Computing Community Consortium Blog

The goal of the Computing Community Consortium (CCC) is to catalyze the computing research community to debate longer range, more audacious research challenges; to build consensus around research visions; to evolve the most promising visions toward clearly defined initiatives; and to work with the funding organizations to move challenges and visions toward funding initiatives. The purpose of this blog is to provide a more immediate, online mechanism for dissemination of visioning concepts and community discussion/debate about them.


DoE’s Quadrennial Review Emphasizes IT R&D

September 29th, 2011 / in big science, policy, research horizons, Research News, resources / by Erwin Gianchandani

DoE releases its first Quadrennial Technology Review [image courtesy U.S Department of Energy, First Quadrennial Technology Review, September 2011].At an event in Washington, DC, on Tuesday, the U.S. Department of Energy (DoE) released results of its first Quadrennial Technology Review (QTR) — launched earlier this year at the recommendation of the President’s Council of Advisors on Science and Technology (PCAST) to help the Department identify a set of priorities for its energy technology R&D activities.

As Energy Secretary Steven Chu noted:

Traditionally, the Department’s energy strategy has been organized along individual program lines and based on annual budgets. With this QTR, we bind together multiple energy technologies, as well as multiple DoE energy technology programs, in the common purpose of solving our energy challenges. In addition, this QTR provides a multi-year framework for our planning. Energy investments are multi-year, multi-decade investments. Given this time horizon, we need to take a longer view.

 

We also recognize that the Department is not the sole agent of energy transformation. Our efforts must be well coordinated with other federal agencies, state and local governments, and with the private sector, who are the major owners, operators, and investors of the energy system.

Predicated on two overarching challenges — within transportation, a reliance on oil; within energy, economic competitiveness as well as the reliability and sustainability of energy production and use — the QTR specifies a framework of six strategies:

  • Increase vehicle efficiency;
  • Electrify the vehicle fleet;
  • Deploy alternative hydrocarbon fuels;
  • Increase building and industrial efficiency;
  • Modernize the grid; and
  • Deploy clean electricity.

The QTR has framed six strategies to address national energy challenges [image courtesy U.S. Department of Energy, First Quadrennial Technology Review, September 2011].
What’s most interesting is that the report emphasizes — at great length, in fact — the very critical role of information technology R&D in pursuing at least two of these strategies:

Energy efficiency in buildings and industry:

An integrated approach to building design and operation can cost-effectively yield energy savings exceeding 50% in new builds. More than 40% savings have been demonstrated in retrofits in a variety of climates, including more than 20% savings over the current minimum requirements. Key enablers include calibrated data through distributed sensors, validated modeling, and real-time control of a building’s components and their interactions with the electrical grid…

 

Viewing whole buildings as systems, as opposed to treating them as collections of components, opens the door for additional energy efficiency opportunities. A systems-integrated building efficiency approach has three primary strategies: reduce internal loads through system integration; improve the building envelope; and design for, and maintain, efficient building performance. Understanding buildings as integrated entities through validated modeling and data collection enables greater control over building operations and energy use. A multi-building view of neighborhood or district energy use can reveal further opportunities for optimization, particularly of heat flows. The Department’s whole-building R&D portfolio will focus on gaining a better understanding of how buildings operate as a system, including the development of sensors, controls, and validated building energy models. This will guide R&D in component and envelope technologies, as well as the development of the next generation of model codes and building labels.

 

System integration can reduce loads by integrating building design (such as size, siting, and daylighting) with intelligently coordinated components and controls. Today’s design and construction practices, which treat a building as a collection of components, are not conducive to an integrated view. Holistic consideration of the building envelope (the walls, roof, and windows) can reduce load while improving indoor environmental quality, task lighting, and management of energy flows through the building. A variety of approaches are required for different climates, as well as for new and existing buildings. Integrated design would leverage modeling advances to improve decision-making throughout design, construction, commissioning, and operations.

 

Designing for and maintaining efficient building performance requires an understanding of how energy flows throughout the building. Wireless sensors and controls, linked with software, can help optimize energy use, lower maintenance costs, and improve thermal comfort and air quality. Sensor-generated data can both validate building models and provide actionable information for energy users, allowing continuous real-time tuning of the building HVAC and lighting to increase comfort while decreasing energy costs. Building retro-commissioning, which relies on measurements of building performance, can save energy by optimizing operations.

