The impacts of climate change and a decreasing Energy Return on Energy Invested (EROI) of fossil fuels1 have sparked interest in increasing the penetration of intermittent renewables into the world’s electric power grids. While pursuing this goal, however, “Perhaps the most important message in this book is that it is not effective to pre-commit to the deployment of a certain fixed capacity of renewables without understanding the characteristics of the existing resources and customer preferences” say authors Marija Ilic, Le Xie, and Qixing Liu in their book Engineering IT-Enabled Sustainable Electricity Services, The Tale of Two Low-Cost Green Azores Islands, published by Springer in 2013. And, “If new technologies are deployed as mandated by the regulators without enhancing today’s operating and planning industry practices it will be very difficult to manage new resources efficiently and reliably.”
Marija Ilić holds a joint appointment at Carnegie Mellon University as professor of electrical & computer engineering and engineering & public policy as well as honorary chaired professor for control of future electricity network operations, Delft University of Technology. Her principal fields of interest include electric power systems modeling; control of large-scale dynamic systems; nonlinear network and systems theory; modeling and control of economic and technical interactions in dynamical systems. Le Xie is an assistant professor in the Department of Electrical and Computer Engineering at Texas A&M University. His research interests include modeling and control of large-scale complex systems, smart grid applications in support of renewable energy integration, and electricity markets. Qixing Liu is a PhD candidate at the Department of Electrical and Computer Engineering, Carnegie Mellon University.
The Azores Archipelago consists of nine islands and is located about 900 miles off the coast of Portugal. Their book reports on results of simulations of the electricity flows on two islands based on a ten-year study demonstrating proof-of-concept using combinations of intermittent resources, complex software methods and adaptive hardware technologies in order to reliably manage system imbalances caused by intermittency. These concepts have the potential to scale to large interconnected power grids such as those found in the US, China and Europe.
Engineering IT-Enabled Sustainable Electricity Services, the Tale of Two Low-Cost Green Azores Islands is divided into seven parts.
Part I introduces traditional unit commitment scheduling practices and their hidden inefficiencies. The authors demonstrate why planning and operations of future electricity systems need to become more complex in order to take advantage of efficiencies created by an increasingly data rich environment. They cite examples of current inefficiencies including the uncoordinated objectives between transmission and distribution system networks and, the opportunity to relax proxy line flow limits and replace them with real thermal line flow limits. They introduce a new operations and planning architecture called dynamic monitoring and decision systems (DYMONDS) that can be used to balance supply and demand reliably and efficiently. They show that it is possible to make existing conventional electricity systems cleaner, more sustainable and cost effective by carefully deploying IT-enabled systems which are supported by automation. The concept of Socio-Ecological Energy Systems (SEES) is introduced using governance principles that earned the late ecological economist, Elinor Ostrom, a Nobel Prize in economics.2, 3 Ilic collaborated with Ostrom in developing these SEES concepts. The authors base their work on these principles saying, “Governance systems which rely more on proactive decision making by the resources and users than on the centralized coordination for managing intertemporal dependencies under major uncertainties support very different technologies and outcomes, because risks are managed in a distributed way over many resources and users and over time as decisions for managing risks are not over a prespecified time for all.”
Part 2 describes the electrical network, generation and demand characteristics of the Azores islands and especially the two islands, Flores and Sao Miguel. Flores is one of the smallest islands with about 4000 inhabitants and has a peak load of nearly 2 MW. It is powered by a fleet of diesel generators, a reservoir hydro plant, and synchronous wind turbine-generators. Sao Miguel Island is the capital and the largest island with 140,000 inhabitants. It has a peak load of 74 MW and is powered by geothermal, run-of-the-river hydropower, and diesel generators using heavy fuel oil.
Part 3 explains conventional generation dispatch methods in power systems that contain intermittent resources. The authors then develop the concept of look-ahead generation dispatch models, which are subject to ramp constraints and use advance knowledge of stochastic wind power and load demand forecasts. They show how these models can result in lower cost and less polluting economic dispatch solutions. “When longer-term predictions are available, it becomes possible to adjust the power outputs of the slow-responding power plants so that one avoids the need to schedule expensive fast-responding units in near real time.” Their models are decomposed into low-frequency signals for long-term policy adaptation and generation investment; medium-frequency signals for detecting seasonal weather variations; and, high-frequency signals for intra-day and intra-week variations. These models are used to simulate island generator control under various dispatch conditions in order to determine a low cost solution. The ability of different types of loads to participate in Adaptive Load Management (ALM) is assessed. The benefits and costs of centralized vs. decentralized economic dispatch are analyzed. They end by discussing the role of electric vehicles in supporting higher penetrations of renewable energy generation.
