Metering, Monitoring and Making Sense of
Energy Use in ‘Mixed-Use’ Buildings
Rajesh K. Gupta
• Professor & Chair, Computer Science & Engineering
• Associate Director, California Institute for
Telecommunications & Information Technology
University of California, San Diego
Our Team
Yuvraj Agarwal, Rajesh Gupta
Thomas Bharath Seemanta Sathya
John
Kaisen
Buildings are an important research focus

All electricity in the US: 3,500 TWh



~500 power plants @7TWh
BuildSys
Buildings: 2,500 TWh
All electronics: 290 TWh
1 PC per 200 sq. foot
1 PC = $100
1W saved = ~2W less imported
= 5W less produced.
Bruce Nordman, LBNL
Buildings consume significant energy
>70% of total US electricity consumption
>40% of total carbon emissions
Energy Dashboard
http://energy.ucsd.edu
Looking across 5 types of buildings
more
IT
From: Yuvraj Agarwal, et al, BuildSys 2009, Berkeley, CA.
Modern Buildings Are IT Dominated:
50% of peak load, 80% of baseload
Two Steps to Improving Energy
Efficiency
1.
Reduce energy consumption by IT equipment



2.
Servers and PCs left on to maintain network presence
Key Idea: “Duty-Cycle” computers aggressively
SleepSever: maintains seamless network presence
Reduce energy consumption by the HVAC
system



Energy use is not proportional to number of occupants
Key Idea: Use real-time occupancy to drive HVAC
Synergy wireless occupancy node
6
Duty Cycling: Processors, HVAC

Why not power-down machines that are not
working?


Or power-down building HVAC systems
Runs into several use model problems

“Always ON” abstraction of the internet


Unlike light-bulb, ‘when not in room, turn off the light’
Use model for the user/application and the
infrastructure are different

Network, enterprise system maintenance: distributed
control of duty-cycling has its own usability problems.
Collaborating Processors