 

Acquiring “real world” energy-use data is a critical aspect of the Department’s R&D activity on both the component and systems levels. Such data will allow DOE to identify common inefficiencies, best practices, and opportunities for retrofits. Understanding the use patterns of appliances and equipment, how they interact, and how real buildings operate is critical to: (1) characterizing energy use for regulation and code development, (2) developing R&D programs that address real-world energy challenges, and (3) validating building energy models. Peak power generation is expensive, inefficient, and polluting. Buildings could help electric utilities meet peak load requirements by reducing loads as needed. Dynamic information exchange between the grid and buildings, combined with building controls, enables such demand response. Standardization and demonstrations are underway to enable these tools.

And grid modernization, spanning two directions:

Improve How the Grid is Observed, Understood, and Operated

 

The grid of the 20th century was operated in ways that are increasingly inadequate. Distribution circuits had simple characteristics, and issues ranging from power outages to system topology (circuit configuration) were verified by physical inspection. The changing generation and load mixes are increasing the uncertainty of grid dynamics, which requires better monitoring and control of power across the system. Introducing information technology for data awareness and communication will transform business models through new mechanisms of system diagnosis and operations, but raises potential cybersecurity concerns. Two-way communication and ubiquitous high-quality data from a range of devices (e.g., smart meters and phasor measurement units [PMUs]) will induce unprecedented software-based innovation in the system, including integrated management of both loads and generation.

 

These changing fundamentals alone require modernization of the grid, but the task is made more urgent by growing demand and aging infrastructure. For example, distribution transformers are nearing an average age of 40 years and have a life expectancy of no more than 50 years. While the need to replace aging components creates opportunities for modernization, maintaining system reliability is a challenge to rapid deployment of new technologies with uncertain performance characteristics and higher capital costs…

 

Dynamic control of the entire system will revolutionize the grid. The forthcoming information technologies and digital controls will entail swift—if unpredictable—introduction of new technologies, services, and business models. Data, modeling, and simulation will be key to dynamic control.

 

Sensors and Data

 

Improving grid awareness requires advances in data coverage, resolution, fidelity, and access. Such improvements will allow operators to observe and track dynamic events that are invisible to current monitoring technology, allowing for early response to sub-critical system failures and thereby preventing the cascading failures that lead to major disruptions (Figure 20 [below]).

 

High Quality Data Recorded Prior to 2003 Blackout. Failures began occurring at 1:31 PM, more than two hours before the blackout at 4:10 PM. These failures ("trips"), seen as changes in dynamic intensity, are indicated on the event axis. Because early failure signatures lasted only a few seconds (i.e., at frequencies higher than 0.25 Hz), these incidents were invisible to conventional monitoring technologies that only measure data every several seconds. One of the earliest PMUs deployed captured this data, which was not available to operators at the time of the blackout. (Image courtesy U.S. Department of Energy, First Quadrennial Technology Review, September 2011)

High Quality Data Recorded Prior to 2003 Blackout. Failures began occurring at 1:31 PM, more than two hours before the blackout at 4:10 PM. These failures ("trips"), seen as changes in dynamic intensity, are indicated on the event axis. Because early failure signatures lasted only a few seconds (i.e., at frequencies higher than 0.25 Hz), these incidents were invisible to conventional monitoring technologies that only measure data every several seconds. One of the earliest PMUs deployed captured this data, which was not available to operators at the time of the blackout. (Image courtesy U.S Department of Energy, First Quadrennial Technology Review, September 2011)

 

Technology will improve data quality and availability across the grid, not just in the transmission system. Sparse measurements of energy consumption (e.g., monthly meter readings) will transition to real-time data on voltage, power flow, phase angle, and other physical parameters at all scales. Using and protecting this data requires new data management capabilities, from communications and cybersecurity through data processing and visualization.

 

Modeling and Simulation

 

To optimize operation and planning, stakeholders must have a comprehensive understanding of how the grid responds to stress and how it would operate under different technology scenarios. Lacking this understanding, stakeholders are less able to predict which situations will lead to failures, and so must operate the grid more conservatively; they are also less willing to adopt new technologies. Although it is impossible to simultaneously account for all aspects of the grid, validated computer models would help reduce uncertainty. However, current tools are inadequate. Figure 21 [below] shows two examples where computer models fail to reproduce observations, even qualitatively.

