Michael Lemmon University of Notre Dame

1
CSOnet A Metropolitan Scale Wireless
Sensor-Actuator Network

• M.D. LemmonDept. of Electrical
  EngineeringUniversity of Notre DameL.
  MontestruqueEmNet LLC, Granger, Indiana
• MODUS Workshop - St. Louis, Missouri - April
  21, 2008.

2
Outline

• Combined Sewer Overflow Problem
• Distributed in-line Storage Strategy
• CSOnet
• Ad hoc wireless sensor actuator network
• Notre Dame, EmNet LLC, Purdue, City of South
  Bend,Greeley-HansenIndianas 21st Century
  Technology Fund
• System Architecture
• Hardware/Middleware Components
• Control Application Algorithms
• Current Status
• Miami-Ireland Prototype (2005)
• 100 sensor node deployment (2008)
• Completion of actuator deployment (2009)

3
Combined Sewer Overflow Events

• Combined sewer overflow (CSO) events occur when a
  municipality dumps untreated water from combined
  storm and sanitary sewer flows into a
  river/stream.
• Such exceedances can pose risk to human
  health, threaten aquatic life and its habitat,
  and impair the use and enjoyment of the Nations
  waterways.
• EPA fines for CSO events- 1994 CSO Control Act-
  Fines are Significant
• Problem is Large-scaleOver 772 cities nationwide

Interceptor Sewer
EPA, Combined Sewer Overflow Control Policy,
April 19, 1994. (www.epa.gov)
4
Solution Strategies
Off-line Storage TunnelsChicagos TARP project
Sewer Separation
Expansion of WWTP
These strategies all require significant
investment in new infrastructure
5
In-line Storage

• Take advantage of excess capacity in existing
  sewer system.
• Centralized Approach
• Sensors measure storm inflow and forward to WWTP
• WWTP forwards control decisions to valves
• CSOnetDecentralized Approach
• Ad hoc sensor-actuator network
• Local decision making

• Advantages of CSOnet
• Reduced Cost
• Scalable Performance
• Fault Tolerance

6
Ireland/Miami Network (Summer 2005)
Prototype system controls retention basin based
on flow measurements at CSO 22 diversion structure
CSO 22Diversion Point

• 7 Relay Rnodes (radios)
• 3 Instrumentation Inodes (sensors)
• 1 Gateway Gnode connect to the internet
• Automated valve
• Network is fully deployed and operational

R
Combined SewerTrunk Line
I
G
V
G
V
Retention Basin
First month of service the system prevented 2
million gallon CSO discharge Continuous operation
since summer 2005.
CSO22Area
Slide provided by courtesy of EmNet LLC
7
CSOnet System Components

• INode sensor measurements, battery powered, can
  be placed in manhole.
• RNode in charge of forwarding INode messages to
  other nodes, battery or solar powered. Current
  range 1,000-2,500 ft, depending on line of sight
  clarity.
• GNode can connect to ethernet based network or
  cellular network, access to wireless network,
  controls structures.

Slide provided by ourtesy of EmNet LLC
8
CSOnet Inode Emplacement
Rnode
To next Rnode
Stoplight Pole
Manhole Cover Antenna
Inode
microprocessor
Sensor
Deployment plate
Slide provided by courtesy of EmNet LLC
9
Smart Manhole Cover

• Manhole is extremely corrosive environment
• Initial prototypes rusted away with a few months

• Integration of processor, radio transceiver, and
  antenna into manhole cover
• William Chappell - Purdue
• Ceramic modules

Slide provided by courtesy of Bill Chappell
(Purdue)
10
Chasqui Node
Chasqui 2.0 with antenna
Chasqui 1.0
Chasqui 1.1
Chasqui 2.0

• Based on UCB Mica2 Module
• MaxStream Radio (115 kbps/900 MHz)
• Rugged Sensor-Actuator I/F
• Precision Real-time Clock (2ppm drift)

Slide provided by courtesy of EmNet LLC
11
Power Management

• Management of system duty cycle
• 2 percent duty cycle - 5 minute period
• During sleep cycle, microprocessor put into deep
  sleep mode. External timer is used to wake the
  system back up
• Requires tight clock synchronization
• Chasqui uses Dallas DS3231 RTC with 2 ppm drift.
  
