Search

Near-memory processing (NMP) is a prospective paradigm enabling memory-centric computing. By moving the compute capability next to the main memory (DRAM modules), it can fundamentally address the CPU-memory bandwidth bottleneck and thus effectively improve the performance of memory-constrained workloads. Using the personalized recommendation system as a driving example, we developed a scalable, practical DIMM-based NMP solution tailor-designed for accelerating the inference serving. Our solution is demonstrated on a versatile FPGA-enabled NMP platform called AxDIMM that allows rapid prototyping and evaluation of NMP's performance potential on real hardware under a realistic system setting using industry-representative recommendation framework. We experimentally validated the performance of a two-ranked AxDIMM prototype which achieves up to 1.89x speedup in latency and 31.6% memory energy saving for embedding operations. For end-to-end recommendation inference serving, AxDIMM improves the throughput up to 1.5x and latency-bounded throughput up to 1.77x, respectively.

Purpose: The final purpose of this study is to minimize the cost and time required in the green remodeling decision making and further promote the green remodeling market, and in this work, a standard model for each type of public building was developed as a first step to develop judgement tools to support pre-decision making. Method: The standard model refers to a series of databases with pre-set input variables for each type of building in order to minimize existing complex energy simulation input variables and increase accessibility to non-specialists. The database was established using existing public building licensing data, building energy efficiency certification cases, legal standards, and related research. Result: Based on the standard model database, we will develop support tools for pre-decision making of green remodeling and develop a web-based system that can easily compare energy requirements before and after green remodeling improvement and calculate expected construction costs.

This paper presents a direct identification algorithm for continuous-time low-order models both with delays and free of delays based on the method of weighted moments matching. In this approach, it is sufficient to obtain response data stimulated by a single pulse, such as a rectangular or a half-sine pulse. From the weighted time moments of both input and output data, the parameters of low-order transfer function models are determined by solving linear equations in a matrix form. Numerical examples show that the proposed algorithm provides robust identification results despite the fact that the system considered is oscillatory with a long tail and contains measurement noise.

Biological desulfurization processes of landfill gas yield an enormous amount of biologically produced S (BPS) as a byproduct. Capability of BPS to remove Cd

Biological desulfurization processes of landfill gas yield an enormous amount of biologically produced S (BPS) as a byproduct. Capability of BPS to remove Cd2+ from aqueous solutions was tested and its removal efficiency was compared to that of granular activated carbon (GAC). Kinetics of Cd2+ removal by BPS was a two-stage process with an initial rapid adsorption showing 45% of initial Cd2+ was removed within 5 min, followed by a slower adsorption. Cadmium adsorption onto the BPS fitted the Langmuir isotherm model and maximum adsorption capacity of the BPS (63.3 mg g−1) was 1.8 times higher than that of GAC (36.1 mg g−1). Thermodynamic parameters showed that Cd2+ adsorption by BPS was favorable and endothermic. Data from XPS proved the main adsorption mechanism to be complexation of Cd2+ with sulfides in the BPS. Results demonstrated that BPS can be recycled as a novel adsorbent for Cd2+ removal from wastewater.

Consider a linear fourth-order system with no zero that is represented in terms of four specific parameters: two damping ratios and two natural frequencies. We investigate several interesting questions about the maximum overshoot of the system with respect to the four-tuple parameters. Some remarkable results are presented.

Commutation failure in line commutated converter based HVDC systems cause severe damages on the entire power grid system. For LCC–HVDC, thyristor valves are turned on by a firing signal but turn off control is governed by the external applied AC voltage from surrounding network. When the fault occurs in AC system, turn-off control of thyristor valves is unavailable due to the voltage collapse of point of common coupling (PCC), which causes the commutation failure in LCC–HVDC link. Due to the commutation failure, the power transfer interruption, dc voltage drop and severe voltage fluctuation in the AC system could be occurred. In a severe situation, it might cause the protection system to block the valves. In this paper, as a solution to prevent the voltage collapse on PCC and to limit the fault current, the application study of resistive superconducting fault current limiter (SFCL) on LCC–HVDC grid system was performed with mathematical and simulation analyses. The simulation model was designed by Matlab/Simulink considering Haenam-Jeju HVDC power grid in Korea which includes conventional AC system and onshore wind farm and resistive SFCL model. From the result, it was observed that the application of SFCL on LCC–HVDC system is an effective solution to mitigate the commutation failure. And then the process to determine optimum quench resistance of SFCL which enables the recovery of commutation failure was deeply investigated.

