Transcript An integrated cyberinfrastructure for real

An integrated cyberinfrastructure for real-time
data acquisition and decision making in smart
buildings
Peter Shin
CSE UCSD
Advisors
• Tony Fountain
– Director of Cyberinfrastructure Laboratory for Environmental
Observing Systems (CLEOS), SDSC/CalIT2
• Paul Linden
– Founding director of UCSD Sustainability Solution Institute
– Chair, Department of Mechanical and Aerospace Engineering (MAE)
• Jan Kleissl
– Assistant Professor, Environmental Engineering, MAE
– Initiator of Decision Making using Real-time Observations for
Environmental Sustainability (DEMROES)
Overview
• Problem:
– Modern campus buildings consume large amounts of power (to control internal climate) and
are often inefficient in power utilization.
• Objective:
– Efficient energy utilization in building operations through automated real-time control system
• Approach: Integrated CI system
– Small, cheap sensors (weather stations)
– Wireless/wired networks
– Cyberinfrastructure for data acquisition and decision support
• Result:
– Integrated real-time sense-and-control system
Architecture
• Deploy weather stations around
campus
• Use streaming data middleware
to aggregate and integrate
streaming data from weather
stations
– DataTurbine
(www.dataturbine.org)
• Use complex event processing
engine to automate real-time
decision making
– ESPER
(http://esper.codehaus.org)
Streaming data
middleware:
DataTurbine
Complex Event
Processing
Enging:
ESPER
Automatic Building
Control System
Weather Stations (DEMROES)
•
Measurements:
•
Sampling Frequency:
•
DEMROES Sites:
– temperature, humidity, wind
direction, wind speed, rain fall
drops, solar radiation, solar panel
voltage, solar panel current
– 5 minutes per sample
– La Jolla west campus (14 stations),
east campus (3 stations), Scripps
Institute for Oceanography (4
stations)
– Hillcrest Medical Center: 1 station
– Camp Elliott Field: 1 station
– UCSD Shuttle buses, trucks (mobile
sensing): 2 stations
Streaming Data Middleware:
DataTurbine
• Open source, Java based
network ring buffer for all sorts of
data.
• Solution for accessing both
streaming and static data, from
different vendor systems, via a
common interface
• Provides real high performance
data streaming, 10Mb/sec,
1000 frames/sec on PCs
• Supported by NASA SBIR,
15 years in development
• Scalable: DataTurbine servers
can be interconnected to handle
large streams
• Can manipulate the streams:
fast forward or slow motion
playback (TiVo-like)
Complex Event Processing Engine:
ESPER
•
Complex Event Processing
Engine:
•
technology to process events
and discover complex patterns
among multiple streams of
event data.
•
deals with the task of
processing multiple streams of
event data with the goal of
identifying the meaningful
events within those streams,
and deriving meaningful
information from them.
Results
• A proof of concept for intelligent control of building
• A generic architecture for a broad suite of real-time
sense-and-control systems
– customizable for various applications
• Integrated software tools, i.e., the coupling of
DataTurbine with ESPER
– open-software products will be available to a broad
community of system developers