Fundamental Problem: Our Notions of Power States
 Hosts (PCs) are either Awake (Active) or Sleep (Inactive)
 Power consumed when Awake = 100X power in Sleep!
Users want machines with the availability of active machine, power of
a sleeping machine.
Somniloquy
SleepServers
Host PC
Apps
Somniloquy
daemon
Operating system,
including networking
stack
Host processor,
RAM, peripherals, etc.
Maintain availability across
the entire protocol stack,
e.g. ARP(layer 2), ICMP(layer
3), SSH (Application layer)
Secondary processor
Network interface
hardware
wakeup
filters
Appln.
stubs
Embedded OS,
including
networking stack
Embedded
CPU, RAM,
flash
Power Consumption
(Watts)
Host Only
Stateful applications:
Web download “stub”
on the gumstix
Somniloquy
200
150
200MB flash, download
when Desktop PC is
asleep
100
50
1
0
1
600
601
1200
1201
1800
Wake up PC to upload
data whenever needed
2400
1801
2401
Timethan
(seconds)
92% less energy
using host PC.
Increase battery life
from <6 hrs to >60 hrs
Somniloquy exploits heterogeneity to save power and maintain availability
SleepServers for Enterprises: Architecture
Respond: ARPs, ICMP, DHCP
Wake-UP: SSH, RDP, VoIP call
Proxy: Web/P2P downloads, IM
Average Power
96 Watts
Average Power
26 Watts
DE
Deployed SleepServers across 50 users
Energy Savings: 27% - 85% (average 70%)
Total estimated Savings for
CSE (>900PCs) : $60K/year
Scenario: CSE Energy Use Reductions
• Deploy Somniloquy / Sleepserver
– Machine room
– PC Plug loads
80 kBTU/ft2
: 142 kW  71 kW
: 130 kW  70 kW
• Ventilation system:
– New fans, chillers : 65 kW  52 kW
• Lighting:
– Fluorescent lighting  LED
– Motion-detector controlled hallway lighting 42 kBTU/ft2
evenings & weekends: 50 kW  11 kW
12
Could CSE become a ZNEB?
• Solar energy
: 2700 m2 roof
• Solar PhotoVoltaic: 20% efficient
• How do we achieve 42 kBTU/ft2 ?
– Tracking solar PV
: add 30% irradiance
– Increase PV efficiency : 29% efficient
111 kBTU/ft2
22 kBTU/ft2
28 kBTU/ft2
42 kBTU/ft2
Dramatic improvements in energy efficiency and solar
conversion efficiency needed for ZNEB
13
Wait for global warming or
better solar cells?
Is that it?
Buildings 2.0: Occupancy-Driven Smart Buildings
Use occupancy and activity to drive energy efficiency in HVAC system usage.
Reduced cooling when a
room is empty.
Increased HVAC when a
room has more
occupants.
Occupancy
Performability
Adaptive Envelope
When there are less people in the room, reduce cooling. When there are
more, increase cooling as required to maintain comfort.
15
HVAC: Central control and Static Schedules
HVAC ON
5:15AM
HVAC starts at
this time
6:30PM
Un-Occupied Periods
HVAC stops at
this time
Some people actually
arrive 2 hours later!
16
16
Energy Consumption in a Mixed-Use Building
• HVAC loads significant: Electrical ( >25%) and Thermal
– Electrical (air handlers, fans, etc), thermal (chilled water loop)
– HVAC load independent of the actual occupancy of building
17
17
Relating HVAC Energy Use and Occupancy
• Controlled experiment in CSE over 3 days: Fri, Sat, Sun
– Friday: Operate HVAC system normally
– Weekend: HVAC duty-cycled on a floor-by-floor basis
– 1 floor (10am – 11am), 2 floors (11am – 12pm), ….., …..
• Occupancy affects HVAC energy
– Points to the benefits of fine-grained control
18
18
Occupancy Driven HVAC control
Synergy Occupancy Node
• CC2530 based design
• 8051 uC + 802.15.4 radio
• Zigbee compliant stack
• PIR + Magnetic reed switch
Key Design Requirements:
• Inexpensive (less than 10$)
• Battery powered – 4-5 year life
• Multiple sensors for accuracy
19
19
Accuracy of Occupancy Detection
• Over 96% occupancy accuracy with Synergy node
20
20
21
Deployment across 2nd floor of CSE
Floormap: 2nd Floor
- 50 Offices, 20 Labs.
- 8 Synergy Base Stations
Control individual HVAC
zones based on real-time
occupancy information!
22
22
23
Implementation: Interfacing with the EMS
Occupancy Data
Analysis Server (ODAS)
Database
Occupancy
nodes
Sheeva Plug
base stations
Windows Server
with OPC Tunneller
Database
BACnet OPC DA
Server
HVAC
Control
NAE
…
NAE
NAE
Metasys ADX
Occupancy Data Analysis Server
• Database to store mapping , MetaSys EMS – proprietary protocols
• OPC tunnel to communicate with EMS
• Actuation based on modifying status for individual thermal zones
• Use priorities levels -- co-exist with current campus policies.
• Occupancy data not visible externally
24
24
HVAC Energy Savings
HVAC Energy Consumption (Electrical and Thermal) during the baseline day.
HVAC Energy Consumption (Electrical and Thermal) for a test day with a similar
weather profile. HVAC energy savings are significant: over 13% (HVAC-Electrical)
and 15.6% (HVAC-Thermal) for just the 2nd floor
Estimated 40% savings if deployed across entire CSE!
Detailed occupancy can be used to drive other systems.25
25
Summary
• HVAC energy not proportional to occupancy
– Use of static schedules is common
– Significant energy wasted
• Fine-grained occupancy driven HVAC control
– Occupancy node: accurate, low cost, wireless
– Interface with existing building SCADA systems
• Evaluation: Deployment in the CSE building/UCSD
– 11.6% (electrical) and 12.4% (thermal) savings
– Estimate over 40% savings across entire building
26
26
Some (Recent) Pointers
• “Evaluating the Effectiveness of Model-Based Power Characterization”, USENIX
Advanced Technical Conference (ATC), 2011.
• "Duty-Cycling Buildings Aggressively: The Next Frontier in HVAC Control" ,
ACM/IEEE IPSN/SPOTS, 2011.
• "Occupancy-Driven Energy Management for Smart Building Automation" ,
ACM BuildSys 2010.
• "SleepServer: A Software-Only Approach for Reducing the Energy Consumption
of PCs within Enterprise Environments" , USENIX ATC, 2010.
• "Cyber-Physical Energy Systems: Focus on Smart Buildings" , DAC 2010.
• "The Energy Dashboard: Improving the Visibility of Energy Consumption at a
Campus-Wide Scale“, ACM BuildSys 2009.
• "Somniloquy: Augmenting Network Interfaces to Reduce PC Energy Usage" ,
NSDI 2009.
27
27
Thank You
An exciting time to be
doing research in
embedded systems
with tremendous
potential to solve
society’s most
pressing problems.
Rajesh Gupta
gupta@ucsd.edu