Recently, studies on HVDC circuit breaker (CB) prototypes have shown successful test results. Nevertheless, effective and reliable solutions regarding massive fault energy during dc fault interruption have not yet been commercialized, and dc current breaking topologies on methods of achieving artificial zero should be somewhat modified. As an alternative, one feasible solution is to combine fault current limiting technologies with dc breaking topologies. In this paper, we studied the application of resistive superconducting fault current limiters (SFCLs) on various types of HVDC CB in order to estimate the effects of combining fault current limiters and conventional dc breakers. For the simulation works, four types of dc breaker topologies were modeled, including a mechanical CB using black-box arc model, a passive resonance CB (PRCB), an inverse current injection CB, and a hybrid HVDC CB. In addition, a resistive SFCL was simulated and added to the dc breakers to verify its interruption characteristic and distributed energy across HVDC CB. From the simulation results, we found that the maximum fault current, interruption time, and dissipated energy stress on the HVDC CB could be decreased by applying SFCL. In addition, it was observed that, among four types of HVDC CB, PRCB with SFCL exhibited the best observable enhancement.

Among the accessories of the dc high-temperature superconducting (HTS) power cable, a stop joint box (SJB) is an essential component for the long-distance transmission cable connections and the separation of the cooling section. The main insulating material of the SJB is a polypropylene laminated paper (PPLP), which is already generally used for dc HTS power cable. In order to separate the cooling section, an epoxy spacer is located at the center of the SJB. Due to the weakness of surface breakdown characteristics compared with that of punctual breakdown and different nature of dc electric field distribution determined by the ratio of conductivities of dielectric materials, the epoxy spacer configuration of the SJB should be carefully designed based on dc electric field analysis. In this paper, the optimum configuration of the epoxy spacer of the SJB was investigated based on dc electric field analysis. On the basis of the initial SJB model, which is already designed for the joint box of the conventional ac oil-filled power cable, a variety of improved SJB models with different epoxy spacer configurations were considered. In dc electric field simulations, the main objective was to reduce the tangential dc electric field intensity at the interface between PPLP and epoxy spacer. To find the optimum epoxy spacer configuration, a parametric sweep technique was applied, and an appropriate SJB model could be identified. However, its tangential dc electric field intensity remained still higher than that of the design criteria for the dc electric field from the experimental results. Therefore, as a novel method, a thin Kraft layer between the PPLP and the epoxy spacer was added to reduce electric field intensification. The simulation result of the newly designed insulation structure revealed the possibility of lowering the tangential electric field. Furthermore, different electrical conductivity values were applied to the additional layer instead of the original Kraft value in order to meet the critical electric field intensity. Finally, 1E-11 S/m, among those values, showed the best simulation result, which was lower than the design criteria for the dc electric field. To verify the effect of the additional layer for dc electric field mitigation, miniature SJB specimens were fabricated, and dc breakdown tests were performed using a step-by-step test method. Based on the experimental results, SJB specimens with an additional layer showed the highest breakdown strength compared with that of SJB specimens without an additional layer. Thus, the optimum configuration of SJB determined by dc electric field analysis was approved by experimental works. Consequently, these insulation design skills based on dc electric field analysis could be usefully adopted in developing the SJB of 80-kV dc HTS power cables.

The key obstacle in integrating high-voltage direct current (HVDC) point-to-point networks into meshed multiterminal HVDC networks (MTDC) is the absence of dc circuit breakers (DCCBs), which can timely and reliably isolate the faulty HVDC network from the MTDC. In this paper, a novel hybrid-type superconducting DCCB model (SDCCB) is proposed. The SDCCB has a superconducting fault current limiter (SFCL) located in the main line, to limit the fault current until the final trip signal to the SDCCB is given. After the trip signal, insulated-gate bipolar transistor (IGBT) switches located in the main line will commutate the fault current into a parallel line, where dc current is forced to zero by combination of IGBTs and surge arresters. DC fault current behavior in MTDC and fundamental requirements of DCCB for MTDC were described, followed by an explanation of the working principles of the SDCCB. To prove the viability of the SDCCB, a simulation analysis demonstrating SDCCB current interruption performance was done for changing the intensity of dc fault current. It was observed that the passive current limiting by SFCL caused significant reduction in fault current interruption stress for SDCCB. Furthermore, fundamental design requirements for SFCL, including the effect of SFCL quenching impedance on SFCL voltage rating and energy dissipation capacity, were investigated. Finally, advantages and limitations of the SDCCB were highlighted.

