IEEE TRANSACTIONS ON POWER ELECTRONICS, VOL. 30, NO. 2, FEBRUARY 2015 603 Networked-Based Hybrid Distributed Power Sharing and Control for Islanded Microgrid Systems Alireza Kahrobaeian, Student Member, IEEE, and Yasser Abdel-Rady Ibrahim Mohamed, Senior Member, IEEE Abstract—Distributed generation (DG) microgrid systems are forming the building blocks for smart distribution grids. Enhanced networked-based control structure is needed not only to eliminate the frequency deviations, power-sharing errors, and stability con- cerns associated with conventional droop control in microgrids but also to yield: 1) improved microgrid dynamic performance, 2) minimized active/reactive power-sharing errors under unknown line impedances, and 3) high reliability and robustness against net- work failures or communication delays. This paper proposes a new hybrid distributed networked-based power control scheme that ad- dresses the aforementioned problems in a distributed manner. The new method consists of a set of distributed power regulators that are located at each DG unit to ensure perfect tracking of the op- timized set points assigned by the centralized energy management unit (EMU). The average power measurements are transmitted to the EMU to calculate the share of each unit of the total power demand based on real-time optimization criteria; therefore, a low- bandwidth communication system can be used. In the proposed method, the distributed nature of the power regulators allows them to adopt the delay-free local power measurements as the required feedback signals. Therefore, the proposed structure pro- vides great robustness against communication delays. Further, this paper presents a generalized and computationally efficient model- ing approach that captures the dominant dynamics of a microgrid system. The model can be used to study the impact of power-sharing controllers and delays in microgrid stability. Comparative simu- lation and experimental results are presented to show the validity and effectiveness of the proposed controller. Index Terms—Distributed generation (DG), energy manage- ment, microgrids, networked control systems, stability. I. INTRODUCTION D RIVEN by the economic, environmental, and technical reasons, the energy sector is showing increasing interest in adopting smart grid technologies (e.g., advanced communica- tion, control, protection, and monitoring algorithms) to improve the efficiency and reliability of future power grids [1]. In partic- ular, distributed generation (DG) microgrid systems are forming the building blocks for smart distribution grids. This vision is in line with the recently developed IEEE Std. 1547.6, which pro- poses microgrid clusters as building blocks of future distribution systems [2]. In this paradigm, networked control of microgrids is essential to optimize the microgrid performance in real-time, Manuscript received October 15, 2013; revised January 28, 2014; accepted March 10, 2014. Date of publication March 19, 2014; date of current version October 7, 2014. Recommended for publication by Associate Editor Y. Sozer. The authors are with the Department of Electrical and Computer Engi- neering, University of Alberta, Edmonton, AB T6G 2V4, Canada (e-mail: kahrobae@ualberta.ca; yasser_rady@ieee.org). Color versions of one or more of the figures in this paper are available online at http://ieeexplore.ieee.org. Digital Object Identifier 10.1109/TPEL.2014.2312425 particularly under high penetration level of DG resources. An energy management controller can be implemented in the upper hierarchy of the microgrid system to optimize the energy flow in the microgrid system by controlling the set points of DG units and other controllable devices (e.g., energy storage devices, loads und the demand-side management program and reactive power sources). Wide band of energy management objectives can be considered in a typical microgrid depending on customer, and utility needs. For example, these objectives include mini- mizing the fuel cost in a microgrid, minimizing the emission and maximizing the network security [3]–[7]. The utilization of an energy management unit (EMU) brings more intelligence and efficiency to the microgrid system where more flexible opera- tion is needed [8]. This is also in line with recent efforts carried out to adopt communication-based networked control systems in power systems [9]–[14]. The majority of microgrid control structures available in the literature are based on autonomous droop controllers, which eliminate the communication requirements between individual DG units. However, in droop-based microgrids the coupling between active and reactive power dynamics can compromise system stability especially when the ratio of the line resistance to its reactance is considerable. This problem can be exacerbated if the power dynamics are to be enhanced by increasing the droop gains. Although cost-optimization algorithms can be applied in a droop-based control system via changing their droop gain ratios in order to manage how the power demand is shared among the DG units; however as suggested in [12], system stability is highly sensitive to the selected droop gains and one should consider this criteria in an online/off-line manner prior to adopt any energy management strategy. Furthermore, in a system with considerable X/R ratio, the conventional autonomous droop-based control is based on drooping the frequency and voltage in order to share the active and reactive powers, respectively. However, this will cause the frequency and voltage levels to drop from their nominal values and frequency/voltage chattering under variable load demand. Therefore, networked-based secondary control is recently proposed to remove the frequency deviations [13]. However, the secondary control does not provide a framework to optimize power sharing among the DG units and hence tertiary con- trollers are often needed for global cost-optimization purposes. Moreover, the traditional droop control fails to share the reactive power accurately among the units. A communication-based inverter control scheme has been proposed in [14] where the system stability is enhanced through supplementary control signals. The proposed method eliminates the inaccurate reactive power-sharing problem when the communication is adopted. 0885-8993 © 2014 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications standards/publications/rights/index.html for more information.