28-11-2012, 02:45 PM
Modeling and Simulation of a Distributed Generation-Integrated Intelligent Microgrid
Modeling and Simulation.pdf (Size: 5.26 MB / Downloads: 79)
Abstract
A reliable, efficient and secure electric power system is necessary for the operation of critical
buildings in a base or the whole base itself. This is also applicable for deployed force in forward
bases, which have to be put into service quickly and reliably. At present, there is a need to design
a distributed and autonomous subset of a larger grid or a microgrid to increase the security and
reliability of electricity supply. The objective of this work was to model and simulate a
specialized microgrid called an Intelligent Distributed Autonomous Power Systems (IDAPS),
which play a crucial role in building a scalable power grid that facilitates the use of renewable
energy technologies. Microgrid device models, including distributed energy sources and loads, as
well as their control algorithms, were developed. Several case studies were simulated to evaluate
the operation of the IDAPS microgrid during parallel and islanded operation modes. Simulation
results indicated that the proposed IDAPS control model was able to: (i) perform demand
management during normal operating condition; (ii) island the microgrid from the main grid
once an upstream fault is detected; (iii) secure critical loads and shed non-critical loads according
to the given priority list during emergencies; and (iv) resynchronize the microgrid to the main
grid after an upstream fault is cleared.
Objective
In response to the Statement of Need (SON) NUMBER: SISON-08-04: Scalable Power Grids
that Facilitate the Use of Renewable Energy Technologies, we proposed to model and analyze
the operation of an Intelligent Distributed Autonomous Power System (IDAPS), which provided
the opportunities for load control and dispatch of distributed energy sources, especially
renewables. This effort resulted in the intelligent distributed autonomous power grid that could
integrate renewable energy technologies and minimize reliance on external energy resources, and
thereby reducing fossil fuel consumption. This capability would facilitate the implementation of
renewable energy projects and enhance energy security and reliability for the mission-critical
parts of military bases and campus-type facilities.
Background
SERDP Relevant
Within the DoD and the new Army Energy Strategy for Installations, one of the major objectives
is to decrease dependence on fossil fuels and increase energy security http://armyenergy.
hqda.pentagon.mil/programs/plan.asp. To achieve this objective, it is necessary to
diversify DoD current use of the local electric utility by integrating various types of distributed
energy sources, including renewable energy systems such as wind, solar, and other advanced
non-polluting Distributed Energy Resource (DER) technologies (e.g., fuel cells and
microturbines).
Previous Work
Discussed below is previous and related work related to microgrid research, development of
DER models, together with their control and communication architecture.
Microgrid R&D
A microgrid comprises the interconnected distributed generation (PV, wind, diesel,
microturbines, fuel cells, etc) – along with energy storage devices (conventional batteries,
hydrogen storage, flywheels, etc) – and controllable loads at low-voltage distribution levels.
Such systems can operate in parallel with the local utility or in an islanded mode during
emergency conditions. Microgrid is one of the key technologies recommended by policy makers
drafting technology roadmaps for electricity delivery in many countries, including the United
States [1, 2], the European Union [3] and Japan [4]. Subsequent studies have all promoted the
microgrid concept citing the increased reliability and power quality it provides to the local
utility.
At the time of writing this report, there were two major efforts for microgrid development:
CERTS/Sandia Labs Microgrid Test Bed [5] and US Army CERL/Sandia Labs Energy Surety
Project [6]. The former developed the microgrid test-bed demonstration with American Electric
Power. The test-bed comprised three low voltage feeders at 480V and a couple of distributed
generators [7]. The work was very comprehensive and covered many aspects of generator
controls and parallel operation of various distributed energy sources. The latter addressed the
Energy Surety Microgrid project. It was one of the very few reported work that focused on
microgrid development at a military base with the objective of creating analytical tools and a
methodology to evaluate the impact of infrastructure disruption on base missions [8].
Battery storage model
Batteries are used to store excess electrical energy from the generation sources and supply the
load during the times when the sources are not available. Batteries are charged when an external
potential is applied to the batteries’ terminals. Batteries are discharged when an external load is
connected to the batteries’ terminals. When batteries are discharged, the batteries’ chemical
reaction is reversed and the absorbed energy is delivered.
Many battery models were developed over past years [32, 33]. A Thevenin equivalent
mathematical battery model was developed by [34]. A kinetic battery model composed of
capacity and voltage models was developed by [35] which represents the sensitivity of storage
capacity to the rate of discharge. Several other battery models [36, 37] were developed based on
the physical and electrochemical processes. Based on some of these previous works, a complete
battery model developed by [38] available in Matlab/Simulink comprised a controlled voltage
source and an internal resistance with current discharge characteristics.
Proof of Concept and its Technical Challenges
This work focused on modeling and simulation of a microgrid that networked variety of
distributed energy sources. Key technology gaps addressed included the development of a
distributed agent system that controlled and networked distributed energy sources, as well as
analyzing the dynamic response of distributed control strategies. In a given situation, an agent
must be able to issue a control signal in response to an event sensed from the external
environment quickly enough to manage the microgrid in a timely fashion. The simulation and
evaluation case studies were conducted to test network interoperability in both on- and off-grid
operation. As previously discussed in Section 2, our success criteria were to demonstrate that the
proposed microgrid could: (i) perform demand management during normal operating condition;
(ii) island the microgrid from the main grid once an upstream fault is detected; (iii) secure critical
loads and shed non-critical loads according to the given priority list during emergencies; and (iv)
resynchronize the microgrid to the main grid after an upstream fault is cleared. Through the
development of a “plug-and-play” interconnected power grid, it is expected that this work can
contribute to allowing the installation or deployed force to: (1) install future renewable energy
systems and (2) effectively control and optimally benefit from the power that is generated from
DERs. The power grid should provide these capabilities during both normal grid-connected and
emergency or islanded operating conditions.
Load Models
This study focused on household loads and their usage characteristics. Hourly residential load
curves of an average household used in this study were extracted from the RELOAD database
[74], which was used by the Electricity Module of the National Energy Modeling System
(NEMS). The hourly residential load curve data were available for twelve months (January to
December), three day types (typical weekday, typical weekend and typical peak day) and nine
load types (space cooling, space heating, water heating, cooking, cloth drying, refrigeration,
freezing, lighting and others). As the load curves in the RELOAD database represented hourly
fractions of the yearly load, the load curves were scaled up by the annual household consumption
and divided by the number of hours in a year, which is 8760.
Defining the Agent Architecture (Type, Role, Interaction)
Defining Types of Agents:
In this study, the idea behind any multi-agent system is to break down a complex problem
handled by a single entity into smaller simpler problems handled by several entities. This is
called a distributed and decentralized system. The proposed multi-agent system consisted of four
agents, namely control agent, DER agent, user agent and database agent. The four agents
together formed the multi-agent system that performed actions to achieve the goal of the system,
that is, to perform demand management during normal operating condition, and increase the
system’s resilience by performing islanding and securing critical loads. The architecture of an
IDAPS multi-agent system was developed and presented in Fig. 13.