Control of a DC Microgrid
Marko Gulin
University of Zagreb, Faculty of Electrical Engineering and Computing Department of Control and Computer Engineering
Unska 3, 10000 Zagreb, Croatia marko.gulin@fer.hr
Abstract—A microgrid is a part of a distribution network embedding multiple distributed generation systems (mostly non-conventional renewable energy sources like photovoltaic panels, small wind turbines etc.) and storage systems with local loads, which can be disconnected from the upstream network under emergency conditions or as planned. The microgrid concept naturally arose to cope with the penetration of renewable energy sources, which can be realistic if the final user is able to generate, store, control and manage part of the energy that it will consume. The power connection between microgrid components can be done through a direct current (DC) link or an alternating current (AC) link. In this paper we describe operation modes (grid-connected, islanded) and control methods of a DC microgrid.
Index Terms—Distributed generation and storage, DC micro-grid, Microgrid operation modes, Active load sharing, Droop control methods
I. INTRODUCTION
Around the world, conventional fossil-fuelled power system is facing problems of gradual depletion of fossil fuel resources, poor energy efficiency and environmental pollution [1]. These problems have led to a new trend of generating power locally at distribution voltage level by using small-scale conventional biomass-fuelled energy sources like gas and diesel micro-turbines, together with non-conventional renewable energy sources like photovoltaic panels, wind turbines etc., and other non-conventional sources like fuel cells. This type of power generation is termed as distributed generation and the involved energy sources are termed as distributed generation sources. Distributed generation can offer considerable social and eco-nomic benefits, including reduced power network losses and the exploitation of renewable energy resources [2].
The integration of renewable energy sources poses a chal-lenge because their output is intermittent and variable and in principle requires an energy storage to enable time-shift between energy production and consumption. If only one re-newable energy source is considered, the integration is simple – for stand-alone use the source is connected with a storage and load, while in the grid-connected case the source injects the power directly into the power network, whereas the issues of power balancing are left to be handled by distribution and/or transmission system operators. Considerable improvements may however be achieved when heterogeneous distributed energy sources are used in a bulk for local or grid power supply, like easier grid integration and smart power manage-ment, with benefits both locally and on the grid-side. The required power and information communication infrastructure
to enable it is called a microgrid. Most commonly used energy storage devices in a microgrid are batteries, supercapacitors, flywheels, and fuel cells with electrolyser (EL). This type of energy storage is termed as distributed storage and the energy storage devices are termed as distributed storage devices.
Microgrid is defined as a cluster of distributed generation sources, distributed storage devices and distributed loads that operate so as to improve the reliability and quality of the local power supply and of the power system in a controlled manner [3]. The microgrid concept naturally arose to cope with the penetration of renewable energy sources, which can be realistic if the final user is able to generate, store, control and manage part of the energy that it will consume [4]. The power connection between microgrid components, i.e. distributed generation sources, storages and loads, can be done through a direct current (DC) link or an alternating current (AC) link. In this paper a DC link microgrid is considered, with emphasis on its control and power management in grid-connected and islanded operation mode.
Microgrid control must insure that: (i) new distributed generation and storage systems can be added or removed from the microgrid seamlessly, (ii) equal and stable current sharing between parallel power converters (i.e. sources) is enabled, (iii) output voltage fluctuations can be corrected, and (iv) desired power flow from/to the microgrid together with technically and economically viable operation is enabled. There is a fairly large number of methods for paralleling power converters (PCs). From the viewpoint of the operating mechanism to current sharing and output voltage level management, control methods are classified into two basic categories: (i) active load sharing, and (ii) droop control methods.
The report is structured as follows. In Section II a micro-grid concept is introduced. In Section III a micromicro-grid power management in grid-connected and islanded operation mode is described. In Section IV a commonly used DC microgrid control methods are described.
II. AMICROGRID CONCEPT
Due to the ever-increasing demand for high-quality and reliable electric power, the concept of distributed generation and energy storage has attracted widespread attention in recent years. Distributed generation and storage systems consist of relatively small-scale generation and energy storage devices that are interfaced with low- or medium-voltage distribution networks through power converters and can offset the local
power consumption, or even export power to the upstream network if their generation surpasses the local consumption. An upcoming philosophy of operation which is expected to enhance the utilization of distributed generation and energy storage is known as the microgrid concept.
