Decentralized Voltage Control to Minimize Distribution Losses in an Islanded Microgrid

Microgrids can bring electricity power to rural communities or isolated military forward operation bases. For many rural areas, renewable powers could be the cheapest power sources available. Utilizing renewable power sources and energy storage systems such as batteries requires new power and voltage control strategies. In addition, these microgrids may be reconfigured frequently. Therefore, the control strategies must be implemented in a plug-and-play fashion at individual nodes, and ideally they should not require much communication with neighboring nodes. Another issue of microgrids is the resistance loss in distribution lines due to the lower operating voltage. To reduce power losses, voltage control is required which again must work in a plug-and-play fashion. In this paper, we propose a decentralized real-time voltage control algorithm that minimizes power losses for a microgrid supported by inverter based distributed power sources. Its optimality and plug-and-play nature are demonstrated through simulations. INTRODUCTION Distributed renewable power sources are increasingly used because of the push for low carbon electricity. They were treated as “negative loads” in the past but will need to be treated as intermittent supplies when they provide a significant portion of the grid power. The intermittency can cause grid instability if inadequate reserve and regulation services are not present. The intermittency problem is more profound for microgrids relying on renewable power sources [1-4]. For military applications, the microgrid concept is especially appealing considering energy security and independence. In this paper, we will focus on microgrids that are islanded, i.e., not connected to the mega grid. In these microgrids, it is desirable to deploy renewable energy sources such as solar and wind to reduce reliance on fossil fuels. In addition, electrified vehicles can play an important role using vehicle-to-grid (V2G) technologies [5-7]. For these military applications, it makes sense to use the batteries and engines of electrified vehicles to support the microgrid and explore their synergistic design and usage for improved overall grid efficiency. Islanded microgrids have much smaller inertia than the conventional grids. For distributed power sources supported by inverters, there may not exist any inertia at all. In such cases, regulating the grid frequency becomes challenging and a number of control strategies have been studied [8-11]. Another important attribute of islanded microgrids is that the lower operation voltage results in high distribution losses than on high voltage grids. Voltage or reactive power allocation over the microgrid network can be used to reduce distribution losses. In conventional grid systems, voltage/reactive power allocation is determined through optimization using the knowledge of the grid structure and operating conditions [1215]. A few approaches were suggested to determine the optimal voltage profiles in real-time, but they require the knowledge of the grid structure [16] and typically require a centralized implementation [17]. For microgrids, the control strategy must be plug-and-play because the frequent reconfiguration. In addition, it is desirable to develop a control algorithm that requires limited communications and measurements. New load and supply nodes could be added, i.e., the grid configuration can change and may not be known to the control algorithm. The main contribution of this paper is a modelASME 2012 5th Annual Dynamic Systems and Control Conference joint with the JSME 2012 11th Motion and Vibration Conference DSCC2012-MOVIC2012 1 Copyright © 2012 by ASME DSCC2012-MOVIC2012-8539 October 17-19, 2012, Fort Lauderdale, Florida, USA Downloaded From: http://asmedigitalcollection.asme.org/ on 01/29/2014 Terms of Use: http://asme.org/terms free, distributed voltage control algorithm for islanded microgrids that minimizes distribution losses. This “plug-and-play” control algorithm consists of two levels. The low level control regulates the power output and the terminal voltage to follow the set values that are determined by a high level controller. It is designed based on inverter and phase-locked loop (PLL) systems. The high level controller is designed using a cost function on distribution power losses. We derive a condition that guarantees the power loss minimization without requiring central coordination. DECENTRALIZED CONTROL DESIGN The structure of the microgrid supported by distributed power sources is shown in Figure 1, where a communication network enables small amount of information sharing. The DC power sources can be solar panels, batteries, or power from wind turbines that has been converted to DC. The detailed view of the controller is shown in Figure 2 where the controller consists of a high level controller, a low level controller, a phase locked loop, and a measurement and computation block. Low-Level Controllers The low level controllers consist of servo loops for controlling an inverter. In this section, we describe the inverter and controllers both located in the ‘Inverter Controller’ block of Figure 1. The inverter model is shown in Figure 3, which consists of a DC to AC converter and an inverter-grid interface. The voltage at the inverter bus is synthesized to an AC voltage wave form and the voltage at the terminal bus is common with the grid side. The primary goals of the inverter are to regulate the terminal bus voltage Vt and the active power delivered to the grid Pgen. This is achieved by controlling the modulation index m of the inverter, which controls the inverter voltage Vi through the relationship