Note: Descriptions are shown in the official language in which they were submitted.
CA 02531842 1998-09-11
-1-
METHOD OF OPTIMIZINGi TRAIN OPERATION .
This application is a divisional of co-pending
application 2,303,281, filed September 11, 1998.
The present invention relates generally to
locomotive display and more specifically to a method
of optimizing train operations and training and its
use with, for example, a Locomotive Engineers Assist
*
Display and Event Recorder (LEADER).
The LEADER System is a real-time, enhanced
version of the Train Dynamics Analyzer (TDA), a long
standing Locomotive Engineer training tool offered by
the Train Dynamics Services Group of New York Air
Brake. LEADER has the ability to display a real-time
or "live' representation of a train on the current
track, the trackage ahead, the dynamic interaction of
the cars and locomotives (both head end and remote).
and the current state of the pneumatic brake system.
As a tool for the Locomotive Engineer, LEADER will
allow insight into the effect of throttle changes and
t
brake applications throughout the train providing
feedback and information to the Locomotive Engineer
not currently available. The information LEADER
offers provides an opportunity for both safer and more
efficient train handling leading to enormous potential
economic benefits.
*Trade-mark
CA 02531842 1998-09-11
-la-
The LEADER System has all the necessary
information to predict the. future state of the train
given a range of future command changes (what if
scenarios). With this a2~ility., LEADER can assist the
railroads in identifying and implementing a desired
operating goal; minimize time to destination, maximize
fuel efficiency, minimize in train forces, (etc.) or
a weighted combination thereof. LEADER will perform
calculations based on the operational goal and the
current state of the train to make recommendations to
CA 02531842 1998-09-11
-2-
the Locomotive Crew on what operating changes will best
achieve these goals.
The TDA functionality was enhanced to assist in
training Locomotive Engineer how to better handle their
trains. Designs of simulators with math models are shown in
U.S. Patents 4,041,283 (issued Aug. 9, 1977); 4,827,438
(issued May 2, 1989) and 4,853,883 (issued Aug 1, 1989).
Further capability was added to investigate accidents by
playing the event recorder data through the TDA, monitoring
critical physical parameters. Through the years data was
collected from instrumented trains and laboratory
experiments, allowing the models used by the TDA to be
refined. On board data collection for off-loading is shown
in U.S. Patents 4,561,057 (issued Dec. 24, 1985) and
4,794,548 (issued Dec. 27, 1988).
As more Locomotive Engineers became familiar with the
TDA display through training sessions, it became apparent
that a real time version of the TDA in the cab of a
locomotive would offer substantial benefits in improved
train handling. Improved train handling would in turn foster
safety and economic benefits. Earlier designs for on board
computer controllers is shown in U.S. Patent 4,042,810
(issued Aug. 16, 1977) with a description of math models.
The LEADER system provides safe and effective control
of a train through display or control of the dynamically
changing parameters. It accurately provides train speed
within designated speed limits. It maintains in-train
coupling forces with safe limits to prevent train break-in-
twos. It maintains safe levels of lateral forces between the
wheels and the rails of all cars to prevent cars from
departing from the track and derailing. It provides control
of slack action for both draft and buff between cars to
reduce
CA 02531842 1998-09-11
-3-
damage to valuable lading and to prevent potential
train separation or break-in-twos. It maintains train
stop and slow downs to prevent the train from entering
unauthorized territories that could cause accidents
with other train traffic. It determines the optimum
locomotive throttle setting and train brake
application to minimize fuel consumption and wear of
brake shoes and wheels. It monitors total locomotive
performance, train brake performance and it provides
advisement if performance is faulty. It forecasts the
estimate time of arrival of train to various switch
points, signals locations or final destinations to
advise the engineer and rail traffic control centers.
It records various key data for later downloaded
analysis for operational studies and accident
investigations as well as engineer qualifications.
A method of optimizing train operation includes
determining conditions of location, track profile and
train forces of the train. Next, a set of preliminary
train restraint operating parameters are determined
from the determined conditions. Also, at least one of
a set of preliminary train optimizable operating
parameters to minimize train forces, to maximize fuel
efficiency and to minimize time to destination is
determined. The determined set of preliminary train
operating parameters are weighted and combined.
Optimized train operating parameters are determined
from the weighted and combined preliminary train
operating parameters.
The determination of whether optimization should
be performed is made from train location track and
profile. This includes determining the location of
CA 02531842 1998-09-11
_q_
train with respect to one or more of hill, valley,
curve, signal and siding.
Determining optimized operating parameters
includes determining dynamic and fluid braking. The
fluid braking is determined individually for each
locomotive and car in the train which can be
individually controlled. The .dynamic and fluid
braking for each locomotive will be determined for
each locomotive individually. Dynamic and fluid
braking may be one of the sets of the preliminary
train optimizable operating parameters and will be
weighed and combined with the other preliminary train
operating parameters. The weighing of the dynamic and
fluid braking is a function of location on the track
profile. The train operating parameters include one
or more of train fluid braking, locomotive fluid
braking, locomotive dynamic braking and locomotive
propulsion. One of the train optimized operating
parameters is shutting down or restarting the
propulsion of individual locomotives. The optimized
operating parameters may be displayed and/or the train
controlled to the determined optimizing parameters.
Determining the preliminary train restraint
operating parameters includes one or more of speed
limits, slow orders, speed restriction zones, meets
and passes, track occupancy permits, general operating
bulletins, drawbar limits and slack action limits.