The main benefits of microgrids are high energy efficiency, high quality and reliability of the delivered electric power, more flexible power network operation, and environmental and economical benefits [5]. However, to achieve a stable and secure operation, a number of technical, regulatory and economic issues have to be resolved before microgrids can become commonplace. In this paper we deal with control methods for integrating distributed generation and energy stor-age systems into a microgrid, as well with power manstor-agement in grid-connected and islanded operation mode.
The main components of a microgrid are: (i) distributed generation sources such as photovoltaic panels, small wind turbines, fuel cells, diesel and gas microturbines etc., (ii) distributed energy storage devices such as batteries, super-capacitors, flywheels etc., and (iii) critical and non-critical loads. Energy storage devices are employed to compensate for the power shortage or surplus within the microgrid. They also prevent transient instability of the microgrid by providing power in transient. The transient power shortage in a microgrid can be compensated for by fast energy storage devices in the microgrid, or by the utility grid through a bidirectional power converter when operating in grid-connected mode.
The issue of the power quality in microgrids is an im-portant issue due to the presence of an appreciable number of sensitive loads whose performance and lifespan can be adversely affected by voltage sags, harmonics and imbalances. In a microgrid, most distributed generation sources and storage devices employ power converters which can rapidly correct indicated imperfections, even in the presence of nonlinear and unbalanced loads [5]. The selection of an appropriate power converter mainly depends on the generation source and storage device type, and on the used power connection between the microgrid components.
The power connection between microgrid components can be done through a DC link or an AC link. Many non-conventional energy sources generate low-voltage DC power, e.g. photovoltaic panels, fuel cells etc. Most of these sources supply power to an AC utility grid and require costly and inefficient power converters, even where the power may ultimately be delivered to a DC device. However, power transmission through a low-voltage DC link produces more losses than transmission through a high-voltage AC link. With development of a microgrid control methods along with cost-effective and efficient power converters, a DC link microgrid can become a promising solution for integrating distributed generation sources, storages and loads. Adding intelligence to a DC microgrid controllers further enables consumer engagement with utility grid through smart metering and ultimately with dynamic demand management, and this could reduce costs associated with periods of high and low power consumption.
Figure 1 shows the schematic diagram of a microgrid which embeds (i) distributed generation sources such as photovoltaic panels, small wind turbine, and fuel cells, (ii) distributed storage devices such as batteries, supercapacitors and flywheel, and (iii) distributed loads. Each distributed generation source and storage device is interfaced with a common link through a power converter. The microgrid is galvanically isolated from the utility grid and can be easily disconnected from the grid through the main switch for maintenance purposes. In a case of power shortage that can occur when utility grid is not available, non-critical loads can be disconnected from the microgrid through an emergency switch. Worth noting, microgrid can also embed combined heat and power systems that exploit waste heat for domestic purposes where heat flows can be managed in addition to electrical energy flows.
Batteries Flywheel
PV panels Wind turbine Fuel cells H2 tank
Utility grid
PC PC PC
Distributed generation systems
PC PC
PC
Distributed storage systems
EL PC Distributed loads DC LINK PC FER building
Fig. 1. Schematic diagram of a DC microgrid
III. MICROGRID OPERATION MODES
A microgrid is connected into the utility grid through a bidirectional power converter, that continuously monitors both sides and manages power flow between them. If there is a fault in the utility grid, the power converter will disconnect the microgrid from the grid, creating an islanded energy system. The microgrid can continue to operate in the islanded mode, that is primarily intended to enhance system reliability and service continuity, and it is typically unplanned. However, it can also be introduced intentionally for maintenance purposes through the main switch. In some cases, islanded operation is the only mode of operation, e.g. in off-grid remote electrifica-tion system. Concludingly, there are two operaelectrifica-tion modes for a microgrid: (i) grid-connected, and (ii) islanded mode.
Consider a DC microgrid that consists of (i) distributed generation sources such as photovoltaic panels, wind turbine and fuel cells stack with electrolyser, (ii) distributed storage devices such as batteries and supercapacitors, and (iii) critical and non-critical loads, all connected in parallel into the com-mon DC link through corresponding power converters. The power flow of the systems in the considered DC microgrid is shown in Figure 2.
PV WT FC Generation systems
BAT SC
Storage systems Critical loads Non-critical loads HT Bidirectional power converter AC grid EL PLOAD PGRID H2
Fig. 2. A DC microgrid power flow
The sum of the output power of the photovoltaic panels, the wind turbine and the fuel cells, i.e. distributed generation sources, is defined as:
PDG =PP V +PW T +PF C, (1) where PP V, PW T and PF C are photovoltaic panels, wind turbine and fuel cells output power.