• Resync network clocks every six hours
• Chasqui service lifetime
• 2 years between service
• 3.6 volt - 19 amp-hour lithium source
• Currently more cost effective to replace
  batteries than to use renewable power systems
  such as solar.

11
12
Communication

• Physical layer
• Maxstream Radio
• 900 MHz, 1Watt, FHSS, 115kbps
• 1000-2500 range in urban environment
• GNODE serves as beacon
• Routing tables built up using flooding protocol
• Messages forwarded to GNODE
• Multiple destinations
• High Reliability
• 95 over 8 hops
• Positive acknowledgements
• Adaptive link switching
• GNODE connects to Internet
• Wired ethernet
• Cellular telephone

12
13
Wireless Reprogramming

• Stream Reprogramming protocol developed by Dr
  Saurabh Bagchi (Purdue)
• Less overhead than Deluge
• Stream segments the program image into Stream-RS
  (Stream Reprogramming Support) and Stream-AS
  (Stream Application Support)
• Stream-RS
• Core reprogramming component
• Preinstalled, before deployment, in all nodes
• Stream-AS
• A small subset of reprogramming component that is
  attached to the user application
• Instead of wirelessly transferring through the
  network user application plus the entire
  reprogramming component, Stream transfers
  Stream-AS plus the user application

Flash
Program memory
Stream-RS (Image-0)
Currently executing program
Stream-AS User application
(Image-1)
Unused portion
Slide provided by ourtesy of Saurabh Bagchi
(Purdue)
14
Actuators
Actuated Valve
Pneumatic Bladder
Slide provided by courtesy of EmNet LLC
15
Distributed Real-time Control

• Model Variables
• wi storm inflow
• Oi overflow
• ui diverted flow
• Hi water height (head)
• Q I pipe flow rate

Optimal Control Problem

• Maximize diverted flow
• Subject to
• Conservation of Mass
• Conservation of Momentum
• Admissible control
• No flooding
• WWTP capacity limit

16
Distributed Control Strategy

• Solution to Previous optimal control problem
• Open valves until flooding constraint is active
• Then reduce diverted inflow to prevent violation
  of flooding constraint
• Distributed Implementation
• Local controller enforces flooding constraint
• 2-3 order dynamical model of each node
• Local regulation only using pressure feedack is
  sufficient to ensure overall system stability

STORM FLOW
VALVE
OVERFLOW
FLOODING CONSTRAINT ACTIVE
UPSTREAMTOTAL FLOW DOWNSTREAMPRESSURE
HEIGHT
INODE
Pressure Sensor(water height)
TIME
17
Simulation Results
1. Two Week Storm (80) (0.485 of rain in 11
hrs) 2. One Month Storm (90) (0.799 of rain
in 13 hrs) 3. One Year Storm (99) (2.046 of
rain in 19 hrs)
Storm Existing System Overflow (ft3 x 106) Controlled System Overflow (ft3 x 106) Overflow Volume Decrease (ft3 x 106) Overflow Decrease ()
Storm 1 1.50 1.10 0.40 27
Storm 2 3.46 2.68 0.78 23
Storm 3 13.6 9.45 4.15 31
Table A. Scenario AMoving Uniform Rainfall
18
South Bend CSOnet

• PHASE 1 (summer 2008)
• Monitor 36 outfalls, 27 interceptor locations, 42
  trunkline locations, and 5 basins
• PHASE 2 (summer 2009)
• Control at least 18 outfalls
• Control storage in retention basins
• Control remediation pumps wet weather discharges

Slide provided by courtesy of EmNet LLC
18
19
CSOnet concept
Wastewater Treatment Plant
CSOnet Sensor I Node
City Engineer
Data Acquisition Point - GNode
Slide provided by ourtesy of EmNet LLC
19
20
CSOnet Website
Overflow Event
Slide provided by courtesy of EmNet LLC
Sensor Location
21
Cyber-Physical Systems
CSOnet is an example of a Cyber-Physical
Systemused in the Management of Critical
Infrastructure

• Cy-Phy Systems are large-scale systems
  integrating computation and networking to manage
  a physical system
• Scalability and Complexity
• Performance spanning Multiple time/spatial scales
• Testbed based on private, public, and academic
  sector collaborations
• MODULAR (compositional) design and analysis
• But not achieved through a separation of
  concerns
• FEEDBACK will be an essential principle of CPS
• powerful principle in cellular networks
• Cy-Phy Systems are the signaling cascades of our
  civil society