load and temperature changes on partial discharge patterns and using amplitude correction factor in high frequency measurements, it is possible to diagnose machines more accurately and have a better estimation of stator’s insulation status.

Due to space limitation along with the necessity to the increase of the substations, gas insulated substations (GISs) have been favorable option owing to their minimum required space for installation. One of the main challenge in this substation is very fast transient overvoltage (VFTO) giving rise to the operation of disconnector. These waves travel through various components. Interestingly, along with waves, a wave is inducted on the ducts called transient enclosure voltage (TEV) or transient ground potential rise (TGPR). This paper has been devoted to investigate these waves through modeling GIS and presenting some analytical approach. Subsequently, the impact of GIS design, CVT, arrester, and ground strap have been discussed through simulations in EMTP-RV. Finally, the impact of high frequency ground behavior has been discussed to show the importance of the design of grounding system in GISs from high frequency point of view.

This paper proposes a novel non-communication method for detection of uneven phase operation of transmission lines. The proposed algorithm is triggered by receiving the line opening/ closing command and works on the principle of comparison between the amplitude of bus voltage signal and line voltage signal. This protection function can be easily implemented in the numerical line relay and it can quickly detect this problem in power system, if applied.

In this paper, a new method is proposed for dynamic modeling of multi-machine power systems including HVDC-links facing symmetrical and asymmetrical faults. In the proposed method, the space phasors based sequence networks are used to power systems modeling during the fault time. In the sequence networks, synchronous generators and HVDC-links are modelled by injected current space phasors and loads are modelled by constant impedances. In each step time of simulation, after solving Y-matrix equations of the sequence networks, fault current and generators terminal sequence voltages are obtained in the synchronous reference frame. Then, for each generator, these voltages are transferred to the rotor reference frame and the machine dynamic equations are solved. By solving the synchronous machines and HVDC-links dynamic equations the state variables are updated and the current space phasors are calculated and the procedure is repeated for the next time steps. In a single-machine-infinite-bus (SMIB) network and a two-open-conductor faults. Moreover, the simulation results of the method have been compared to EMTP for the standard 33-bus-bar IEEE power system including a HVDC-link, the simulation results of the proposed method and EMTP software have been compared for a single-phase-to-ground and a two-open-conductor faults. The simulations results show the proposed method is reliable.

The synchronous switching reduces the transients caused by switching of the transmission line. By considering the coupling voltages between phases, the proposed algorithm controls the synchronous closing of all three phases of an uncompensated transmission, whenever the other end of the line is open. The algorithm by calculating the delay times for sending commands to each phase of CB, based on mathematical relations, controls the switching of the three phases of CB. It can be applied for different types of configuration of the transmission line (whether they are transposed or not, and balanced or not), various line lengths, the mechanical operation time for each phase of CB, and independent of the opening time of each phase. The results show that by applying this algorithm each phase of CB is closed in the zero-voltage state with a maximum error of three samples.

In this paper, a method for improving the performance of the network protection system in the presence of DGs by use of intelligent devices is proposed. The method, which is implemented on a hierarchical multi-agent structure, can improve the coordination problems between overcurrent relays by using a telecommunications platform based on the IEC-61850 standard information model. By this way, the protection scheme is implemented on three levels on the relays. Each level has tasks that are implemented on a layered multi-agent structure, depending on the layout of the network under certain conditions. At the first protection level performed by relay agents, the relay protection characteristic curve is used along with the functions defined on the intelligent devices. At the second level, a database is used to record the events and the pattern of behavior chosen according to them in order to prevent the flow of information on the network and to speed up the performance of the protection system. Functional patterns recorded in the database allow agents to obtain appropriate settings without sending information to the central unit if similar conditions occur. At the third protection level, in the absence of a similar pattern, the new relay settings are calculated according to the network conditions and transmitted to the network monitoring unit. Finally, in order to verify and validate the proposed method, the proposed scheme is simulated on an experimental network in the ETAP software.

Executing demand response programs will be effective when they are accompanied by the considerable cooperation of participants in addition to appropriate planning. Therefore, distribution companies always emphasize on the higher cooperation of participants in demand response programs. However, lack of proper planning may reduce electricity sale or may have negative impacts on the technical indices of the distribution network. Of the issues which have not been considered in most studies related to demand response are the rate of change in load and the impact of distribution network flexibility in reducing the negative impact of load response programs. In this paper, a model has been proposed that can determine the negative effect threshold for participantschr('39') cooperation in demand response programs by considering a wide range of distribution network features that change due to load shifting. Moreover, by considering flexibility in distribution network and introducing new indices, proper prioritization is proposed for applying reward and punishment policies or for determining the necessary technical changes in order to rate of change in load on each feeder. Based on the proposed model, maximum peak power reduction in determined in the RBTS-BUS5 test system. The results confirm the accuracy of the proposed model.

