Presentation to

1
BARTs AATC Project Status Update

• Presentation to
• CBTC and Beyond Fifth International Conference
  on Communications-Based Train Control
• Washington, DC
• May 7-8, 2003

2
Overview
Brief Recap of AATC Concept
Brief Recap of AATC Concept
Project Status
Next Steps
3
Radio Ranging
Measure the round-trip propagation time and
calculate the distance.
4
Radio Ranging
Ranging with Multiple Radios
Multiple measurements Better accuracy
5
Message Relaying

6
Track Circuits detect a train...
RELAY INTERLOCKING
And the Station MUX sends speed commands to the
train.
7
RELAY INTERLOCKING
Existing interlocking receives track circuit
data...
and sends permissive route data to the existing
MUX
8
Overlay System
Ranging data provides train detection...
AATC sends train performance commands via the
radio network
9
Overlay System
When the AATC system is complete...
existing equipment will be removed.
10
Overview
Brief Recap of AATC Concept
Project Status
Project Status
Next Steps
11
Program Phases
Phase 1(4/96)
Phase 2(6/02)
Fruitvale, Lake Merritt and the Wye
Phase 3(7/05)
A-line thru Bay Fair
M-line thru Daly City
Hayward Test Track
12
Safety Certification

• Mixed Mode Certified for Revenue Service
  Effective July 1, 2002
• Full Mode Certification deferred. As of July 1,
  2002
• Dry-Run Tests Nearly Complete
• Formal Tests, VV and Report Remaining to
  Complete

13
Phase 2 Accomplishments

• Radio-based train location safety-certified
• Robust train-to-wayside communications proven
• Continuous speed, acceleration/braking commands
  achieved
• Closed loop braking from wayside demonstrated
• Run time savings proven
• Energy savings measured

14
Train Speed vs. Location
80
AATC speed
AATC vs ATC - Fruitvale to Lake Merritt, A2 track
ATC speed
70
60
50
Speed (mph)
40
30
20
10
Track Location (feet)
0
0
500
-500
1000
1500
2000
2500
3000
3500
4000
4500
5000
5500
6000
6500
7000
7500
8000
8500
9000
9500
-1000
10000
10500
11000
11500
12000
12500
13000
13500
14000
14500
15
Train Speed vs. Time
80
AATC vs ATC - Fruitvale to Lake Merritt, A2 track
AATC speed
70
ATC speed
60
37 seconds
Speed (mph)
50
192 seconds
229 seconds
40
30
20
10
Time (seconds)
0
0
5
10
15
20
25
30
35
40
45
50
55
60
65
70
75
80
85
90
95
110
115
120
125
130
135
140
145
150
155
160
165
170
175
180
185
190
195
200
205
210
215
220
225
230
235
100
105
16
Energy Measurements
Line Voltage
9.30 kWh
2.31 kWh
Fruitvale to Lake Merritt -- ATC
17
Energy Measurements
Line Voltage
7.81 kWh
1.66 kWh
Fruitvale to Lake Merritt -- AATC
18
Energy Measurements
Sample Runs Fruitvale to Lake Merritt (kWh/car)
Mode kWh In (Used) kWh Out (Regen.) Net Used
ATC 9.30 2.31 6.99
AATC 7.81 1.66 6.15
Savings (AATC vs. ATC) 16 12
Note Energy regenerated depends upon other train
traffic in area
19
Current Status

• System is not yet in routine operations
• Only 5 trains equipped
• Operations Staff not yet trained
• Need Test Director on-board
• Special operating instructions

20
Remaining Design Issues

• All remaining issues are in the adaptation of the
  technology to underlying train control system.
  For example
• Track circuit failure modes
• Fuzzy track circuits limits, compounded by train
  positioning margins
• Integration with existing system (SORS)
• Result is difficulty in locating a silent train
  in MANUAL if tightly sandwiched between two
  AATC trains

21
Operating Scenarios

• Thorough look at how system safety and
  performance is affected by
• Loss of radio communication (silent train)
• Equipment failure
• Unusual or improper operations
• 30 years in AUTO. Fall back to MANUAL is
  inoperable and less safe. Need for hardened AUTO
  system.
• Analysis could not have been done before target
  design was completed
• Joint effort by Supplier, BART project team, and
  staff from Operations Control Center

22
Modifications Underway

• Modify Silent Train Detection algorithms to use 2
  track circuits instead of 1
• Eliminate automatic transitions between AATC
  Mixed Mode and AATC Full Mode. However, still
  have fall back
• From AATC Full Mode to ATC with restricted speeds
• From AATC Mixed Mode to ATC w/o restricted speeds
• Position Accuracy improvements (partial)
• Tools for increased deployment efficiency

23
Pending Modifications (subject to funding)

• AATC Performance Improvements
• Additional scope originally identified in 1999
  but postponed until funding is available
• Position Uncertainty (PU) Improvements
• Reduced Brake Rate (RBR) Improvements
• Slow Speed Orders
• Defer all but 1 of 8 potential PU Improvements
• Territory (Oakland Wye Retrofit, M-Line)

24
Status of Phase 3 Installation
25
Remaining Phase 3

• Install wayside radios
• Install station computers in 5 TC rooms
• Collect radio ranging coefficients
• Perform database VV at each location
• Install AATC equipment on 289 cars and 40 MV
• Training
• Revenue Service Preparation

26
Schedule
Phase 2 Safety Certification June 2002
NTP Remaining Phase 3 August 2002
New test area ready (A30-A40) May 2003
Modifications Release 1 May 2004
Modifications Release 2 May 2005
Wye Re-layout Possibly Defer
Vehicle Retrofit May 03-Jan 2004
A-Line Ready (incl. Vehicles,Training) May 2004
M-Line July 2005 (Defer?)
27
Summary

• AATC concept is solid
• Interaction with existing system more complex
  than expected
• Two key Lessons Learned
• Continuously involve Operations Staff
• Try to break the design with Operating
  Scenarios analysis
• Found nothing than time and money cant fix!