a2) United States Patent
`US 6,487,478 B1
`(10) Patent No.:
`Nov.26, 2002
`(45) Date of Patent:
`Azzaro et all.
`
`US006487478B1
`
`(54) ON-BOARD MONITOR FOR RAILROAD
`LOCOMOTIVE
`
`(75)
`
`Inventors: Steven Hector Azzaro, Schenectady,
`5
`we
`noine
`NY oe iinanhaben, Rextord,
`Y
`(US);
`Vinay
`Bhaskar
`Jammu,
`Niskayuna, NY (US), Thomas George
`Cook, Fairview, PA (US),Thomas
`Timothy Booth, Indialantic, FL (US)
`
`(73) Assignee: General Electric Company,
`Schenectady, NY (US)
`
`(*) Notice:
`
`Subject to any disclaimer,the term ofthis
`patent is extended or adjusted under 35
`US.C. 154(b) by 32 days.
`
`(21) Appl. No.: 09/696,368
`(22)
`Filed:
`Oct. 25, 2000
`
`(60)
`
`Related U.S. Application Data
`Provisional application No. 60/161,965, filed on Oct. 28,
`1999.
`_
`Tint. Cd ee eeeceteeenenresecenerevenee GO06F 19/00
`(51)
`(52) U.S. Che ceececececcccscsssssssersnene 701/24; 701/29; 701/31;
`340/3.1
`701/19, 24, 25
`(58) Field of Search
`701/29. 30, 31, 35,36: 246/122 R 103.
`eee '340/3 1: 370/316
`~~
`:
`
`(56)
`
`References Cited
`U.S. PATENT DOCUMENTS
`
`4,517,468 A
`4,561,057 A
`
`5/1985 Kemperet al.
`12/1985 Haley, Iv.etal.
`
`4,827,438 A
`4,885,689 A
`5,065,321 A
`pe 4
`35,590
`5,806,011 A
`5,845,272 A
`5,884,202 A *
`5,961,567 A
`6,264,950 BL *
`
`5/1989 Nickles et al.
`12/1989 Kaneetal.
`11/1991 Bezosetal.
`et007 pesos “ -
`/
`ezos et al.
`9/1998 Azzaro etal.
`12/1998 Morjariaetal.
`3/1999 Axjomand oo... 701/29
`10/1999 Azzaro etal.
`6/2001 Bessler et al. 0.0.0... 701/99
`
`OTHER PUBLICATIONS
`.
`. a
`Fry, K.N., BSc; Dicscl Locomotive Reliability Improvement
`By System Monitoring; British Rail Research, Railway
`Technical Centre, Derby; 1995; Proceedings of the Institute
`of Mechanical Engineers, vol. 209.
`* cited by examiner
`:
`Primary Examiner—William A. Cuchlinski, Jr.
`Assistant Fxaminer—Arthur D. Donnelly
`(74) Attorney, Agent, or Firm—JohnL. DeAngelis, Jr.; Carl
`A. Rowold; Beusse Brownlee Bowdoin & Wolter, P.A.
`(57)
`ABSTRACT
`
`An on-board monitor for a railroad locomotiveis disclosed.
`The on-board monitor interfaces with the controller sub-
`systemsof the locomotive to collect parametric performance
`data. The specific data to be collected and the collection
`intervals are defined at a remote service center and trans-
`mitted to the on-board monitor. The on-board monitor also
`includes the capability to collect additional data or collect
`data more frequently in response to the results of certain
`triggering events.
`
`37 Claims, 2 Drawing Sheets
`
` ¥
`
`~~ 104
`
`116 a
`18
`{
`© CALHOMES
`¥_INTIATE GALL
`2HOMESEOUENGE
`(Aw
`HOME SEGUENCE
`N
`122
`MODIFY DATA
`Y
`NA ORTRIGGER
`
`EQUATIONif|COLLECTIONmoO? PROCESS
`
`EX1010
`Petitioner Hum (223)
`
`Page 1 of 10
`
`POWER UF L~100
`
`¥
`EXCHANGE CONFIGURATION
`
`AND STARTUP INFORMATION [~~ 102
`
`
`
`SYNG TIME
`TRANSMIT
`i
`;
`POLLING \ y
`
`NFORMATICN
`REQUEST
`
`ae:
`N
`108
`
`5,
`
`¥
`
`M2 STORE
`
`
`
`
`
`
`
`
`
`106
`
`10~}
`
`
`
`
`Page 1 of 10
`
`EX1010
`Petitioner Hum (223)
`
`