The preliminary train optimizable parameters are
determined using the operational restraints. The
method also includes determining train
characteristics. The train characteristics for each
car includes one or more of length, weight, position
of the train, braking equipment description, types of
CA 02531842 1998-09-11
_5_
bearings and wind drag areas. The characteristics for
each locomotive includes one or more of length,
weight, position of the train, traction performance,
dynamic braking performance, fuel consumption as it
relates to power control settings and locomotive
speeds.
The determination of train forces includes
determining forces experienced by and throughout the
train for the ensuing track. This includes
determining coupler forces and slack action throughout
the train. The determination of preliminary train
restraint operating parameters includes determining
preliminary train operating parameters to maintain
coupler forces below a set limit. If slack action has
been determined, the optimized train operating.
parameters are determined to achieve zero acceleration
within the train. If slack action and coupler forces
exceed predetermined limits, the operator is notified.
If the operator does not take appropriate action, the
train is controlled to the determined operating
parameters. The determining of forces includes
determining at least one of steady state draft and
buff, transit draft and buff, slack action and lateral
over vertical force ratio.
The determining forces also includes determining
forces exerted by the grade and rolling resistance
over the ensuing track. The preliminary train
optimizable operating parameters are determined to
maintain grade from the determined rolling resistance
and force exerted by a grade. A preliminary train
optimizable parameters are determined to stop at a
minimum distance from the determined rolling
resistance force and force exerted by the grade. The
CA 02531842 1998-09-11
-6-
method of determining the optimizable operating parameters
to stop at minimum distance is repeated until the train has
stopped. The preliminary train restraint and optimizable
operating parameters are determined to adhere to posted
speed restrictions.
The method of training a locomotive engineer in a
moving train includes displaying on the train conditions of
the train and a situation to which the engineer responds.
The engineer's response is determined and the display of the
train's conditions is changed in response to the engineer's
response. The display may also be changed to reflect the
actual conditions of the train resulting from the engineer's
response. The display may also suggest a response to the
displayed situation and the display can be changed to
reflect the conditions of the train resulting from the
suggested response. The engineer's response is recorded as
it relates to the displayed situation. The engineer's
response is compared to the suggested response to the
situation. The engineer is qualified from the comparison of
the engineer's response to the suggested response to the
situation. The response of the train to the engineer's
response as it relates to the displayed situation is also
recorded.
The method of training would also include determining
the qualification level of the engineer prior to training
from the engineer's inputs. The qualification level of the
engineer and an engineer identification may be inputted by
using an encoded device. The engineer is qualified based on
the determined response and the qualification level is
updated based on the determined qualification. Access to the
CA 02531842 1998-09-11
locomotive controls may also be controlled using a user
identification and determining the qualification level of
the user. The system is enabled if the qualification level
of the user meets the locomotive requirements. The user
identification and qualification levels may be inputted by
the previously discussed encoded device. The control system,
for example, the display, may be customized using the user's
identification and/or qualification level.
Other objects, advantages and novel features of the
present invention will become apparent from the following
detailed description of the invention when considered in
conjunction with the accompanying drawings.
BRIEF DESCRIPTION OF THE DRAWINGS
Figure 1 is a diagrammatic view of the energy balance
system according to the principles of the present invention.
Figure 2 is a block diagram of the system components of
a locomotive assist display and event recorder according to
the principles of the present invention.
Figure 3 is a flow chart for the control of the braking
and propulsion system according to the principles of the
present invention.
Figure 4 is a flow chart for the control of braking and
propulsion system car by car according to the principles of
the present invention.
Figure 5 is a LEADER display incorporating the
principles of the present invention.
Figure 6 is a block diagram with flow of information
from the train to the display incorporating the principles
of the present invention.
CA 02531842 1998-09-11
-8-
Figure 7 is a flow chart far location and
determination according to the principles of the
present invention.
Figure 8 is a flow chart for fuel optimization
using throttle setting according to the principles of
the present invention.
Figure 9 is a flow chart for fuel savings using
shutdown/restart of locomotives.
Figure 10 is a flow chart for balancing grade
according to the principles of the present invention.
Figure 11 is a flow chart for excessive force
warning according to the principles of the present
invention.
Figure 12 is a flow chart for slack related force
correction according to the principles of the present
invention.
Figure 13 is a flow chart for stop distance
calculation according to the principles of the present
invention.
Figure 14 is a flow chart for on job
training/qualification according to the principles of
the present invention according to the principles of
the present invention.
Figure 15 is a flow chart for controlling access
to the locomotive controls according to the present
invention.
CA 02531842 1998-09-11
_g_
DETAILFn D~~RIPTION OF TNF PRFE,F'RRF.~MHODIMF~NTS
LEADER operates on~the principle of conservation
of energy, kinetic and potential as illustrated in
Figure 1. Some events increase the amount of kinetic
or potential energy in the system, while others reduce
it. Burning fuel converts matter to energy (motion
via horsepower and heat), while 'braking converts
kinetic energy into heat, slowing the train. Energy
changes state but, the total sum of energy in the
system must be a constant. A train in motion is
constantly converting fuel to energy, converting
kinetic energy into potential energy traveling uphill
against gravity, converting potential energy into
kinetic energy traveling downhill, and removing
kinetic energy in the form of heat from the friction
and dynamic brake systems.
Math models of the LEADER System, monitors
parameters and performs calculations based on the
current energy state of the train to create a real-
time display of train dynamics. The power of LEADER
resides in its ability to provide information allowing
the crew to better control the train, minimizing loss
of energy. Loss of energy via over-braking represents
fuel unnecessarily consumed. Energy imparted to the
cargo of the train represents potential damage to
lading, equipment and rail. Both phenomena are
undesirable and addressable with LEADER.