The distributed generation systems supply unidirectional power to the DC microgrid and play a role as the main energy source. Since energy storage devices control the power balance of a DC microgrid by charge and discharge, the power flow is bidirectional and the reference power for energy storage devices is defined as:
PDS=PBAT+PSC+PEL=PGRID+PDG−PLOAD, (2) wherePBAT andPSC are batteries and supercapacitors charg-ing power, PELis the electrolyser power,PLOAD is required power of all loads connected into the DC microgrid, critical and non-critical, and PGRID is the utility grid power.
The loads are assumed to demand unidirectional power from the microgrid. According to a varying local demand, the distributed storage systems realize a power balance, and thus make a continuous high-quality power supply to the load possible [6]. In a case of power shortage that can occur when utility grid is not available, non-critical loads can be disconnected from the microgrid.
In the following subsections, a simple algorithm of power management for DC microgrid is described. It must be noted that this is not the only option power management. In future work, the problem of the optimal power management will be handeled by the control algorithm.
A. Grid-connected mode
In the grid-connected operation mode, the grid-tied power converter has control over the DC link voltage level. If the output sum of the power of the distributed generation systems is sufficient to charge the storage devices, any excessive power is supplied to the utility grid. If the sum of the power of the distributed generation and storage systems is deficient with respect to the load demand, the required power is supplied from the utility grid. In the grid-connected mode, power man-agement is performed in a complementary manner between storage devices and as a result a DC microgrid can operate safely and efficiently.
B. Islanded mode
When a DC microgrid must be separated from the utility grid and switch to the islanded mode, the grid-tied power converter releases control of the DC link voltage level, and one of the converters in the microgrid must take over that control. Since each converter of distributed generation sources is used for optimal control of its belonging source, only the converters of the energy storage elements are free to regulate the DC link voltage level. During the islanded mode, the battery plays the main role in regulating the DC link voltage level, and the supercapacitor plays a secondary role in responding of the sudden power requirement as an auxiliary source/sag, i.e. for peak shaving during transients.
IV. MICROGRID CONTROL METHODS
Microgrid control must insure that: (i) new distributed generation and storage systems can be added or removed from the microgrid seamlessly, (ii) equal and stable current sharing between parallel power converters (i.e. sources) is enabled, (iii) output voltage fluctuations can be corrected, and (iv) desired power flow from/to the microgrid together with technically and economically viable operation is enabled. There is a fairly large number of methods for paralleling power converters. From the viewpoint of the operating mechanism to current sharing and output voltage level management, control methods are classified into two basic categories: (i) active load sharing, and (ii) droop control methods. It is also possible to design a hybrid control method combining good aspects of active load sharing and droop control method, but this will not be further discussed. A microgrid control is often implemented in a hierarchical manner, with three control loops: (i) tertiary loop manages the power flow from/to the microgrid, (ii) secondary loop corrects output voltage fluctuations, and (iii) primary loop performs current sharing control between power converters.
Figure 3 shows the equivalent circuit of two DC power supplies connected in parallel sharing a common load through resistive output impedances. If there is some voltage difference between sources, this will circulate a current between DC sources, and in order to reduce the circulating current a primary control loop is applied.
+ − V1 I1 R1 R2 + − V2 I2 IL Vo
Fig. 3. Two parallel-connected DC power supplies
The output voltage Vo, i.e. the DC link voltage, can be expressed as: Vo Rp = V1 R1 + V2 R2 − IL, (3)
where IL is the total load current, and Rp is the parallel resistance defined as:
1 Rp = 1 R1 + 1 R2 . (4)
In a case of sudden rise of the load current IL, the output voltageVo drops. In order to restore its nominal voltage level, a secondary control loop is applied.
V. ACTIVE LOAD SHARING
The first category of control methods, named the ac-tive load-sharing technique [7]–[26], need intercommunication link. Although these links limit the flexibility of the microgrid and degrade its redundancy, both tight current sharing and low-output-voltage fluctuations can be achieved. The following section provides a review of the existing active load sharing control methods for parallel converters available in literature [4]. The active load sharing control methods can be classified into three different types: (i) centralized control, (ii) master-slave control (MS), and (iii) circular chain control (3C). A. Centralized control
The centralized control, shown in Figure 4(a), consists of dividing the total load current iL by the number of modules (MODs) N, so that this value becomes the current reference i∗k of each module k:
i∗k= iL
N, k= 1,2, . . . , N. (5) The current reference value is subtracted by the current of each module, obtaining the current error ∆ik, which is processed through a current control loop (CL). An outer control loop in the centralized control, i.e. voltage control loop (VL), adjusts the load voltage. Using this approach, it is necessary to measure the total load current iL, so it cannot be used in a large distributed systems. Consequently, a central control board (CCB) is necessary.