Power systems have been traditionally designed and operated based on the main principles of reliability, i.e. security and adequacy. These principles can properly deal with the known failures in the power infrastructure components. However, recently occurred High Impact Low Probability (HILP) threats have raised the concerns about the classical reliability-oriented view. Hence, resilient distribution networks against various threats and enhancement of energy security in critical situations are the most significant challenges that must be taken into consideration. Distribution network plays vital role in power systems because of its connection point to the end users and due to its wide range, security of electricity supply is strongly dependent on it. Hence, a lot of efforts have been done on distribution networks scheduling, planning and operation. So, optimal resilient planning of conventional medium voltage distribution network is aimed at mitigating vulnerability to natural disasters. In this paper, a new method is suggested to make a relation between network component fragility curves, component geographical location and natural disasters spatial risk index. To achieve this goal, the resiliency concept and effective solutions for resiliency increment, elicitation of vulnerability atlas of distribution networks and natural disaster risk based modelling have been discussed and assessed comprehensively in this paper. The simulation of proposed method is applied on a large scale distribution network. The obtained results validate the effectiveness of proposed methodology.

In this paper, restoration of the customers in distribution network considering the annual financial losses of the process trips has been studied. The most purpose for restoration of customers is to the restore the maximum possible load. For evaluation of the financial losses of the process trips, a new concept in the field of power quality named process immunity time have been used. Impact of two equipmentchr('39')s – disconnector and automated sectionalizing switches on the geographic area, financial losses and reliability indices will be reviewed. In this paper, restoration priority will be with the customers outside of the fault section. IEEE 34 node test feeder is simulated as the base network and the results will be analyzed and evaluated. Then two DGchr('39')s are connected in two arbitrary locations. Eventually, DG will be transferred to near the industrial load in order to examine the results. In this paper, restoration priority is with customers outside of the fault section.

Given the lack of access to the electricity distribution network in remote areas, rising fossil fuel prices and transportation costs, the importance of using renewable energy in these areas is increasing. Due to the high price of equipment such as wind turbine, solar panels and batteries, capacity sizing of the equipment is vital. In this research, the optimal design and practical implementation of renewable energy systems in a remote area in Sistan and Baluchestan province have been investigated and the net present value function has been modeled and solved by the GAMS software as MILP. In the following, practical implementation based on computer simulation has been considered and the operating scenario is set to 4 kW of solar panels, 28 seald acid batteries, 1200 W wind turbines and 4 kW inverter. Economic analysis shows that renewable energy systems will be utilized by a 35% reduction in costs relative to energy from diesel generators.

In this paper, to implement the reconfiguration of the intelligent distribution network and sizing the distributed generations (DGs), simultaneously, a stochastic multi-objective framework is presented in which the goals of reducing the energy losses, operating costs and emissions of environmental pollutants are followed. In most of the researches carried out for the network reconfiguration, the customer and distribution substation loads have been assumed to be fixed, while in reality these loads are changing. Generally, in the power system planning, the worst case which is the maximum possible load is considered, but the obtained configuration with this assumption is certainly not the optimal configuration. Therefore, in this paper, in order to cover the uncertainties associated with the load, the framework for the reconfiguration and DGs sizing problem is presented as a chance constrained programming problem and the Monte Carlo simulation is used in it. The problem of reconfiguration and DGs sizing at the same time in order to achieve predetermined goals is a multi-objective optimization problem. In order to solve it, this paper uses a newly introduced powerful optimization method called human-behavior based optimization approach. Finally, the proposed stochastic multi-objective framework is implemented on an IEEE 33-bus network, and the results are analyzed.

This paper presents a new Switched-Capacitor Multilevel Inverter (SCMLI) structure. The presented configuration generates a staircase almost sinusoidal AC voltage by using a single DC source and a few capacitors which its amplitude is several times the input voltage. The operation of the converter is controlled by the help of NLC modulation strategy to choose the best switching states in a way that they always ensure the self-balancing capability of the capacitors voltages. Low switching frequency, simple control and inherent bipolar output are some of the advantages of the presented SCMLI. In comparison with other existing topologies, to reach a specific voltage level, the structure requires less circuit elements since the capacitors in follower modules can be charged up to 3 times the input voltage. Having a modular structure, facilitates the expansion of the circuit to attain higher voltage levels. In order to investigate the performance of the presented topology, MATLAB software was used. Furthermore, a 13-level laboratory prototype was performed to affirm the efficacy of the circuit under different loads using AVR ATMEGA 16A microcontroller. Both simulation and experimental results accredit the appropriate performance of the converter. Finally, after calculating losses, a theoretical efficiency of 92.72% is reached.

Multilevel inverters are a new generation of DC-AC converters at medium and high voltage and power levels. In this paper, a new single-phase cascaded multilevel inverter is presented. For this purpose, a basic cell is first presented and then by series connection of these cells, the new multilevel inverter structure is presented. The proposed new cell is only capable of generating positive voltage levels, and therefore, to produce zero and negative voltage levels, the proposed structure is constructed based on H-bridge module. A comprehensive comparison is carried out between the proposed multilevel inverter with the classical and recently introduced structures in terms of the number of switching devices, the number of drivers, the total blocking voltage of the switches as well as the loss and efficiency. The accuracy of the proposed inverter’s performance is simulated in Matlab/Simulink in symmetric and asymmetric topologies fora 21-level and 37-level output voltage respectively, and then evaluated by the laboratory prototype.