Collecting and Managing Building Data Jorge Ortiz Action
Collecting and Managing Building Data Jorge Ortiz Action. Webs Weekly Meeting September 1, 2010
Buildings large electrical energy consumer • 72% of electrical consumption, 40% of total consumption, 42% of GHG footprint • 370 B$ in US annual utility bill • 9. 5% of GDP in bldg construction/renovation • Primarily Coal generation • 10 -20% of the energy is wasted 1 Coal consumption by sector Electricity source – Visibility and intelligence can help reduce waste 2 1 K. W. Roth et al. , Energy Impact of Commercial Building Controls and Performance Diagnostics, TIAX (http: //apps 1. eere. energy. gov/buildings/publications/pdfs/corporate/pnnl-15149_market_assessment. pdf, 2005). 2
Talk overview • Building energy management goals • Cory Hall project • Simple Monitoring and Actuation Protocol (s. MAP) – Live, raw data • Integrated Sensor-Stream Storage System (IS 4) – Live, raw, contextualized and processed • Current state and future work
Power proportionality in buildings • Uncover underlying energy usage – What are the usage patterns and models? • Uncover available energy • Electric, solar, wind, etc. – How much does it cost? Etc. • Shed unnecessary waste – Define efficiency metrics • Schedule the remaining load – Optimization REPEAT – How much energy is available?
Uncovering each layer Environments & Activity Climate Plant Load Tree CT: mains power monitoring Vibration panel level power monitoring Humidity Temperature Pressure ACme: plug load energy monitor and controller 5
Applications Uniform Representation of Physical Data Personal Feedback Modeling Visualization Control Storage Location Continuous Commissioning Actuation Debugging Authentication Physical Information s. MAP Water Structural Electrical Weather Geographical Environmental Occupancy Actuator
s. MAP Overview • Simple Monitoring and Action Profile • JSON data format for measuring, sensing, and metering • HTTP/Restful interface to each sensor or meter Term Description Sensing sampling an instantaneous physical quantity. Application Interface HTTP Engine TCP HYDRO Routing Tiny. OS/BLIP Metering determining the integral of some physical quantity over time. 6 lowpan HC measureme nt point Refers to a single physical point of instrumentation. 802. 15. 4 Link measureme nt channel Measurement devices often derive multiple values from a single instrumentation point.
Capturing electric load tree Electrical Load Tree Physical CT mains power monitoring Vibration Circuit/breaker panel level power monitoring Humidity PAR/TSR ACme: plug load energy monitor and controller Temperature
Data-Metadata Bind Naming GET http: //is 4/cory/lt /main/devices/C 3/true_power xform REPLY { "status": "success", "pubid": "68060935 -1457 -414 c-b 330 e 2220969 d 970", "properties": { "formatting": { "Unitof. Time": "second", "Unitof. Measure": "k. W", "Divisor": 10, "Meter. Type": "electric", "Version": 1, … } Main
Cory Hall Load Tree RESTful access
Capturing context Chiller Pump Zone Chiller Vent Pump SF EF AHU
HVAC namespace /<building>/hvac_wet /<CT_source> /devices /<device> /<publisher> /<pump_source> /devices /<device> /<publisher> /<Chiller_source> /devices /<device> …
Namespace for spaces /<building>/spaces/<floor> /<area> /devices /<device> /<publisher> /<room> /devices /<device> /<publisher> /<area> /devices /<device> /<publisher> …
Hierarchical composition • Resource node hierarchy rules for each namespace – i. e. Room may have area child but not a floor child • Some resource have “devices” child • “devices” contain device instances • Device instances contain publishers – Publisher map to s. MAP “measurement points” • Namespaces combined through measurement devices
Combining namespaces
Combining query facilities
IS 4 Architectural Overview Temperature CT mains power ACme: plug load energy monitor and controller monitoring Publish PAR/TSR Humidity JSON N J SO bs Su HTTP/REST Metadata Manager Publish Manager Query Processor Subscribe Manager Data processing Security Manager e ib cr JSON Subscribe Sub IS 4 Storage JSO scri N be Vibration
Current state • 69 raw stream sources available (s. MAP) • 55 contextualized stream sources available (IS 4) • IS 4 release available for download – Version 2. 0 to be released in a few weeks
(Near) future work • Direct all s. MAP feeds through IS 4 • Integrate modeling with querying features • Automatic device configuration
Questions • More information – http: //is 4 server. com – Quickstart tutorial: http: //smote. cs. berkeley. edu: 8000/tracenv/wiki/i s 4_quickstart – jortiz@cs. berkeley. edu
- Slides: 20