B. Master-slave control
In the master-slave control, the master module regulates output voltage. Hence, the master currentimfixes the current references of the rest of the modules (slaves) as:
i∗s=im, s= 2, . . . , N. (6) Consequently, as shown in Figure 4(b), the master acts as a voltage source converter, whereas the slave works as a current source converter. If the master unit fails, another module will take the role of master in order to avoid the overall failure of the system. There exist different variants of this control method, depending on the role of the master: (i) dedicated, where the master is one fix module, (ii) rotary, where the master is arbitrarily chosen, and (iii) high-crest current, where the module that brings the maximum current automatically becomes the master.
C. Circular chain control
The circular chain control, shown in Figure 4(c), consists of the current reference of each module taken from the other module, forming a control ring. Note that the current reference of the first unit is obtained from that of the last unit to form a circular chain information. This strategy can be expressed as:
i∗k=
iN, k= 1, ik−1, k= 2, . . . , N.
(7) The current limitation control is a variant of the circular chain control. In this case, the load voltage is controlled by the master module, whereas the slave modules are only for sharing the load current. Except for the master module, the current command of the slave is generated by its previous module and limited in amplitude. In this scheme, any module can be the master (dedicated, rotating, high-crest current).
MOD 1 CL MOD 2 CL MOD N CL CCB Load i∗1 i1 i∗N iNiloadv o vref i∗2 i2
(a) Centralized control of a DC microgrid
MOD M CL MOD S2 CL MOD SN CL Load vo VL vref im i∗s is2 isN i∗s
(b) Master-slave control of a DC microgrid
MOD 1 CL MOD 2 CL MOD N CL Load i∗1 i1 i∗N iN i∗2 i2
(c) Circular chain control of a DC microgrid Fig. 4. Active load sharing control methods
VI. DROOP CONTROL METHOD
The second category of control methods, named the droop control method [27]–[34], is able to avoid critical communi-cation links. The absence of critical communicommuni-cations between the modules improves the reliability without restricting the physical location of the modules. The droop method is based on a well-known concept in large-scale power systems, which consists of drooping the frequency of the AC generator when its output power increases [4]. The droop method achieves higher reliability and flexibility in the physical location of the modules since it only uses local power measurements.
A. Virtual output impedance
In the virtual output impedance control, shown in Figure 5, current at the module output is sensed and sent back to the module input via virtual impedance RD, where is compared with the output voltage reference at no load:
vo∗=vref−ioRD, (8) whereiois the module output current,RDis the virtual output impedance, andvref is the output voltage reference at no load.
Load CL vref VL MOD 1 vo R D δvo io1 ioN vM G v∗M G CL vref VL MOD N vo R D δvo VL
Fig. 5. Droop control of a DC microgrid via virtual output impedance
This control loop has the inherent load-dependent voltage deviation. To solve the problem of the voltage deviation, the voltage level in the microgrid vM G is sensed and compared with the voltage reference vM G∗ , and the error processed through a compensator is sent to all the modules to restore the output voltage. The controller can be expressed as follows:
δvo=kpeV +ki Z
eVdt, (9a)
eV =v∗M G−vM G, (9b) where kp andki are the control parameters of the microgrid voltage level compensator. Finally, equation (8) becomes:
v∗o=vref +δvo−ioRD. (10) B. Series resistor
In the series resistor control, a resistor is placed in series with the module output to provide a voltage drop in the output. In this control method, all of the paralleled modules have an initial setting that, via a potentiometer, are made almost identical. Obviously, the major disadvantage of this approach is the high power dissipation in the series resistor if the droop in output voltage is large. Because of added power dissipation, this method is used only for low-power linear post-regulators [35]. Microgrid voltage level deviation is corrected in the same way as in virtual output impedance control method.
VII. CONCLUSION
A microgrid is a part of a distribution network embed-ding multiple distributed generation systems (mostly non-conventional renewable energy sources like photovoltaic pan-els, small wind turbines etc.) and storage systems with local loads, which can be disconnected from the upstream network under emergency conditions or as planned. The microgrid con-cept naturally arose to cope with the penetration of renewable energy sources, which can be realistic if the final user is able
to generate, store, control and manage part of the energy that it will consume. The power connection between microgrid components can be done through a direct current (DC) link or an alternating current (AC) link.
