An Information-Centric Energy Infrastructure: the Berkeley Viewଝ

An Information-Centric Energy Infrastructure: the Berkeley Viewଝ

Sustainable Computing: Informatics and Systems 1 (2011) 7–22 Contents lists available at ScienceDirect Sustainable Computing: Informatics and Systems journal homepage: www.elsevier.com/locate/suscom An information-centric energy infrastructure: The Berkeley viewଝ Randy H. Katz ∗, David E. Culler, Seth Sanders, Sara Alspaugh, Yanpei Chen, Stephen Dawson-Haggerty, Prabal Dutta, Mike He, Xiaofan Jiang, Laura Keys, Andrew Krioukov, Ken Lutz, Jorge Ortiz, Prashanth Mohan, Evan Reutzel, Jay Taneja, Jeff Hsu, Sushant Shankar University of California, Berkeley, CA, United States article info abstract Article history: We describe an approach for how to design an essentially more scalable, flexible and resilient elec- Received 8 October 2010 tric power infrastructure – one that encourages efficient use, integrates local generation, and manages Received in revised form 23 October 2010 demand through omnipresent awareness of energy availability and use over time. We are inspired by Accepted 23 October 2010 how the Internet has revolutionized communications infrastructure, by pushing intelligence to the edges while hiding the diversity of underlying technologies through well-defined interfaces. Any end device Keywords: is a traffic source or sink and intelligent endpoints adapt their traffic to what the infrastructure can Smart Grid support. Our challenge is to understand how these principles can be suitably applied in formulating a Energy networks Supply-following loads new information-centric energy network for the 21st Century. We believe that an information-centric Slack and slide approach can achieve significant efficiencies in how electrical energy is distributed and used. The exist- ing Grid assumes energy is cheap and information about its generation, distribution and use is expensive. Looking forward, energy will be dear, but pervasive information will allow us to use it more effectively, by agilely dispatching it to where it is needed, integrating intermittent renewable sources and intelligently adapting loads to match the available energy. © 2010 Elsevier Inc. All rights reserved. “The coming together of distributed communication technolo- tion delays. Local distribution, via step-down transformers, is also gies and distributed renewable energies via an open access, expensive in cost and efficiency, and is a single point of failure for intelligent power grid, represents ‘power to the people.”’ – an entire neighborhood. The system demands end-to-end synchro- Jeremy Rifkin [28] nization, and lacks a pervasive mechanism for energy storage or buffering, thus complicating the integration of renewable gener- ation sources, sharing among grids, or independent operation of 1. Introduction and motivation subgrids during upstream outages. Supplies are meticulously scheduled, or dispatched, to meet pro- 1.1. The energy challenge jected loads, which in turn are oblivious to the available supply. When demand exceeds supply, the only available response is to Today’s energy infrastructure is a marvel of the Industrial Age, shed load through brownouts and blackouts. The utility’s strategy yet it is showing its age. The Grid as it exists today is characterized can best be described as one of load-following supply: the utility by centralized generation via large plants, and a massive, centrally manages its portfolio of increasingly expensive generation sources controlled transmission and distribution system. It delivers high – including if necessary entering a regional marketplace for energy quality power to all consumers simultaneously and is sized to ser- – to scale its supply to meet its load. This approach relies heavily vice the peak aggregate demand at each distribution point. Power is on multiplexing a large population of loads to smooth the aggre- transmitted via high voltage lines over long distances, with associ- gate demand. For example, a home draws on average 25 kWh/day, ated inefficiencies, power losses, and right-of-way costs. Adding which is less than 5% of its 100 A service, but demand swings rapidly transmission capacity is expensive, and involves long construc- from a fraction of a kW to several kW. Consumption correlations, like air conditioners on a hot day, drive demand beyond estimated aggregates. This can result in huge spikes in supply cost, and may ଝ Research supported by NSF Grant #CPS-0932209, the FCRP MuSyC Center, trigger blackouts. Department of Energy/Lawrence Berkeley Laboratories, and industrial support from Part of the challenge of the traditional load-following supply eBay, Fujitsu, Intel, Samsung, Siemens, and Vestas. approach is indicated by Fig. 1, which shows how California’s ∗ Corresponding author at: 465 Soda Hall, CS Division, University of California, Berkeley, CA 94720-1776, United States. Tel.: +1 510 642 8778; fax: +1 510 6437352. statewide daily peak electricity demand varies. For most of the year E-mail address: [email protected] (R.H. Katz). the demand hovers around 38 GW, but it increases during the sum- 2210-5379/$ – see front matter © 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.suscom.2010.10.001 8 R.H. Katz