Attention NAE Members
Starting June 30, 2023, login credentials have changed for improved security. For technical assistance, please contact us at 866-291-3932 or firstname.lastname@example.org. For all other inquiries, please contact our Membership Office at 202-334-2198 or NAEMember@nae.edu.
Click here to login if you're an NAE Member
Recover Your Account Information
Author: Timothy D. Heidel and Craig Miller
Agile and fractal power system design and control methods are needed to realize the benefits of distributed generation and storage technologies.
The methods used to plan and operate the grid since the dawn of electrification have worked well. Indeed, the US grid has set the absolute standard for scale and performance of engineered systems for more than a century, but new technologies, economics, social attitudes, and environmental sensibilities are calling this model into question.
Rapidly falling costs of distributed electricity generation methods such as solar photovoltaics and storage technologies coupled with the growing emphasis on improving electric power system resiliency have motivated the investigation of alternative architectures for planning and operating electric power systems.
In addition, recent advances in power electronics, computation, and communication technologies could provide the opportunity to optimize and control grid operations closer to the locations where power is consumed (Kassakian et al. 2011), offering significant efficiency, cost, reliability, and emissions benefits.
But the methods that have historically been relied on for designing and operating power systems will prevent the full realization of the potential benefits associated with the newer technologies. Power system design and -control methods that are both agile and fractal are needed to fully realize the benefits offered by distributed generation and storage technologies.
In a 2014 speech, then US Secretary of Energy Ernest Moniz defined the electrical grid of North America as “a continent-spanning machine of immense complexity that is at its best when it is invisible” (Moniz 2014). There has probably never been a more succinct and accurate definition of the grid that has grown from the first central power plant opened in 1882 in Manhattan.
For more than a century the electric power delivery system has evolved continuously as generations of engineers have identified improvements enabling greater reliability, resilience, and lower cost. Over time, every component and procedure has been refined and polished. Today, the grid operates with impressive reliability, often making it invisible.
The grid’s generally routine reliability is largely a consequence of the system’s scale, literally its angular momentum. Titanic power flows from the Hoover Dam and its kin, an immense fleet of large-scale central power generation stations throughout the country. Small generators and loads are effortlessly swept into synchronicity by the current flowing from these huge turbines. This has proven to be a very good way to design a system, especially given the economies of scale and increased efficiency of most electricity generation technologies.
Recent Changes Motivating New Adaptation
The recent rapid growth of distributed energy resources located at the far, thin edge of the grid is calling the existing model into question. As these resources continue to proliferate, individual homes, businesses, and factories will begin to have a far larger influence on the operation of the grid both locally and throughout the system (Kristov 2015).
Distributed and, particularly, customer-owned generation, thermal and electrical storage, and load control technology such as communicating thermostats and building management systems all raise the question of what constitutes a grid. Is the grid the continent-spanning totality, or is it one utility, one feeder, one portion of a feeder, or one building? The answer is, increasingly, all of these. A useful working definition of a grid is a collection of electrical assets (generation, load, storage, transport) that can be controlled by a single entity. By this definition, grids range from individual buildings to regional transmission organizations (RTOs) spanning multiple states.
A building energy management system may control rooftop photovoltaic or gas-powered combined heat and power technologies, loads, energy storage, and purchases from or sales to the grid. It is an electrical grid in every sense except scale and presents many of the same problems in optimal control. It must also act in harmony with all the other actors in the grid. This will become increasingly critical as the electric power system as a whole evolves to rely ever more heavily on distributed energy resources.
This is a unique time of challenges to adapt the grid for new and changing needs. The challenges present an opportunity to think beyond incremental improvement to a fundamental reimagining and reinvention, building on emergent technology in distributed generation and sensor technology and advances in communications and industrial controls.
Agile and Fractal Grids
The hierarchical model of the grid challenges the old simplifying dichotomy in which generation and transmission companies thought of the distribution system as an exogenous, slowly varying, uncontrollable load, and distribution companies treated the transmission systems as an infinite bus. With many systems and actors involved, the fundamental problem in operations moves from pure control to harmonization.