In this paper we describe operation modes and control methods of a DC microgrid. A microgrid can operate in a grid-connected mode or in an islanded operation mode. From the viewpoint of the operating mechanism to current sharing and output voltage management, control methods can be classified as an active load sharing and droop methods. The main differ-ence between aforementioned control methods is that droop control methods do not require fast communication between components (i.e. generation sources and storage devices), thus improving system reliability and flexibility at the cost of the DC link voltage level stability.
REFERENCES
[1] Microgrids and Active Distribution Networks. IET Renewable Energy Series, The Institution of Engineering and Technology, 2009. [2] N. Jenkins and R. Allan and P. Crossley and D. Kirschen and G. Strbac,
Embedded Generation. IET Power and Energy Series, The Institution of Engineering and Technology, 2000.
[3] F. A. Farret and M. G. Sim˜oes,Integration of Alternative Sources of Energy. John Wiley & Sons Inc., 2006.
[4] J. M. Guerrero and L. Hang and J. Uceda, “Control of Distributed Un-interruptible Power Supply Systems,”IEEE Transactions on Industrial Electronics, vol. 55, no. 8, pp. 2845–2859, 2008.
[5] N. Hatziargyriou and H. Asano and R. Iravani and C. Marnay, “Mi-crogrids,”IEEE Power & Energy Magazine, vol. 5, no. 4, pp. 78–94, 2007.
[6] J. H. Lee and H. J. Kim and B. M. Han and Y. S. Jeong and H. S. Yang and H. J. Cha, “DC Micro-grid Operational Analysis with a Detailed Simulation Model for Distributed Generation,” Journal of Power Electronics, vol. 11, no. 3, pp. 350–359, 2011.
[7] A. P. Martins and A. S. Carvalho and A. S. Araujo, “Design and Imple-mentation of a Current Controller for the Parallel Operation of Standard UPSs,” in21st Internation Conference on Industrial Electronics, Control and Instrumentation, pp. 584–589, 1995.
[8] T. Iwade and S. Komiyama and Y. Tanimura, “A novel small-scale UPS using a parallel redundant operation system,” in Telecommunications Energy Conference, pp. 480–484, 2003.
[9] J. Holtz and W. Lotzkat and K. H. Werner, “A high-power multitransistor-inverter uninterruptible power supply system,” IEEE Transactions on Power Electronics, vol. 3, no. 3, pp. 278–285, 1988. [10] J. Holtz and K. H. Werner, “Multi-inverter UPS system with redundant
load sharing control,” IEEE Transactions on Industrial Electronics, vol. 37, no. 6, pp. 506–513, 1990.
[11] H. van der Broeck and U. Boeke, “A simple method for parallel opera-tion of inverters,” inTelecommunications Energy Conference, pp. 143– 150, 1998.
[12] C. S. Lee and S. Kim and C. B. Kim and S. C. Hong and J. S. Yoo and S. W. Kim and C. H. Kim and S. H. Who and S. Y. Sun, “Parallel UPS with a instantaneous current sharing control,” in24th Annual Conference of the IEEE Industrial Electronics Society, pp. 568–573, 1998. [13] Y. Pei and G. Jiang and X. Yang and Z. Wang, “Auto-master-slave
control technique of parallel inverters in distributed AC power systems and UPS,” inPower Electronics Specialists Conference, pp. 2050–2053, 2004.
[14] W.-C. Lee and T.-K. Lee and S.-H. Lee K.-H. Kim and D.-S. Hyun and I.-Y. Suh, “A master and slave control strategy for parallel operation of three-phase UPS systems with different ratings,” inApplied Power Electronics Conference and Exposition, pp. 456–462, 2004.
[15] J. F. Chen and C.-L. Chu, “Combination voltage-controlled and current-controlled PWM PWM inverters for UPS parallel operation,” IEEE Transactions on Power Electronics, vol. 10, no. 5, pp. 547–558, 1995. [16] S. Tamai and M. Kinoshita, “Parallel operation of digital controlled UPS
system,” inIndustrial Electronics, Control and Instrumentation, pp. 326– 331, 1991.