The LEADER system is comprised of a number of
subsystems each with specific duties. Figure 2 shows
a generic LEADER architecture. The user interface of
the LEADER System is the real-time display which shows
a graphical and numerical representation of the
current state of the train as shown in Figure .5.
CA 02531842 1998-09-11
-10-
Radio communication is established between the lead
locomotive, the trailing locomotives in the lead
consist, and locomotives in the remote consist to
report the necessary parameters from each of these
locomotives necessary to perform LEADER Calculations.
Consist information is entered via the key pad on the
real-time display, a wired communication source
(laptop PC or removable storage device) or via wayside
radio conammication. Position is determined from
wheel movement sensors and a Global Positioning System
(GPS). The Input/output (I/O) Concentrator gathers
all of the various locomotive parameters necessary for
LEADER algorithm calculations and reports the
information to the LEADER Computer. The LEADER
Processor, a high throughput capacity computer
platform using a Real Time Operating System (RTOS),
then performs the calculations required by the LEADER
algorithms and the real-time display is updated. All
of these sub-systems combine to form the LEADER
System.
Each locomotive in a LEADER train will require at
a minimum, the I/O Concentrator with communication
capability to the head end. A LEADER Processor and
Display are only required for the lead locomotive.
The decision to equip all locomotives with a full
LEADER installation (Processor, Display in addition to
the I/O Concentrator) should be based on the Railroads
ability to permanently designate a locomotive as lead
or trail in its duties.
The development of LEADER began over 20 years ago
with early efforts to create the Train Dynamics
Analyzer (TDA), a computer math model used to predict
in-train forces. The train dynamic modeling
CA 02531842 1998-09-11
-11-
techniques and algorithms embodied in the TDA are
described in U.S. Patent 4,041,283(issued Aug. 9, 1977).
LEADER provides a number of unique features that
are available due to LEADER~s comprehensive knowledge
of the state of the train. Many of these features
allow the Railroad to convey and enforce operating
policies by programming limits into the LEADER System
such that when the limits are exceeded, LEADER will
audibly and/or visibly alert the crew to the
situation.
Figure S shows a "frozen" LEADER Display. Each
LEADER feature is identified by a block which points
to the appropriate screen location. The sections
following use the same paragraph number as the
identification block detail the operation of each
feature.
The LEADER Display shown in Figure 5 represents
one particular configuration for the display of LEADER
information. The display format~can--be customized on
customer request by adding information, removing
information, changing the color scheme, rearranging
the position of the sections of information, and/or
varying the size of ariy particular graphic.
In the following descriptions of features on the
LEADER Display the term function will be used to
deseribe the plot of the magnitude of a particular
parameter across the length of the train varying with
time.
5.1 Track Profile
The top portion of the LEADER Display shows the
track profile in three views. The train consist is
represented with different colored blocks for the
CA 02531842 1998-09-11
-12-
locomotive units and for the cars. The length of the
displayed train is proportional to the length of the
actual train. Milepost marks are represented by lines
running vertically through the track profile portion
of the display.
5.2 Horizontal View of Track
The horizontal view of the track profile shows
the grade on which the train is currently positioned
and the grade of the track profile for a number of
miles ahead . The Horizontal View of the track
profile will show the position of the entire train on
the track, both current location and geographic shape
(uphill or downhill) as a vertical slice of the track
profile in real-time.
5.3 Track Curvature Representation
The top graphic.. of the track prof ile section is
made up of blocks that represent track curvature. A
block above the dividing line represents a curve to
the right, a block below the dividing line represents
a curve to the left. The longer the block the longer
the curve. The higher the block the more severe the
curve.
5.4 Overhead View/Supplemental Information
Just above the horizontal view is the overhead
view. This view incorporates symbols to represent
track structures such as crossings, signals,
overpasses, underpasses, and sidings.
CA 02531842 1998-09-11
-13-
5.5 In-Train Forces
Directly below the train represented on the
LEADER display is the portion of the screen dedicated
to showing in-train forces. All in-train forces are
displayed as a graphic that maps to each car in the
train. Follow any point on any of the force functions
straight up the display and it will intersect with a
point on the train where that particular level of
force is currently present. Three types of forces are
represented on two different graphics. The two
graphics can be identified as the draft/buff magnitude
force functions and the Lateral divided by Vertical
(L/V) force ratio function.
The draft/buff force graphic represents draft
forces as a function above the 0 kilo-pound line and
buff forces as a function below the 0 kilo-pound line.
Draft and buff forces can be divided into two
categories, steady state and transient. Steady state
forces are shown, in general, by a smooth, relatively
slow changing function. Transient draft and buff
forces (run-in/run-out or slack induced forces? are
shown by "spikes" of force that travel through the
train. LEADER accurately calculates and displays
both. Slack induced forces represent momentum
transfers between the cars resulting in potential
lading and car damage.
The Lateral over Vertical (L/V) force ratio
graphic is shown above the draft/buff force graphic.
L/V forces appear in cars as they traverse curves in
the track.
CA 02531842 1998-09-11
-14-
5.6 Hrake Pipe Pressure/Hrake Cylinder Pressures
Directly below the force graphics are two
functions that represent brake pipe pressure
throughout the train and brake cylinder pressure
throughout the train. Again, these functions map to
a location in the train representation directly above.
Because the functions are real-time-representations of -
the brake system, it is possible to monitor a brake
application or release as it travels through the
entire train.