High Voltage Direct Current (HVDC) system has been evaluated as the optimum solution for the renewable energy transmission and long-distance power grid connections. In spite of the various advantages of HVDC system, it still has been regarded as an unreliable system compared to AC system due to its vulnerable characteristics on the power system fault. Furthermore, unlike AC system, optimum protection and switching device has not been fully developed yet. Therefore, in order to enhance the reliability of the HVDC systems mitigation of power system fault and reliable fault current limiting and switching devices should be developed. In this paper, in order to mitigate HVDC fault, both for Line Commutated Converter HVDC (LCC-HVDC) and Voltage Source Converter HVDC (VSC-HVDC) system, an application of resistive superconducting fault current limiter which has been known as optimum solution to cope with the power system fault was considered. Firstly, simulation models for two types of LCC-HVDC and VSC-HVDC system which has point to point connection model were developed. From the designed model, fault current characteristics of faulty condition were analyzed. Second, application of SFCL on each types of HVDC system and comparative study of modified fault current characteristics were analyzed. Consequently, it was deduced that an application of AC-SFCL on LCC-HVDC system with point to point connection was desirable solution to mitigate the fault current stresses and to prevent commutation failure in HVDC electric power system interconnected with AC grid.

Voltage source converter-based HVDC systems (VSC-HVDC) are a better alternative than conventional thyristor-based HVDC systems, especially for developing multi-terminal HVDC systems (MTDC). However, one of the key obstacles in developing MTDC is the absence of an adequate protection system that can quickly detect faults, locate the faulty line and trip the HVDC circuit breakers (DCCBs) to interrupt the DC fault current. In this paper, a novel hybrid-type superconducting circuit breaker (SDCCB) is proposed and feasibility analyses of its application in MTDC are presented. The SDCCB has a superconducting fault current limiter (SFCL) located in the main current path to limit fault currents until the final trip signal is received. After the trip signal the IGBT located in the main line commutates the current into a parallel line where DC current is forced to zero by the combination of IGBTs and surge arresters. Fault simulations for three-, four- and five-terminal MTDC were performed and SDCCB performance was evaluated in these MTDC. Passive current limitation by SFCL caused a significant reduction of fault current interruption stress in the SDCCB. It was observed that the DC current could change direction in MTDC after a fault and the SDCCB was modified to break the DC current in both the forward and reverse directions. The simulation results suggest that the proposed SDCCB could successfully suppress the DC fault current, cause a timely interruption, and isolate the faulty HVDC line in MTDC.

High-voltage direct current (HVDC) technology is considered to have some important advantages over traditional high-voltage alternating current, such as higher overall efficiency and smaller power losses for long-distance transmission. In addition, applications of superconducting cables in dc electric power networks may realize real zero impedance, and the economic and technical advantages could be maximized. Therefore, many research institutes have tried to develop advanced superconducting cables for HVDC grids with higher reliability, by considering insulation diagnosis in order to avoid unexpected failures. As one of the plausible diagnostic methods for power cables applied to the ac grid, the detection of partial discharges (PDs) taking place inside the apparatus has been widely investigated. With regard to the related PD pattern analysis, a phase resolved PD analysis (PRPDA), which was first developed in the early 1970s, accounts for the phase information of the applied ac voltage. In 2001, we also proposed a method for pattern recognition, i.e., chaotic analysis of PD (CAPD), that considers three normalized parameters obtained from the values between two consecutive PD pulses: amplitude difference (Pt), occurring time difference (Tt), and applied voltage difference (Vt). However, none of the proposed methods of pattern analysis can be employed for PD under dc stress. Therefore, in this paper, we propose a modified CAPD for the related pattern recognition of possible defects inside a joint box and termination of an HVDC superconducting cable. PDs are produced from four artificial defects and are then detected by a self-designed and fabricated sensor, for which the analysis was performed based on our newly modified CAPD.