A conglomeration of today’s distinct and incompatible methods of operating buildings, campuses, feeders, distribution systems, generation and transmission systems, RTOs, and independent system operators (ISOs) will not enable high potential agility. Furthermore, it will create a morass of interoperability standards and local, ad hoc, idiosyncratic methods of coordination.
Efforts to establish commonality in the problems of grid operations across many scales can move the system closer to a grid that continuously adapts, collaborates, and harmonizes to achieve greater reliability, resiliency, and efficiency. We believe that such a grid must be agile and fractal.
What Is an Agile Grid?
To be agile in this context, a grid must be capable of dynamically reconfiguring and optimizing based on rapidly changing local conditions.
Even under ideal conditions the grid is constantly changing—components are installed and retired every day, and load varies with weather, season, and the vagaries of human activity. Beyond these (literally) “blue sky” variations, storms, natural disasters, equipment failures, and other factors disrupt normal grid operations.
Variations have always been present, but they are poised to have more significant impacts as new weather-dependent generation sources (such as solar and wind) and new electricity uses (such as electric vehicles) become ubiquitous. Efforts to design and build the grid of the future must therefore be based not on a static approach but on a design process that is constantly evolving and that allows the routine and continuous adaptation of operations to account for changing conditions and circumstances.
As new technologies enable a more efficient grid, the fiction of a static grid—designed to a fixed point and then simply operated as designed—will be further undermined. A campus or individual building in an office park may sometimes operate autonomously, sometimes focus on local coordination, sometimes operate as part of a much larger whole. The future grid must be envisioned as a grid of grids of grids, dynamically adapting when challenged.
What Is a Fractal Grid?
Fractal design is an essential element to achieve desired grid agility. Taking inspiration from fractal geometric figures, fractal grids will exhibit the same control and operational characteristics at every scale.
In a fractal grid, any part of the overall power system will be capable of performing all of the functions of the full grid today. With fractal design, parts of the grid can safely isolate from the rest of the power system if and when it is optimal to do so (e.g., in response to local weather conditions, changes in fuel costs) but return to the broader system when conditions change. Decisions on how and when to segment parts of the system will be based on economic, engineering, and business considerations.
Figure 1 illustrates the concept of an agile, fractal grid. Figure 1a illustrates the current operation of a distribution grid. Energy flows to customers from two substations and the system operates with a tree structure. A normally open switch isolates the green and blue portions of the distribution system. Individual customers with generation or storage can use power generated locally and may in some circumstances be able to feed that power back to the local grid.
Figure 1b illustrates how the grid might be reconfigured after an equipment or line failure. In this scenario, energy is still fed from two substations, but certain customers are now receiving power from Substation B instead of Substation A. This scenario is becoming increasingly common as utilities install automated switching technologies in distribution systems.
Finally, figure 1c illustrates the potential for a portion of the grid, corresponding to a small group of customers, to further isolate from the rest of the system for business or economic reasons. That portion of the system would consume power, in this specific scenario, purely based on locally available generation resources. Eventually, one would generally expect the operation of the system to return to that shown in figure 1a.
Rearchitecting the control of electric power systems will not be achieved quickly or simply. Indeed, the study of grid architecture is emerging as an important new research domain (e.g., Taft and Becker-Dippmann 2015). But the technology needed is there or nearly there.
Achieving the transition to an agile and fractal future grid will rely primarily on three classes of innovation:
(1) precise state awareness,
(2) precise controls, and
(3) advanced analytics (including forecasting and optimization technologies).
Precise State Awareness
Successful grid operations in an agile, fractal environment will require precise knowledge of the state of the grid at all times and locations. Grid operators need to understand the operating state and the real-time capability of loads, generators, and storage devices. Ensuring the safety of utility personnel and customers will also require a precise understanding, at all times, of what parts of the system are connected to each other (and what reconfiguration options are permitted).
Fortunately, recent years have seen dramatic advances in sensor technologies that can contribute to state awareness, such as communicating digital consumption meters and distribution system phasor measurement units (von Meier et al. 2017). The rapidly falling costs of communication technologies also enable grid operators at all levels to communicate state-related information more often and on a more granular basis.