5.7 Trajectory/Telemetry Information
The lower right and lower center sections of the
screen have real time trajectory and status __
information displayed in both digital and analog bar
graph format. The following list contains the
parameters currently shown on the LEADER Display:
Sections are numbered to match the identification
blocks of Figure 5.
5.7.1 Head End Information
Location is a digital representation of mile
market location of the head end locomotive. Slope
is the grade of the track at the location of the
head end locomotive. curvature is the degree of
curvature of the track at the location of the head
end locomotive.
5.7.2 Speed is shown as a digital read out
followed by an analog bar graph representing the
speed of the head end locomotive at each instant in
time. The bar graph will turn from a normal color
of green to red if the speed limit is exceeded.
5.7.3 Acceleration is shown as a digital read
out followed by an analog bar graph representing the
CA 02531842 1998-09-11
v 4
-15-
acceleration of the head end locomotive at each
instant in time.
5.?.4 Current Speed Limit is shown as a
digital read out of the speed limit for the current
position of the head end locomotive.
5.7.5 Feed Valve Setting is the pressure at
which the~feed valve is set, displayed in pounds per
square inch (psi).
5.7.6 Fuel is the amount of fuel consumed
since the counter was last reset.
5.7.7 Consist Length is a digital read out of
the length of consist shown in feet.
5.7.8 Time is the digital read out of the
current time.
5.7.9 Hrake Pipe Reduction (or EP Brake
Command) This graphic takes on two roles; one for
conventional pneumatic brake equipped trains and one
for EP Brake equipped trains. In Conventional, the
graphic is a digital read out followed by an analog
bar graph representing the brake pipe pressure
reduction at the head end locomotive at each instant
in time. The LEADER system has the capability to
support trains equipped with EP Brake Systems rather
than conventional displacement valves. In an EP
equipped train the graphic is a digital read out
followed by an analog bar graph representing the
percent of brake commanded to the EP System.
5.7.10 Draw Bar Farces is a digital read out
followed by an analog bar graph representing the
instantaneous locomotive draw bar force.
5.7.11 Fuel Consumption Rate is a digital read
out followed by an analog bar graph representing the
CA 02531842 1998-09-11
-16-
instantaneous fuel consumption rate of the entire
train shown in Gallons Per Hour (GPH).
5.8.1 Excessive Speed Warning is an audio
and/or visual warning that will appear on the LEADER
Display when the speed of the locomotive exceeds the
speed limit for the current track location. The
speed restriction for the track can be modified to
reflect slow orders for a section of trackage. A
mark is shown on the speed indication graphic that
represents the current speed limit.
5.8.2 Minimum Safe Pneumatic Brake Reduction
is of interest for safe train operation. As brake
applications are applied and released the charge
state of the pneumatic brake system can become such
that an undesired release of brakes will occur if
the next brake application requested is not deep
enough. LEADER will calculate the safe brake
application level and visually display a target on
the Brake Reduction bar graph. If the brake
application requested is not deep enough, a visual
warning will be posted by the LEADER display.
5.8.3 Gluttony is a measure of how the train
is being moved with regards to fuel efficiency.
Gluttony is calculated and displayed in
gallons/thousand-ton-miles.
The LEADER display is equipped with eight
function keys at the bottom of the display. The
definition of each function key is shown in the
representation of the key on the LCD panel directly
above it. The function keys allow user input to the
system, accessing various setup and configuration
menus and querying information from the LEADER
system.
CA 02531842 1998-09-11
-17-
LEADER is capable of three operating modes,
each building on the previous mode. The three modes
advance LEADER from a real time display passively
providing information to the locomotive engineer
(information only mode) to a LEADER system that will
suggestions to the locomotive engineer on how to
better handle the train (driver assist mode) and
finally to a control system that is capable of
issuing commands to optimally control the locomotive
(cruise control mode).
In the information only mode, the locomotive
engineer makes all of the decisions and solely
activates the various control systems in a manual
mode. The LEADER provides information to the
engineer that is not currently available to him to
use to manage various locomotive control systems.
In driver assist mode, LEADER determines and
displays the optimum locomotive power dynamic brake
throttle setting and the locomotive and car brake
control settings. These settings are determined for
the head end locomotives and the remotely controlled
locomotives. These recommendations are desired
settings displayed to'the locomotive engineer who
can then elect to manually move the various controls
to achieve these settings.. In the cruise control
mode, leader derived settings are used to
automatically control the locomotive power and
braking systems, the train brake system of each car
and ancillary systems which effect train movement.
The locomotive engineer services an operational
supervisor with the ability to manually override the
cruise control. Cruise control can also be
effectuated by communication links between the
CA 02531842 1998-09-11
-18-
leader and the railroad central traffic control
center.
LEADER would receive and activate a variety of
commands received from the central traffic control
center. This received information would include
impeding slow orders, speed limit changes, aspects
of approaching signals, etc. LEADER would provide
status updates to the control center, including
current train speed, current track location,
detected faults or equipment malfunctions, estimated
time of arrival, etc. The traffic control can use
this information to manage the complete safe and
effective movement of train traffic under its
control.