Due to fewer risk of commutation failures, harmonic occurrences and reactive power consumptions, Voltage Source Converter (VSC) based HVDC system is known as the optimum solution of HVDC power system for the future power grid. However, the absence of suitable fault protection devices for HVDC system hinders the efficient VSC-HVDC power grid design. In order to enhance the reliability of the VSC-HVDC power grid against the fault current problems, the application of resistive Superconducting Fault Current Limiters (SFCLs) could be considered. Also, SFCLs could be applied to the VSC-HVDC system with integrated AC Power Systems in order to enhance the transient response and the robustness of the system. In this paper, in order to evaluate the role of SFCLs in VSC-HVDC systems and to determine the suitable position of SFCLs in VSC-HVDC power systems integrated with AC power System, a simulation model based on Korea Jeju-Haenam HVDC power system was designed in Matlab Simulink/SimPowerSystems. This designed model was composed of VSC-HVDC system connected with an AC microgrid. Utilizing the designed VSC-HVDC systems, the feasible locations of resistive SFCLs were evaluated when DC line-to-line, DC line-to-ground and three phase AC faults were occurred. Consequently, it was found that the simulation model was effective to evaluate the positive effects of resistive SFCLs for the effective suppression of fault currents in VSC-HVDC systems as well as in integrated AC Systems. Finally, the optimum locations of SFCLs in VSC-HVDC transmission systems were suggested based on the simulation results.

High temperature superconducting (HTS) cable has been paid much attention due to its high efficiency and high current transportation capability, and it is also regarded as eco-friendly power cable for the next generation. Especially for DC HTS cable, it has more sustainable and stable properties compared to AC HTS cable due to the absence of AC loss in DC HTS cable. Recently, DC HTS cable has been investigated competitively all over the world, and one of the key components of DC HTS cable to be developed is a cable joint box considering HVDC environment. In order to achieve the optimum insulation design of the joint box, analysis of DC electric field distribution of the joint box is a fundamental process to develop DC HTS cable. Generally, AC electric field distribution depends on relative permittivity of dielectric materials but in case of DC, electrical conductivity of dielectric material is a dominant factor which determines electric field distribution. In this study, in order to evaluate DC electric field characteristics of the joint box for DC HTS cable, polypropylene laminated paper (PPLP) specimen has been prepared and its DC electric field distribution was analyzed based on the measurement of electrical conductivity of PPLP in liquid nitrogen (LN 2 ). Electrical conductivity of PPLP in LN 2 has not been reported yet but it should be measured for DC electric field analysis. The experimental works for measuring electrical conductivity of PPLP in LN 2 were presented in this paper. Based on the experimental works, DC electric field distribution of PPLP specimen was fully analyzed considering the steady state and the transient state of DC. Consequently, it was possible to determine the electric field distribution characteristics considering different DC applying stages including DC switching on, DC switching off and polarity reversal conditions.

—Superconducting equipment has been actively investigated for securing the environment and energy technology (ET) in various parts of the world. Despite these movements, a high voltage cryogenic bushing, which plays an important role of interconnection between the electric power systems and superconducting devices, has not been fully developed due to severe insulation requirements. gas A insulated cryogenic bushing has been investigated as one of our projects since 2010. a basic step to obtain the design As parameters for cryogenic bushing, we focused on the surface flashover characteristics of glass fiber reinforced plastic (FRP) in several insulation gases. the danger of reduction of the dielectric performanceFor the surface flashover tests, several insulation gases including SF6, CF 4 and N 2 gas were prepared. Various length of FRP specimens were fabricated in order to obtain the data for fundamentalcreepage distance of FRP. The first specimen group was from 2 mm to 10 mm with 2 mm intervals and the second specimen group was from 20 mm to 100 mm with 20 mm intervals. And the gas pressure was varied from 1 bar to 4 bar. n AC Aovervoltage test and a lightning impulse test were performed. Then the experimental results of surface flashover were obtained and analyzedBased on these results, it would be . possible to design the optimum creepagedistance of FRP in a cryogenic bushing.