`

`U.S. Patent
`
`Nov.26, 2002
`
`Sheet 1 of 2
`
`US 6,487,478 BI
`
`Sod1ddlM
`
`JOIAIA
`
`Ovv
`
`YOSSII0Ud
`
`NIV
`
`FOVAYFINISd9
`
`9g
`
`NolvaINnWWwod||||Yoss390ud0g
`
`YALNJOJIAUASYAMOd
`SOLLSONOVIGANLOWO9018ONIYOLINOW
`
`Ov/M9010HALSVIN
`
`
`AYOWIWWd
`Se
`
`dal
`
`AYOWAWHSV14
`FOVIUIINI59Sd74OVIYALNI
`
`YIMOd
`
`AlddN'
`
`OL
`
`ov
`
`/\
`
`Sdd
`
`YSAAIFOSY
`
`vi
`
`Al
`
`9
`
`BI
`
`XV
`
`Ox
`
`avo
`
`Osd
`
`NOIS1NdOud
`
`WALSAS
`
`Page 2 of 10
`
`
`Page 2 of 10
`
`
`
`
`
`
`

`

`U.S. Patent
`
`Nov.26, 2002
`
`Sheet 2 of2
`
`US 6,487,478 BI
`
`
`
`COLLECT
`DATA
`
`
`
`STORE DATA
`
`
`
`
`
`N
`
`
` CALL HOME
`
`
`
`
`
`MODIFY DATA
`
`COLLECTION
`
`
`PROCESS
`
`
`FAULT
`?
`
`
`
`INITIATE CALL
`HOME SEQUENCE
`
`ANOMALY
`OR TRIGGER
`EQUATION
`
`Page 3 of 10
`
`
`Page 3 of 10
`
`