The display of Figure 5 is produced by the flow
chart of Figure 6. The track profile data bases
include track profile or topology including grades,
curves, mile posts, town names, absolute latitude ,
and longitude coordinates, signal locations and
other various track structures. Besides the
physical characteristics of the track profile, also
the condition of upcoming signals, slow orders,
speed limits and location of speed restricting zones
are taken into account. The LEADER system includes
communication capability between the locomotive and
central dispatcher such that new orders can be
loaded into the system in real time_ THE LEADER
system also include multiple distance counters to
aid the locomotive engineer with navigating various
speed restrictions zones. The LEADER has the
capability to start, stop, reset and alert the crew
upon arbitrary condition of multiple counters shown
on the LEADER display. Thus, depending upon the
CA 02531842 1998-09-11
-19-
length of the train, the train may be in multiple
zones at a time.
Various physical conditions and train
characteristics are need for the methods to be
discussed below. The consist information or train
characteristics includes car length, car weight, car
position in the consist, braking equipment
description, types of bearings, wind drag areas. It
also includes information on the locomotive
including locomotive length, weight and position of
the consist, traction performance, dynamic braking
performance, fuel consumption as related to power
control settings and locomotive speeds.
The track profile data base and the consist
information can be loaded externally by the engineer
or communication links with the railroad central
traffic control center. Similarly, the information
on the consist may be derived automatically from
individual controllers on the locomotives on the
cars during power up as described in U.S_ Patents
6,049,296 (issued April 11, 2000) and 6,012,681
issued Jan. 11, 2000). Other information
that may be loaded would include operational
constraints for exceptional speed limits and special
slow orders, speed restriction zones, meets and
passes, track occupancy permits, general operating
bulletins, upcoming signal designations, desired
time of arrival and final destination location and
limit of authority or track locations the train move
to without the possibility of interference with the
movement of the trains in the area.
CA 02531842 1998-09-11
-20-
The track profile data base, location and
consist information are used to draw the track
profile in the horizontal view, curvature view and
overhead view and provided to the LEADER display.
The LEADER train dynamics calculations uses a
combination of instrumented measurements at the
locomotive and if available from the cars and a
computer calculation based on the fundamental laws
of physics, engineering mechanics along with various
empirical derivations. These measurements and
calculations are used to facilitate the estimated
required feedbacks. The measurements include
throttle and/or dynamic brake settings at the
locomotive, control pressures of various locomotive
air brake control systems, and locomotive speed and
increment wheel rotation. It also includes end of
train brake control pressures and brake pressures to
the individual cars available. Power produced
and/or anticipated within the locomotives traction
system and locomotive wheel slip. Activation of the
locomotive sanders and whistle and or bells. The
absolute track location and time of day are also
inputs. This is in addition to the consist
information and track profile previously described.
From this information, LEADER calculates the
steady state and transient draft/buff force
magnitude and its position of the consist, lateral
over vertical force ratio magnitude and position in
the consist, brake pressure magnitude and position
in the consist and brake cylinder pressure magnitude
and positions in the consist. These tour parameters
are then mapped to the train symbols and current
CA 02531842 1998-09-11
-21-
track profile locations and provided to the LEADER
display.
A flow chart for determining the location is
illustrated in Figure 7. Three different sources of
information are derived, weighted and combined as an
estimation of location. A first source for
information is transmitted to the train. This is
available from geographical positions system tGPS)
which is used to detect the precise latitude and
longitude of the head' 'bf the train . Another
commercially available system uses transponders
buried at spaced intervals under the rail. The
locomotive equipped with a transceiver activates
transponders as they pass over them. The
transponder then transmits an encoded message that,
when deciphered, provides track location. As
another alternative, the train engineer can enter
mile post signs which would then be compared against
the track data base to determine location.
The flow chart of Figure 7 also indicates two
incremental methods of estimating location. A turn
rate indicator provides a signal which is used to
1
index against the track data base to produce an
estimate of the location. Another is an end of axle
pulse counter which detects incremental distance
travel by the locomotive and is used to index the
track data base to an estimate location. The turn
rate indicator and the end of axle pulse counter
will operate to an increment from the previously
measured position. All three of the estimated
locations are then weighted and combined to produce
the ultimate estimate of location to be used in the
CA 02531842 1998-09-11
-22-
display and various calculations based on track
pro:ile.
The LEADER also has the capability of
determining whether its calculations are inaccurate.
For example, if there is error in actual distance
traveled versus the calculated distance traveled,
this may be brought to the attention of the
operator. This allows the operator to change the
weighting functions if desired in the algorithm.
These errors in the algorithms may be to faulty
measurements, wrong consist information, failures in
the car brake systems and/or failures in the
propulsion system. Even without the ability to
identify the specific source of the error, at least
notification that the error does exist will allow
the operator and the system to make more intelligent
decisions.
The LEADER integrates with electropneumatic
(EP) brake systems to provide braking on a car by
car basis. This can be used to maximize fuel
efficiency, minimizes in-train forces, and/or
minimize the time of the destination. By combining
knowledge of the position and state of the train,
the control of the individual brakes further
improves braking performance. For example, the EP
brake controllers on each car can be commanded to
apply brakes on the cars forward of a certain point
as the train crests the hill and not apply the
brakes on the cars which are below the crest of the
hill. This eliminates the cars on the uphill side
from being dragged with their brakes on until they
crest the hill.
CA 02531842 1998-09-11
-23-
The control of the individual care is
illustrated in the flow chart of Figure 4. The
process begins with LEADER monitoring the track
profile. If there is a geographic feature of
interest, it proceeds to identify train location
relative to geographic feature of interest. If not,
it cycles back to monitoring the track profile.