 

Observed Dynamics and Simulated Results. These simulations, performed after-the-fact, were unable to reproduce the measured grid responses to major failure events. (A.) shows an inability to simulate the unstable grid dynamics that led to the 1996 western blackout affecting 7.5 million customers. The slower than expected recovery of the Eastern interconnect following a generator failure shown in (B.) indicates the system is more susceptible to cascading failures than models predict. (Image courtesy DoE, First Quadrennial Technology Review, September 2011)

Observed Dynamics and Simulated Results. These simulations, performed after-the-fact, were unable to reproduce the measured grid responses to major failure events. (A.) shows an inability to simulate the unstable grid dynamics that led to the 1996 western blackout affecting 7.5 million customers. The slower than expected recovery of the Eastern interconnect following a generator failure shown in (B.) indicates the system is more susceptible to cascading failures than models predict. (Image courtesy DoE, First Quadrennial Technology Review, September 2011)

 

The ability to accurately model and describe grid dynamics and predict system responses across multiple spatial and temporal scales will improve functions from real-time operation to multi-decadal planning. Current models and toolsets cannot scale to the physical and temporal resolution necessary to confidently assess and predict grid operation under stress. Updated software design and improved models of the physical system are required to better represent dynamics. Comprehensive and high-quality data will permit validation, refinement, or rejection of models.

 

Modeling and analysis of the electric system considers both physical and behavioral processes. The response of electricity consumers to dynamic prices or dispatch signals will have a significant impact on future system capacity, reliability, and flexibility. For distribution utilities, the historic models of energy demand and distribution circuits alike must be reinvented to reflect new capabilities being deployed across the grid. Today, there is no validated behavioral model to assess the impacts of these technological capabilities…

 

Improve Control of Energy and Power Flow

 

Distributed generation, demand response, and two-way power flow — all of which rely upon new data collection and communication mechanisms — will integrate the bulk and distribution power systems as never before. Doing so reliably requires knowledge of distribution circuits and their connected technologies, which will be provided by advanced metering infrastructure (AMI).

 

Continual monitoring of grid information at the home will allow for dynamic control of load along distribution circuits. This will improve power quality, enable proactive system optimization, reduce infrastructure costs, and enable new services and business models that benefit utilities and consumers alike.

 

All of this must be done while protecting the cyber and physical security of the grid itself. Absent the proper cybersecurity safeguards, ubiquitous communication could mean near-ubiquitous vulnerability.

 

Communications and Load Control

 

Dynamic response to grid conditions is among the most significant aspects of grid modernization. Properly controlled and integrated into the system, small variations aggregated across many individual loads can mitigate issues ranging from infrastructure capacity constraints to uncertain renewables generation. Conversely, poorly controlled dynamic response could reduce power quality and even system stability.

 

While AMI’s technical capabilities to communicate data are well understood, the real-world potential for demand response to improve grid services is not. Better understanding of how consumers respond to user interfaces and economic signals is needed, requiring integration of social science research with grid operation and planning.

 

Power Flow Control

 

Power electronics underpin the converters, controllers, and switches that regulate power flows on the grid. Advanced power electronics will ease renewable energy integration while improving stability as they can accommodate — and even counteract — voltage swings along circuits and dynamically reroute power in response to varying generation and system conditions. Transitioning to semiconductors with high operating temperatures (such as wide band-gap semiconductors) will allow for improved alternating current–direct current conversion, higher voltage operation, and improved efficiency. The cost and manufacturability of semiconductor materials tolerant of high voltage and temperature is a key challenge.

What’s more, in the QTR, the DoE goes on to assess its FY 2011 energy technology budget as categorized by strategy, concluding:

The DoE's FY 2011 energy technology budget, categorized by strategy [image courtesy U.S Department of Energy, First Quadrennial Technology Review, September 2011].DoE will increase emphasis on efficiency and understanding the grid. These sectors each have thousands of independent stakeholders and points of authority (and millions of energy consumers). DoE can leverage its role as a systems integrator and convener, particularly with the growing importance of information. While technology R&D is a foundation for DoE’s work in both efficiency and the grid, DoE will direct additional effort toward high-impact but relatively inexpensive activities that reduce non-technological barriers and increase coordination.

The QTR — which captures several of the research directions in the roadmap that resulted from the CCC/NSF Workshop on Sustainability and IT earlier this year — seemingly provides the computing research community a unique opportunity to pursue highly interdisciplinary DoE-funded research at the intersection of computing and energy in the years ahead.

To see the full QTR, click here. And to share your thoughts, comment in the space below.

(Contributed by Erwin Gianchandani, CCC Director)

DoE’s Quadrennial Review Emphasizes IT R&D

Comments are closed.