The aim of this paper is to perform fundamental simulation for design of 1.5 kV DC circuit breaker of high speed. The simulation was focused on methods for improvement of the arc driving force. Considering that design for arc runner is a main component which affects the arc driving force, two main design parameters of arc runner: slope of arc runner and electromagnetic force of the blow out coil, were simulated with variable shape, and then research was conducted on exploration of optimum arc runner shape with calculation of Lorentz force about each shape. The electromagnetic field analysis based on Finite Element Method (FEM) was performed by utilizing COMSOL Multiphysics. For analyses, a related model was designed for a practical 1.5 kV DC circuit breaker. In addition, transformative works of model was performed to conduct effective analysis on Lorentz force in the arc runner. As a result, arc is more effectively transferred and focused in driving part using core pin and electrode magnetic field which is formed as applying to Blow out coil. Furthermore, it was identified that the more the slope of arc runner increases, the more increase the density of Lorentz force and acceleration of arc at the same time. Additional researches and developments for advancement in driving force is required.

The purpose of this research is to determine the calibration constants of the wind speed measurement systems required to calculate the wind tunnel velocity in the test section. In the present work, the aerodynamic calibration tests of the test section were conducted in the Agency for Defense Development`s Low-Speed Wind Tunnel. The test speed ranged from 10 to 100 m/s with a reference pitot-static pressure probe. The validity of the calibration results was evaluated by comparing them with the previous calibration constants. The calibration results show that fair to good agreement is obtained with resonable accuracy.

An experimental study of the aerodynamic characteristics of a stealth configuration, the test techniques developed for the testing in the Low Speed Wind Tunnel of Agency for Defense Development(ADD-LSWT), and the lessons learned have been presented. The main objectives of this test are to determine the aerodynamic characteristics of a stealth configuration and to measure the flow field characteristics with a 5-hole pressure probe. The test results are discussed and the effect of the leading edge shape on the aerodynamic characteristics is also given.

Super grids are considered to be the technology of the future that will allow transmission and trade of huge volumes of electricity over long distances. HVDC is the preferred choice for developing super grids as HVAC transmission of bulk power over long distances is less efficient. However, the key obstacle in developing HVDC super grids is the absence of HVDC circuit breakers (DCCB) which can quickly detect and isolate the fault. The huge electrical and thermal stresses subjected to DCCB during DC current interruption are one of the primary challenges in developing DCCB. Especially, IGBT valves in hybrid DCCB are the most expensive and sensitive part and their cost drastically increase with their maximum rating. This research paper presents a comparative study on the electrical and thermal stress distribution across IGBT valves of conventional hybrid DCCB (CDCCB) and Superconducting hybrid DCCB (SDCCB) during fault conditions. Realistic CDCCB and SDCCB models were developed by considering the characteristic values of commercially available IGBT module. To create the electrical and thermal stress profiles, CDCCB and SDCCB were placed in HVDC test bed model and transient simulations were performed. The voltage distribution, current distribution and energy dissipation profiles across IGBT valves were developed for CDCCB and SDCCB. Comparative analysis of the profiles showed that the SDCCB components were subjected to relatively lower electrical and thermal stresses.

Owing to the type and structure of HVDC system, fault current characteristics can be varied. Until the present, most of studies have been dealt with fault current characteristic of HVDC system which focused on point-to-point connection. However, there were few announcements about the fault current characteristic of MTDC system. In this works, as a first step to determine the characteristics of circuit breaker for MTDC network, simulation works of fault current characteristics in MTDC system was investigated. Two types of MTDC system model were designed by Matlab/Simulink; Tree-like topology and Meshed topology. Pole to pole fault which is the severe fault in HVDC system was imposed and its characteristics was analyzed and compared. From the simulation result, the characteristics of fault current in MTDC system were analyzed and basic requirements for HVDC circuit breaker were suggested.

This paper describes the test carried out on an experimental study of fuselage drag and stability characteristics of a helicopter configuration and the test techniques developed for the testing and the lessons learned in the Agency for Defense Development Low Speed Wind Tunnel(ADD-LSWT). The main objective of this test is to determine the drag and stability characteristics of helicopter configurations according to the various configuration changes. The fuselage model with a highly modular structure is a representation of 1:8 scale of the external contour of the conceptual design helicopter configuration with rotating main rotor hub including blade stubs capable of rotating up to 500 rpm. The test results are compared with the available similar data and fair to good agreement is obtained.