et al. / Sustainable Computing: Informatics and Systems 1 (2011) 7–22 Fig. 1. California statewide electricity demand and power content. mer months due to air conditioning loads and reaches a peak of management. Intelligence can be distributed to loads as well as the well over 60 GW for a few hours of a few days. The variation within supply side. But these are steps towards supply-following loads, that each day is equally dramatic. Other parts of the world experience is, loads that adapt their behavior to the available supply. such peaks for other reasons, but, in general, the peak demand far Information technology can improve the reliability, visibility, exceeds the average and is present for a small fraction of the time, as and controllability of the Grid. But making the Grid into a Smart illustrated by Fig. 2. Capital investment in generation, transmission, Grid will require more than the intelligent metering currently being and distribution must address the peak, even though this means discussed in the context of demand response systems. The Inter- that much of the resources will sit idle. net suggests alternative organizing principles for a smarter Smart A second part of the challenge is increasing the penetration of Grid. The Internet succeeded by pushing intelligence to the edges renewable sources beyond the scale that is typical today, as indi- while hiding the diversity of underlying technologies through well- cated for California and a hypothetical retail electricity product in defined interfaces. Any device can be a source or sink of routable the Power Content Label. Renewable resources like wind and solar traffic and intelligent endpoints adapt their behavior to what the bring additional variability due to environmental effects and they infrastructure can deliver in accordance with localized utility func- cannot be dispatched whenever desired. tions. While others have also made these observations (e.g. [17,23]), Utilities are motivated to mitigate energy consumption because we focus more deeply on the underlying information architecture of the expense of deploying new infrastructure to meet growing of the Smart Grid, an intensive area of active research. demands, and they generally seek to reduce the carbon content of Radical proposals to replace existing infrastructures, given their the fuel mix. But, the system is neither agile, making it difficult wide deployment, high capital costs, and well-understood tech- to exploit non-dispatchable renewable supplies, nor is it able to nologies, are unlikely to succeed. Here, too, the Internet offers a mitigate demand except through the blunt instrument of price sig- model – of infrastructural co-existence and service displacement nals, also known as demand response. Alternatively, some loads can over time. The early network was deployed on top of the telephone be shifted to off-peak periods, through a process of demand-side network. It provided a more resilient set of organizing principles, Fig. 2. The load duration curve. R.H. Katz et al. / Sustainable Computing: Informatics and Systems 1 (2011) 7–22 9 became its own infrastructure, and eventually the roles reversed: describes recent trends in Energy Supply and Demand. Section 4 services such Voice over IP (VoIP) telephony are recent additions, introduces Energy System Terminology. We present the concept having been added over time. The same approach can yield a new of Deep Demand Response and Intelligent Loads in Section 5.We architecture for local energy generation and distribution that lever- describe Energy Proportional Computing in Section 6. Our initial ages the existing energy grid, but achieves new levels of efficiency approach for an Architecture for Energy Networks is presented in and robustness. This is similar to how the Internet, built in part on Section 7. Our Summary and Conclusions are given in Section 8. top of the telephony technology of ATM, has improved the phone network. 2. The concept of energy networks Combining intelligent communication protocols with energy transmission in a common architecture makes possible distributed Pervasive information can fundamentally change how electrical control and continuous demand response to pricing signals or to power is produced, distributed and used. The crucial insight is to richer expressions of energy availability. Such an infrastructure integrate information exchange everywhere that power is trans- design would

View Full Text

Details

  • File Type
    pdf
  • Upload Time
    -
  • Content Languages
    English
  • Upload User
    Anonymous/Not logged-in
  • File Pages
    16 Page
  • File Size
    -

Download

Channel Download Status
Express Download Enable

Copyright

We respect the copyrights and intellectual property rights of all users. All uploaded documents are either original works of the uploader or authorized works of the rightful owners.

  • Not to be reproduced or distributed without explicit permission.
  • Not used for commercial purposes outside of approved use cases.
  • Not used to infringe on the rights of the original creators.
  • If you believe any content infringes your copyright, please contact us immediately.

Support

For help with questions, suggestions, or problems, please contact us