Geographic features of interest may include hills,
valleys, curves, signals or siding. These are
examples of features of interest which would benefit
from individual car braking. After identifying the
location of the train relative the geographic
feature, the identity of which cars are at the
geographic feature of interest is determined. This
is determined using car serialization data to order
and identify the cars. Serialization of the car may
be determined using the process described in
copending U.S. Patent 6,049,296 (issued April 11, 2000).
After identifying the location of the car
relative to the geographic feature of interest,
there is a decision made of whether optimization is
possible. If not, it cycles back to the LEADER
monitoring track profile. If optimization is
possible, the operating parameters for operational
constraints are determined. As previously
discussed, they may include limits on forces or
speed. Next, optimized weighted parameters from the
user are derived taking into account the operational
constraints. Examples of method of optimization may
be to maximize fuel efficiency, minimize in-train
forces or minimize time to destination. Also,
dynamic brake affinity and pneumatic brake affinity
CA 02531842 1998-09-11
-24-
can be weighted also. These are determined by the
user. These weighted factors are combined and
provided to the optimized weighted parameter from
user step. If less than all is desired, the weight
for the undesired parameter may be given zero.
The optimized weighted parameters are then used
to calculate optimized operating parameters, for
example, which car to brake and what level of
braking is necessary. The appropriate commands are
issued to the electropneumatic system brakes of each
car. This would include the car ID and level of
braking transmitted over the communication link
throughout the train. For the locomotives, this
would include pneumatic and dynamic braking as well
as propulsion.
If all of the cars are not equipped with EP
system brakes, the only control of the train is
through the brake pipe to the~individual cars as
well as the brakes and the throttle on the
locomotive. Thus, maximization of fuel efficiency,
minimizing the in-train forces and minimizes time to
destination would have to be controlled through the
throttle and brake pi~Se. If the cars do not have
individually controlled brakes, the optimized
operating parameters are determined on the system
level as illustrated in Figure 3.
It should be noted that.even if optimization
may not be possible, the flow charts of Figures 3
and 4 would also be operable to identify the train
location and the location of the individual cars and
determine whether the constraint limits have been or
are going to be reached. If there is an
anticipation that they are going to be reached, then
CA 02531842 1998-09-11
-25-
the calculation of restraint operating parameters
are performed. The calculation of which cars to
brake and what level of brake is necessary as well
as calculation of locomotive commands is performed.
These conanands are then issued to the individual
cars and locomotives. Alternatively, the braking
and propulsion control is performed on the system
level.
The calculations and the optimization of
Figures 3 and 4 are conducted not only for the
present location of the train but looking ahead at
the ensuing track profile. This is used to
determine what change in the train conditions will
result from the change in the track profile and
location_ Without such look ahead capability,
complete optimization is not fully achieved. Also,
anticipation of a diminishing or increasing
condition in track profile and train location will
affect keeping the train within the constraint
boundaries.
Proportional controls of the locomotive
traction power and dynamic braking include
subsystems for managing a locomotive pulling effect
and electric braking efforts. Proportional control
of locomotive air brake subsystems including both
the application and release the independent
locomotive brakes and the train or individual car
brakes. The activation/deactivation of ancillary
locomotive control systems include control of
sanding, the horn and the bells.
As a subset of the optimization of Figures 3
and 4 or as an independent system, fuel optimization
CA 02531842 1998-09-11
-26-
through the throttle setting including idle is
described in Figure 8.
Locomotives operate at maximum efficiency at
certain throttles. While multiple locomotives are
necessary to pull the train over certain areas,
other areas in the same territory may not require
all of the locomotives for operation. LEADER has
the knowledge of the state of the train currently,
the locomotive capacity, the train consist and the
track profile ahead. Using this knowledge, LEADER
can determine the optimal combination of throttle
settings on the locomotive to maximize efficiency.
The setting can be from idle to notch 1-8.
Information on the train consist, locomotive
information, track profile and current throttle
commands are used by the LEADER processing
algorithms to determine the train status_ The
determination is then made of whether the throttle
settings are optimal for fuel efficiency. If they
are, no further processing is required. If they are
not, then a recommended optimal throttle setting is
determined. This setting is then displayed to the
crew. In a semiautomatic mode, the crew response is
determined and if the crew selects the optimal
setting, no further processing is performed. If the
crew does not select the optimal setting, then the
system will automatically select the optimal
setting. Thus, fuel optimization through throttle
setting is achieved.
As a further extension of optimization using
throttle settings is the ability to save fuel
through shutdown and restart of the locomotives.
Over the course of operation of a freight train, for
CA 02531842 1998-09-11
-27-
example, a coal train operating in a loop, a portion
of the train's operating time is spent moving
totally empty. The amount of power required during
a loaded move will generally determine the number
and horse power of the locomotives in the train.
All trains are generally operated during both the
loaded and empty move in such a train, even though
the empty train movement may not require all of the
locomotives from propulsion. The LEADER's
capability to look ahead and calculate required
horse power for a certain section of a move, it is
possible to shutdown an unneeded locomotive and save
sufficient fuel. The LEADER can continue to look
ahead and determine whether and when additional
power will be necessary. They command the engines
to restart with proper warm up time to be available
when required.
The flow chart for such an operation is
illustrated in Figure 9_ User defined operating
parameters are inputted to the LEADER system. The
LEADER looks ahead to calculate or the determined
required horse power to navigate through the next
section of trackage. Then a determination is made
whether sufficient power is available. If not, then
there is a determination whether there dre any
available locomotives to start up. If not, the
system cycles back to the beginning. If there is,
then one or more of the available locomotives are
started up to provide the appropriate horsepower.