[17] Y. J. Cheng and E. K. K. Sng, “A novel communication strategy for decentralized control of paralleled multi-inverter systems,” IEEE Transactions on Power Electronics, vol. 21, no. 1, pp. 148–156, 2006. [18] X. Sun and Y.-S. Lee and D. Xu, “Modeling, analysis, and
implementa-tion of parallel multi-inverter system with instantaneous average-current sharing scheme,” IEEE Transactions on Power Electronics, vol. 18, no. 3, pp. 844–856, 2003.
[19] Y. Xing and L. Huang and S. Sun and Y. Yan, “Novel control for redundant parallel UPSs with instantaneous current sharing,” inPower Conversion Conference, pp. 959–963, 2002.
[20] Y.-K. Chen and T.-F. Wu and Y.-E. Wu and C.-P. Ku, “A current sharing control strategy for paralleled multi-inverter systems using microprocessor-based robust control,” inElectrical and Electronic Tech-nology, pp. 647–653, 2001.
[21] Z. He and Y. Xing and Y. Hu, “Low cost compound current sharing con-trol for inverters in parallel operation,” inPower Electronics Specialists Conference, pp. 222–227, 2004.
[22] H.-M. Hsieh and T.-F. Wu and H.-S. Nien and Y.-E. Wu and Y.-K. Chen, “A compensation strategy for parallel inverters to achieve precise weighting current distribution,” in Industry Application Conference, pp. 954–960, 2005.
[23] H. Oshima and Y. Miyakawa and A. Hirata, “Parallel redundant UPS with instantaneous PWM control,” in13th International Telecommuni-cations Energy Conference, pp. 436–442, 1991.
[24] T.-F. Wu and Y.-K. Chen and Y.-H. Huang, “3C strategy for inverters in parallel operation achieving an equal current distribution,” IEEE Transactions on Industrial Electronics, vol. 47, no. 2, pp. 273–281, 2000. [25] Y.-K. Chen and Y.-E. Wu and C.-P. Ku, “ACSS for paralleled multi-inverter systems with DSP-based robust controls,”IEEE Transactions on Aerospace and Electronic Systems, vol. 39, no. 3, pp. 1002–1015, 2003.
[26] S. Sun and L.-K. Wong and Y.-S. Lee and D. Xu, “Design and analysis of an optimal controller for parallel multi-inverter systems,” IEEE Transactions on Circuits and Systems II: Express Briefs, vol. 52, no. 1, pp. 56–61, 2006.
[27] J. M. Guerrero and J. C. Vasquez and J. Matas and L. G. de Vicu˜na and M. Castilla, “Hierarchical Control of Droop-Controlled AC and DC Microgrids — A General Approach Toward Standardization,”IEEE Transactions on Industrial Electronics, vol. 58, no. 1, pp. 158–172, 2011. [28] P. Kundur,Power System Stability and Control. McGraw-Hill, 1994. [29] J. M. Guerrero and J. C. Vasquez and J. Matas and M. Castilla and
L. G. de Vicu˜na, “Control Strategy for Flexible Microgrid Based on Parallel Line-Interactive UPS Systems,”IEEE Transactions on Industrial Electronics, vol. 56, no. 3, pp. 726–736, 2009.
[30] H. Kakigano and Y. Miura and T. Ise and R. Uchida, “DC Voltage Control of the DC Micro-grid for Super High Quality Distribution,” in
Power Conversion Conference - Nagoya, pp. 518–525, 2007. [31] Z. Ye and D. Boroyevich and K. Xing and F. C. Lee, “Design of parallel
sources in DC distributed power systems by using gain-scheduling technique,” in30th Power Electronics Specialists Conference, pp. 161– 165, 1999.
[32] R. Nilsen and I. Sorfonn, “Hybrid Power Generation Systems,” in13th European Conference on Power Electronics and Applications, pp. 1–9, 2009.
[33] Z. Jiang and X. Yu, “Hybrid DC- and AC-Linked Microgrids: Towards Integration of Distributed Energy Resources,” inEnergy 2030 Confer-ence, pp. 1–8, 2008.
[34] , “DC-Bus Signaling: A Distributed Control Strategy for a Hybrid Renewable Nanogrid,” IEEE Transactions on Industrial Electronics, vol. 53, no. 5, pp. 1453–1460, 2006.
[35] S. Luo and Z. Ye and R. L. Lin and F. C. Lee, “A Classification and Evaluation of Paralleling Methods for Power Supply Modules,” inPower Electronics Specialists Conference, 1999. PESC 99., pp. 901–908, 1999.