Part 4 offers recommendations about the optimal placement of wind power plants in order to minimize both real and reactive power delivery losses. They show the benefits of equipping all wind plants with automatic voltage regulators in order to improve voltage profiles. Today’s voltage dispatch rules are typically based on security constrained off-line DC Optimum Power Flow (DC OPF) models to insure system reliability during worst-case conditions. At higher penetrations of intermittent generation resources, they show that these off-line techniques are no longer adequate. They demonstrate why it is becoming increasingly necessary to use Extended AC Optimal Power Flow (AC XOPF) on-line models to compute needed resource management solutions.
Part 5 shows how intra-dispatch control of supply and demand imbalances caused by wind ramps can reduce the need for fast and expensive regulation reserves, such as batteries and flywheels. Advanced frequency stabilization and regulation techniques are analyzed for large penetration wind systems. Techniques to promote stability are described including the appropriate use of fast controlled energy storage devices such as flywheels and batteries, excitation systems of conventional power plants, Fast AC Transmission Systems (FACTS),4 and Doubly Fed Induction Generators (DFIG) for voltage control of wind turbines. They also identify solutions to promote small signal stability.
Part 6 provides recommendations on how to reconfigure distribution systems to incorporate distributed generation (DG) and demand response (DR) in order to minimize cost and reserve margin as well as to provide differentiated reliability-of-service options. This includes changing the topology and adding more normally-closed and normally-open switches so the distribution system can be easily reconfigured during system faults also allowing for the connection of DG’s to selected customers, often referred to as Islanding. The authors comment that, “One means to decrease the reserve margin of central generation and improve the reliability of a system is to connect DG to the system.” They also point out the critical nature of having remote fast communications systems in place in order to support future reconfigurable smart distribution systems. The authors analyze transient stability problems in networks with high wind penetration and offer a case for using FACTS devices and flywheels to help improve stability and reliability.
Part 7 discusses how generation planning offers a whole new set of challenges when adding a high penetration of variable renewable resources; changing grid topology to incorporate DR, DG’s and islanding; and increasing on-line IT-based hierarchical generation/load control. This final chapter identifies approaches to creating a dynamic optimization model for generation expansion decisions that considers the uncertainty of future load and wind power variations. “The objective is to maximize the sum of discounted social surplus over the planning horizon, considering supply costs and consumer benefits,” say the authors.
Much has been written about the need for and expense of additional fossil fuel generation backup as increasingly higher percentages of renewables are integrated into power systems. Contrary to this conventional thinking, Marija Ilic and her team show there is significant hidden potential in reliably and efficiently combining higher penetrations of distributed renewable energy resources at a low cost and with minimal use of fossil fuels. Their concepts enable intermittent resources to be effectively integrated using embedded model-based sensors, enhanced communications links and decision-making algorithms, combined with Adaptive Load Management (ALM) techniques that include DR and load shifting.
The authors recommendations on scheduling generation and interchanges at shorter intervals to reduce the need for fast ramping reserves are supported by data on the costs and emission impacts of cycling fossil fuel plants contained in the recently released NREL Western Wind and Solar Integration Study.5
Their identification of the need for a more flexible governance policy was echoed in the recently released America’s Power Plan, which advocates operational changes that reduce system costs using a more performance-based regulatory model.6
This data-rich book is an excellent resource for those who are interested in gaining a more thorough understanding of the enabling principles and technologies needed to cost effectively develop a stronger, cleaner and more intelligent electric energy infrastructure.
- Hall and Klitgaard, “Energy and the Wealth of Nations,” Springer 2012, Reviewed in Today’s Engineer, June 2012, http://www.todaysengineer.org/2012/jun/book-review.asp
- E. Ostrom et al., “A general framework for analyzing sustainability of social ecological systems,” Science 325, 419 (2009)
- Ben Ramalingham, “Conversations on Complexity, a Tribute to Elinor Ostrom,” August 16, 2012, http://aidontheedge.info/2012/08/16/conversations-on-complexity-a-tribute-to-elinor-ostrom/
- Peter Fairley, "Flexible AC Transmission: the FACTS Machine," IEEE Spectrum, December, 2010, http://spectrum.ieee.org/energy/the-smarter-grid/flexible-ac-transmission-the-facts-machine
- NREL, Western Wind and Solar Integration Study, Phase 2, September 2013,http://www.nrel.gov/electricity/transmission/western_wind.html#!
- Americas Power Plan; September, 2013, http://americaspowerplan.com/
Jim MacInnes worked as a power engineer in the electric utility industry for Ebasco Services, Inc. and later developed renewable power plants in California. He is chairman of the Michigan Utility Consumer Participation Board and has testified on energy issues before the Michigan House and Senate Energy Policy Committees. He is a member of the IEEE Power and Energy Society, the IEEE USA Energy Policy Committee and the US Society for Ecological Economics. He is a licensed professional engineer and holds BSEE and MBA degrees from the University of California, Irvine.