If sufficient horsepower is available, then a
determination is made of whether one or more of the
locomotives can be shutdown for the power
requirement. If not, then the system cycles back.
CA 02531842 1998-09-11
-28-
If one or more locomotives can be shutdown, then one
or more locomotives are appropriately shutdown.
This system is reiterated continuously monitoring
the ensuing trackage and adjusting the number of
locomotives which are shutdown or restarted.
Another limitation on the abilities to control
the in-train forces and fuel efficiency is that on
most trains, that although the freight brakes can be
gradually applied, they cannot be gradually
released. Once a brake is released by the
locomotive engineers, the brake will completely
release. After a brake application is initiated, it
may take up to one minute or longer to fully apply
the brakes throughout a train having, for example,
100 cars. A full release of a brake system will
also take a minute or longer. The time required is
depended upon the length of the train and the
specific type of brake valves on the cars. Thus, a
common problem faced by the locomotive engineers to
determine the extent of brake to apply.
If an engineer is moving down a grade that
requires a brake application to maintain a safe
speed, a sufficient application must be made to
prevent the train from gaining too much speed. If
too much brake is applied, the train may slow down
and eventually stop or stall. Ideally, sufficient
brake is applied to balance the effect of the grade.
If too much brake is applied, there are two choices.
The brakes can be released and re-applied, which
will take two or more minutes. This may result in
the train gaining too much speed before the brakes
are reapplied and therefore being a "runaway". The,
other alternative is to apply the locomotive power
CA 02531842 1998-09-11
-29-
propulsion to compensate for the excessive brake
application. This is known as "power braking"'.
Although resulting in a reasonable control of train
speed, it increased the cost of energy, fuel
consumption and wear on the brake system.
The LEADER system measures train speed, time,
track location and profile, locomotive throttle and
brake control settings, and the train makeup
determines the optimum brake application required to
maintain a safe speed. The safe speed may be
optimized in combination with a longitudinal and
lateral force level, minimum slack action and fuel
usage. Using standard freight train dynamic
modeling coupled with the measurements and data
bases, the LEADER predicts that brake control
settings will be required to maintain the speed or
to optimize the speed with other forces and fuel
usage. Depending upon the mode of operation, the
predicted brake settings can be conveyed to the
LEADER display for implementation by the engineer or
for control by the LEADER system directly.
A flow chart for optimizing the braking of the
train to maintain grade is illustrated in Figure 10.
LEADER monitors current train speed (V) and track
location (XLOC). LEADER determines the average
grade (G) over the ensuing distance to be traveled,
X. LEADER determines the average train rolling
resistance forces (FR) over the ensuing distance to
be traveled, X. Based on train weight (W), LEADER
calculates the force exerted on the train by the
grade (FG), FG=WG. If the average grade (G) is
downhill, LEADER computes the braking force (FB)
required to balance the grade (G), FB=(FG+FR).
CA 02531842 1998-09-11
-30-
Based on the calculated brake force (FH) required,
LEADER determines the brake control settings to
achieve this brake force. If the average grade is
not downhill, LEADER computes the propulsion force
required to maintain the speed (V). Based on the
propulsion force required to maintain speed, the
LEADER calculates train throttle settings to achieve
this force.
The control settings or the throttle settings
required to achieve the force will then be used with
a determination of whether there is automatic
control. If automatic control is not selected, the
settings are displayed. The determination is made
of whether the crew responded to the displayed
settings. If they do, then the system cycles back
to the beginning. If not, an enforcement is
produced. This enforcement could include a warning
or automatic control. If automatic control is
selected, then the appropriate settings are
actuated. The balancing grade may also be part of
the optimization routine of Figures 3, 4, 8 and/or
9.
The longitudinal forces which, if they exceed
the limit of the coupling system, will result in a
train braking in-two. Pulling a long train up a
grade against gravity and overcome the friction at
the car axles, may produce an acceleration or force
in the couplings which is unsafe. The determination
of unsafe forces is described in the flow chart of
Figure 11. The LEADER determines the forces using
its standard algorithms. The current force levels
include steady state draft and buff, transient draft
and buff and lateral over vertical force ratio.
CA 02531842 1998-09-11
-31-
These are compared with the limits set by the
railroad. If these calculated forces are outside
the limits, the determination is then made whether
the system is an automatic control or not. If it is
an automatic control, then the commands to actuate
forces are actuated. If it is not an automatic
control, then an alert is provided to the user. The
crew response is then determined. If the crew
affectuates the required setting, then the system
cycles back. If the crew does not actuate the
required setting, then an enforcement is actuated.
This enforcement again can be some visual alarm or
indication or automatic control. If they are not
outside the limits, the LEADER force calculation is
continued.
Also, this information can be used as in the
flow chart of Figures 3 and 4 as a weighted value or
by itself to achieve optimum braking_ It can be
used to control the braking as well as to control
the propulsion. By adjusting the propulsion system,
pulling the force exerted can be controlled.
In addition to the coupling forces produced by
the pulling between the cars, there is slack action
which may brake the coupler or damage the coupling
system. There is typically 4 - 6 inches of free
unrestrained movement in the coupling and even at
its relative speed of 1 - 2 miles an hour, can
create large forces. The prediction and control of
slack action is illustrated in the flow chart of
Figure 12. The LEADER measures train speed, time,
track location and track profiles, locomotive
throttle and brake settings and train make-up to
minimize slack related forces. Using standard
CA 02531842 1998-09-11
-32-
freight train dynamic modeling techniques, LEADER
calculates coupler forces and slack conditions
throughout the train. Next, it determines if slack
action is occurring. If not, the existing throttle
and brake settings are used in the display or for
control of the train.