Wind-tunnel tests were conducted to determine the optimum bypass duct area of the inertial particle separator for the aircraft inlet duct. When the area of the bypass duct for the air inlet is excessively increased, the amount of the dumped air to overboard is increased to result in the inlet performance degradation. In the present work, various tests are conducted on an inertial particle separator of the nacelle inlet model with bypass area variations. The tests utilize techniques involving design of experiments and response surface modeling. The effects of the bypass area variation on the bypass ratio of the inertial particle separator system and the performance of the inlet system are investigated. Experimental design optimization is then performed on the basis of two design factors: the bypass area and the engine air flow condition. The experimental results show that the bypass duct area to meet the bypass ratio requirement provided by the aircraft manufacturer can be determined to yield maximum performance on the basis of the regression model.

Dry-air is recognized as an attractive insulating medium for substituting SF 6 gas in high voltage devices. Recently, Japanese manufacture announced that 72.5 kV has been successfully developed using dry-air as a main insulating medium. However, only a few fundamental research works of dry-air have been reported compared to those of SF 6 . In this paper, two kinds of experiments were conducted to clarify the breakdown characteristics of dry-air. First, the effect of humidity on the breakdown characteristics of dry-air was deeply investigated. And secondly, the effect of electrode roughness on the breakdown voltage of dry-air has been investigated. In order to make artificial specific humidity of dry-air, the experiment chamber has been controlled in vacuum condition under 0.2 Torr and some water was injected before filling the dry-air. Dry-air used for the experiments was composed of about 21% Oxygen and 79% nitrogen and its humidity was controlled of ‒85°C dewpoint temperature. Water could be successfully vaporized in dry-air, and then AC withstand test and lightning impulse breakdown test were performed using sphere to plane electrode airgap configurations. Next, similar breakdown tests were performed considering polished condition, and artificially made surface roughness with 16 μm and 25 μm on the high voltage electrode respectively. From the test results, it was shown that impulse breakdown voltages were slightly decreased with increasing electrode roughness. However, AC withstand voltages did not show any remarkable difference regarding surface roughness, and it was revealed that the breakdown voltage of dry-air was not affected by different humidity condition ranging from 300 PPM to 1,500 PPM humidity in dry-air. AC withstand voltage and negative impulse voltage in this humidity range did not show noticeable difference.

Since the discovery of high-temperature superconductor (HTS), liquid nitrogen(LN 2 ) has not been only utilized as a coolant of superconducting electric equipment but also as an insulation material in cryogenic environment due to its dielectric performance. It also has a lot of advantages over other cryogenic liquid such as less expense and harmless substance, thus it has been widely used in the development of superconducting devices. Up to now, a lot of research works dealing with the breakdown characteristics of LN 2 for distribution-class superconducting devices have been presented worldwide but, few research works about breakdown characteristics of liquid nitrogen in extra high voltage class have been reported due to the limitation of cryogenic test facilities in extra high voltage (EHV) class. In order to study the cryogenic EHV insulation technologies, we have built the cryogenic dielectric test facilities including a fiber reinforced plastic (FRP) big cryostat with cryogenic bushing, a 400 kV AC overvoltage and a 1.6 MV lightning impulse test systems. Using these facilities, we focused on the breakdown characteristics of liquid nitrogen in EHV level which is rather different comparing to the distribution level. With real scale big cryostat, AC overvoltage test and impulse tests have been performed. From the test results, the breakdown characteristics of liquid nitrogen in EHV were suggested. And these test results could be used as basic insulation design data to develop transmission-class superconducting electric equipment.

In a smart grid, various kinds of distributed generation (DG) sources could be connected into the main power grid in order to enhance the reliability of the power system. The combination of ac and dc distribution grid are also considered for the efficient connection of renewable power resources. In this case, one of the critical problems due to these integrations is the excessive increase in the fault current because of the presence of DG within the smart grid. In order to protect the smart grid from increasing fault current, a superconducting fault current limiter (SFCL) could be applied, which has negligible power loss and capability to limit initial fault currents effectively. This paper presents feasibility analysis results of the positioning of the SFCL and its effects on reducing fault current in a smart grid having ac and dc microgrid. The detailed power system was implemented with a microgrid having wind farm and low voltage dc grid connected with a photovoltaic farm. Transient analyses were performed for the worst case faults with the different SFCL arrangements. The strategic location of SFCL in the power grid, which could limit fault currents and has no negative effect on the DG sources, was found to be the connection point of integration of the each DG sources in the ac and dc microgrid.