Distributed Real-Time Systems for the Intelligent Power Grid

1
Distributed Real-Time Systemsfor the Intelligent
Power Grid

• Prof. Vincenzo Liberatore

2
Intelligent Power Grid

• Situational awareness by means of time-stamped
  data collection, real-time wide-area
  visualization, and data integrity within and
  outside an operators own area.
• Improvement of the quality of models and
  simulations by continuously matching models with
  measured data, for example to formulate and
  develop improved provisioning and contingency
  plans, and predictive models for security
  assessment and enhancement.
• Timely and accurate information dissemination to
  all key stakeholders, including state and local
  officials, as wells as customer communication
  that is more scalable than one-on-one telephone
  calls.
• Proactive operations of facilities.
• Real-time actions and distributed control of
  protection devices to prevent cascading failures
  or for the graceful degradation of user service
  based on service priorities, etc.
• Real-time wide-area control to minimize power
  generation over-provisioning.
• Context-dependent models and control of massive
  and cascading failures via predictable component
  interactions to achieve robustness,
  fault-tolerance, or graceful performance
  degradation.
• Large-scale distributed real-time embedded
  software development according to the best
  practices in Software Engineering.
• Integration of legacy systems as well as the
  plug-and-play introduction of novel components
  and solutions.
• Support for ubiquitous alternative energy
  generation systems and the seamless integration
  of these systems into utility operation.
• Market dynamics, for example, in the context of
  power routing transactions and regulatory issues

3
Distributed Real-Time Embedded Systems

• Communication Networks
• Connect computers, embedded systems sensors,
  actuators, controllers, signal processors,
  synchronophasors
• Distributed Systems
• Connectivity, programmability, scalability
• Real-Time
• Predictable timing of computation
• Critical for real-time monitoring and control

4
Distributed Real-Time and Power

• Distributed Real-Time Embedded systems provide
  underpinning of
• Communication
• Software Development
• Critical for objectives
• Situational awareness
• Distributed software components monitor phase
  angles and other quantities of interest
• Report to visualization centers, logging
  facilities
• Support cooperative work
• On-line diagnostics
• Off-line simulations, forensics
• Distributed control
• Automatically close the feedback loop

5
Short-Term Challenge Co-Simulation

• Simulate jointly the computer network and the
  grid
• Expertise
• Computer Networks simulations (Prof. Liberatore)
• Hybrid System simulations (Prof. Branicky)
• Previous work
• Ns2 and differential equation solver BLP03,
  etc.
• Oak Ridge National Labs, etc.
• Future work
• Co-simulation of computer networks and power
  systems
• Integration with Modelica
• Formulation of objectives and scenarios

6
Co-Simulation Methodology
Branicky, Liberatore, Phillips ACC03
Co-simulation of systems and networks
7
Medium-Term Challenges (I)

• Real-Time Networked Control (I)
• Close-feedback loop in real-time over a network
• Network Quality-of-Service (QoS)
• Prevent timing failures
• E.g., fully-distributed QoS L04a
• Allocation of network resources ABLP06
• Depends on system requirements (stability,
  performance)
• Fully distributed, asynchronous, scalable
• Dynamic and flexible
• Optimization approach

8
Medium-Term Challenges (II)

• Real-Time Networked Control (II)
• Application adaptability
• End-point adapts to timing failures L06
• In-network synchronization (IEEE PTP) B06
• Real-Time Secure Management

9
Play-Back
Sequence number
Play-back
time
10
Medium-Term Challenges (III)

• Software Engineering
• Large-scale distributed real-time embedded
  systems
• Functional scalability L04
• Software development platforms and middleware
• E.g., RT Corba and power applications
• Multi-agent software systems ACKRNL03
• Integration of software, protocols, and standards

11
Emulation

• Application benchmark
• National Lambda Rail (or GENI)
• NLR is planned to be capable of supporting both
  production and experimental networks.
• Not a single network or a single test bed but
  facilities to build multiple networks and
  multiple test beds at all of layers 1-3 including
  optical, switched, and routed.
• Goal is to have both persistent and flexible
  infrastructure(s)
• Foster network research
• Support QoS
• Real-Time Overlay
• Support end-to-end RT SR

12
Conclusions

• Intelligent Grid
• Distributed Real-Time Embedded Systems
• Immediate need
• Co-simulation
• Long-term needs
• Software Engineering
• Networked Control