Many companies are developing advanced switching and power electronics technologies that can enable more rapid and precise control (Bhattacharya 2017). More advanced protection system devices, reactive power controllers, networked switches, and disconnect-capable meters can enable more agile volt/voltage-ampere reactive (VAR) control throughout the system and a wider range of feasible system reconfiguration options.
Many of these technologies are already being adopted in the utility community to reduce system losses, enhance efficiency through conservation voltage reduction, or improve resiliency during and after storms. The power electronics–based inverters that interface with distributed energy resources such as photovoltaics or storage devices will play an increasingly important role in enabling more precise control of the system.
A new generation of electricity system data analytics is needed (National Academies 2016), with more precise and accurate algorithms for forecasting the evolution of customer needs and generation resource capabilities. Scalable algorithms will also need to be developed to optimize large, diverse fleets of controllable resources (Panciatici et al. 2014). These algorithms will help translate improved state awareness into decisions on how best to deploy distributed energy resources and other controllable devices.
Effectively and securely managing the transport, storage, and analysis of data among a large number of diverse stakeholders will be a key architectural design challenge. Advances in the analysis of corrupted or incomplete data will also be critically important. Many of these advances will rely on techniques for making decisions in the face of significant uncertainty.
Analytically driven and agile control of the grid is being made technologically possible by declining costs of renewable and distributed generation technologies, higher-performance computing, and high-bandwidth communications, coupled with advances in power electronics and related control technologies. Indeed, many of the individual components required to realize agile, fractal grid operations are either already available or in advanced development.
But significant research and development are still needed to determine how to optimally integrate all the required component technologies. A particular challenge will be harmonization of this vision for future grid operation with the reality of continuous incremental change, which is necessary to the engineering of all critical infrastructure technologies. Control systems that are consistent with agile, fractal operation will have to coexist for some time with the control approaches that are used widely today.
As this new architecture for the control of electricity delivery infrastructure becomes widely used, we expect it will be possible to achieve greater reliability, resiliency, and efficiency while also easing the challenge of adapting to future changes. Finally, we believe insights gained throughout this transformation could have important implications for the design of other highly distributed engineered systems.
Bhattacharya S. 2017. Smart transformers will make the grid cleaner and more flexible. IEEE Spectrum, June. Available at http://spectrum.ieee.org/energy/renewables/smart-- transforme rs-will-make-the-grid-cleaner-and-more-flexible.
Kassakian JG, Schmalensee R, Desgroseilliers G, Heidel TD, Afridi K, Farid AM, Grochow JM, Hogan WW, Jacoby HD, Kirtley JL, and 5 others. 2011. The Future of the Electric Grid: An Interdisciplinary MIT Study. -Massachusetts Institute of Technology Energy Initiative, December. Available at http://mit.edu/mitei/research/studies/the--electric-grid- 2011.shtml.
Kristov L. 2015. The future history of tomorrow’s energy network. Fortnightly Magazine, May.
Moniz E. 2014. Keynote Address, IEEE Innovative Smart Grid Technologies Conference, February 19, Washington, DC.
National Academies of Sciences, Engineering, and Medicine. 2016. Analytic Research Foundations for the Next--Generation Electric Grid. Washington: National Academies Press.
Panciatici P, Campi MC, Garatti S, Low SH, Molzahn DK, Sun AX, Wehenkel L. 2014. Advanced optimization -methods for power systems. Proceedings of the 18th Power System Computation Conference, August 18–22, Wroclaw.
Taft JD, A Becker-Dippmann A. 2015. Grid Architecture. Richland WA: Pacific Northwest National Laboratory. Available at http://gridarchitecture.pnnl.gov/media/white-papers/Grid% 20Architecture%20%20-%20DOE%20QER.pdf.
von Meier A, Stewart E, McEachern A, Andersen M, -Mehrmanesh L. 2017. Precision micro-synchrophasors for distribution systems: A summary of applications. IEEE Transactions on Smart Grid, June.