If slack action is occurring, coupled with the
measurements and the stored data base information,
the LEADER predicts the train and brake control
settings that will minimize momentum changes between
the freight cars. Through experimentation it has
been discovered that the magnitude of slack related
forces can be effectively minimized by maintaining
near zero acceleration of the train as the slack
action is occurring. The LEADER determines the
magnitude of locomotive power application or brake
system application that is required to maintain near
zero train acceleration.
Next to determination is whether automatic
control is actuated. If it is, then the throttle
end are actuated to the calculated settings to
minimize slack action. If not, the LEADER displays
a throttle and/or brake settings to minimize slack
action. Next, the crew's response is measured. If
the crew sets the throttle and/or brakes to the
desired setting, then the system cycles back to the
beginning. If not, an enforcement is initiated.
Again, this may be an alarm or indicator that they
have not responded or it may automatic control.
Calculation control settings can either be
conveyed to the locomotive engineer by the LEADER
display for implementation or complete control
settings can be set. Trains having some or all EP
CA 02531842 1998-09-11
-33-
brake cars, slack action will be controlled either
at the locomotive only With propulsion and brake
pipe braking or in combination with EP braking on
the individual cars by sending additional electrical
signals to the individual cars to even out the slack
within the consist.
In addition to providing information or
controlling the train to be within speed limits, or
to optimize the performance of the train, it is also
important to provide information with respect to the
ability to come to a safe stop. The LEADER system
uses the measurement of train speed, time, track
location and characteristics, locomotive throttle
and brake control settings-and car consist
information to determine what is required to stop
the train from its current track location. The
LEADER uses standard freight train dynamic modeling
techniques, coupled with the measurements in data
base to accurately predict the distance required to
stop the train from its present track location. The
predicted stop distance can be conveyed to the
locomotive engineer and the LEADER display and to
ancillary control equipment of systems internal or
external to the locomotive. If the predicted stop
distance indicates tha~ the locomotive will enter a
restricted zone unless a stop is initiated, the
ancillary control equipment can automatically induce
the required brake application to bring the train to
a safe stop.
Figure 13 shows a flow chart for stop distance
calculations. LEADER monitors current train speed
(V) and track location (XLOC). It sets integration
time period, DT=1 sec, initialize stop distance,
CA 02531842 1998-09-11
-34-
XSTOP=0 and initialize atop time, TSTOP=0. Based on track
location (XLOC), speed (V) and train makeup, LEADER
calculates retardation forces imparted to the train (FR)
due to grades, curves, and rolling resistance effects, at
Time = TSTOP. Based on full service brake application,
LEADER calculates brake cylinder pressures on train, at
Time = TSTOP. LEADER calculation train brake retardation
(FB) forces from brake cylinder pressures, and car brake
rigging, at Time = TSTOP. Based on Newton's laws of motion
and train weight (W), LEADER calculates train deceleration
at time = TSTOP, A=K* (FB+FR)/W. LEADER integrates
calculated acceleration to determine speed at time = TSTOP,
V=V+A*DT. LEADER integrates speed to determine stopping
distance and track location at time = TSTOP, XSTOP = XSTOP +
V*DT. XLOC = XLOC + V*DT. TSTOP = TSTOP + DT. Next, it is
determined if velocity V = O. If yes, LEADER displays stop
distance XSTOP. If not, it loops back to begin the
calculation cycle.
The ability of the LEADER to provide information and
to record information allows the training or qualifying of
the locomotive engineer. As illustrated in Figure 14,
training can be performed by providing railroad training
parameters to the LEADER system. The LEADER system can
provide feedback display information to the LEADER display
screen of Figure 5 for guidelines of operation or
situations. The response of the engineer would then be
recorded by the LEADER system. The LEADER display would then
provide the result of the input with suggested corrections
or modifications. The LEADER system also has the ability to
store the history of the trip and the
CA 02531842 1998-09-11
-35-
operator's responses. This information can be
downloaded and used on a simulator, desktop or
portable PC system. This information can be
compared against desired responses to qualify the
engineer. Although such qualifications have
historically been performed on training systems, the
real life data provides a better setting for
qualification. This information may also be used
for train handling, performance analysis, accident
investigation and general operational studies.
Access to the control of the locomotives may be
determined by reading a user I.D. and determining
whether the user is qualified to use the particular
piece of equipment. This information may be on an
encoded device which the user of the locomotive must
insert into the system before the system will be
actuated. A flow chart of the system is illustrated
in Figure 15. The user's I.D. is read. A
determination is then made whether the user, through
his I.D. or through his qualification level is
approved for the particular equipment or locomotive.
If the engineer is not, the system is disabled. An
alarm may be set if desired. If the engineer is
qualified or approved for the equipment, the user
I.D. is recorded. The display and/or systems could
be customized for that-user if desired. If not,
this step can be deleted. The final step for an
improved user is to enable the systems for its use.
The encoded device which includes the user's
I.D. and their qualification may also be used with
the trainer of Figure 14. This training session
would determine, from the user I.D., their level
prior to the training session. Depending upon the
CA 02531842 1998-09-11
-36-
results and the qualification, this user's level
would be updated. Thus, the encoded device would
always carry the latest qualification level of the
user.
Although the present invention has been
described and illustrated in detail, it is to be
clearly understood that the same is by way of
illustration and example only, and is not to be
taken by way of limitation. The spirit and scope of
the present invention are to be limited only by the
terms of the appended claims.