`

`US 6,487,478 B1
`
`1
`ON-BOARD MONITOR FOR RAILROAD
`LOCOMOTIVE
`
`This application claims the benefit of U.S. Provisional
`Application No. 60/161965 filed on Oct. 28, 1999.
`BACKGROUND OF THE INVENTION
`
`The present invention is directed in general to monitoring
`performance and operational parameters and fault-related
`information on a railroad locomotive or other complex
`electromechanical system, and more specifically,
`to a
`method and apparatus for on-board monitoring of perfor-
`manccand fault-related parameters and transmission of the
`data collected to a monitoring and diagnosticsite.
`A railroad locomotive is one example of a complex
`electromechanical system comprised of several complex
`subsystems. Each of these subsystemsis built from compo-
`nents which overtime will fail. When a componentdoesfail,
`it may be difficult to determine the cause of the failed
`component because the effects or problemsthat the failure
`has on the subsystem are often neither readily apparent in
`terms of their source nor are they typically unique.
`The ability to automatically diagnose problems that have
`occurred or will occur in the locomotive subsystems has a
`positive impact on minimizing locomotive downtime. It is
`known that cost efficient operation of a railroad requires
`minimization of line-of-road failures and locomotive down
`time. Failure of a major locomotive subsystem can cause
`serious damage, costly repairs, and significant operational
`delays. A locomotive break-down while in service is an
`especially costly cvent, requiring the dispatch of a replace-
`ment
`locomotive to pull
`the train consist and possibly
`rendering a track segment out of service until the train is
`moved. As a result, the health of the locomotive engine and
`its constituent subsystemsis of significant concern.
`Previous attempts to diagnose problems once they have
`occurred on a locomotive usually involve performing
`inspections by experienced personnel who have in-depth
`individual training and experience in working with locomo-
`tives. Typically, these experienced individuals use available
`informationthat has been recordedin a log. Looking through
`the log, they use their accumulated experience and training
`in mapping incidents occurring in locomotive systems to
`problems that may be causing the incidents. If the incident-
`problem scenario is simple, then this approach worksfairly
`well. However, if the incident-problem scenario is complex,
`then it is very difficult to diagnose and correct anyfailures
`associated with the incidents.
`
`Currently, computer-bascd systems arc being used to
`automatically diagnose problemsin a locomotive in order to
`overcome someofthe disadvantages associated with relying
`completely on experienced personnel. Typically,
`a
`computer-based system utilizes a mapping between the
`observed symptomsof the failures and the equipment prob-
`lems using techniques such as table look ups, a symptom-
`problem matrices, and productionrules.
`There is also no automatic or systematic mechanism for
`the identification of incipient locomotive problems. Instead,
`conventionally, the railroads have relied on regular inspec-
`tions and the observation of performance anomalies bythe
`locomotive operator. Some cursory inspection processes are
`accomplished while the locomolive is in service; more
`thorough inspections require the locomotive to be taken out
`of service for several days. In any case, locomotive down
`time, whether for inspection or repair, represents a signifi-
`
`10
`
`1s
`
`40
`
`45
`
`50
`
`55
`
`60
`
`65
`
`2
`fault diagnosis and predictionofpotential failures represents
`an important cost saving, opportunily for the railroads.
`As a further means to reduce locomotive downtime, there
`has been a focus on the engineering design process with an
`objective of increasing the mean time between failures for
`locomotive subsystems and components. While this is cer-
`tainly a commendable objective, it remains for the railroads
`to continue their cost containment goals through the collec-
`tion and monitoring of real time performance data and fault
`related information directly from the locomotive, and the
`implementation of repairs before the problem requires sig-
`nificant downtime.
`
`BRIEF SUMMARY OF THE INVENTION
`The above-mentioneddifficulties associated with locomo-
`tive operations can be ameliorated by the present invention,
`which relates to a novel and unobvious apparatus and
`method for measuring performance andfault-related param-
`cters of the locomotive during operation. Monitoring the
`locomotive performance can provide timely and important
`indications of expected and immediate failures. With timely
`and continuous access to locomotive performancedata, it is
`possible to predict and/or prevent untimely failures.
`With recent advances in telecommunications
`technologies, it is now possible to collect information from
`a moving locomotive and transfer it to a fixed monitoring
`and diagnostic service center. With today’s advances in
`computing technology, the large amount of data collected
`from a fleet of locomotives can be properly aggregated and
`analyzed. The railroad can nowbetter understand the opera-
`tional and pertormance characteristicsof its individual loco-
`motives and the entire locomotive fleet. Analysis of this
`performance data can allow the railroad to advantageously
`predict and thereby avoid line-of-road failures.
`The present
`invention provides for the collection,
`aggregation, and communication of locomotive perfor-
`mance and fault-related data from an operational locomo-
`tive. Generally, anomalous or fault conditions will be
`brought to the attention of the locomotive operator directly
`by the control system, but the control systems generally lack
`the necessary hardware and software elements to self-
`diagnose the fault. After collection, the performance data is
`communicated to a remote monitoring and diagnostic site,
`where data analysis tools operate on the data to identify the
`source of potential or actual faults. The analysis tools may
`employ case-based or artificial intelligence strategies. In
`addition to computer-based analysis, human operators who
`are experts in locomotive operation and maintenance ana-
`lyze the data received. Historical data and patterns of
`anomalous behavior can be important clues to an accurate
`diagnosis and repair recommendation. The lessons learned
`from failure modes in a single locomotive can then be
`applied to other locomotivesof the class orto the entire fleet
`so that
`the necessary preventative maintenance can be
`performed. When the data analysis processidentifics incipi-
`ent problems, certain performance aspects of the locomotive
`can be derated to avoid further system degradation, and
`further limit violation of operating parameters.
`BRIEF DESCRIPTION OF THE DRAWINGS
`
`The present invention can be more easily understood and
`the further advantages and uses thereof more readily
`apparent, when considered in view of the description of the
`preferred embodiments and the following figures in which:
`FIG. 1 is a block diagram of an on-board monitor con-
`structed according to the teachings of the present invention;
`
`Page 4 of 10
`
`
`Page 4 of 10
`
`

`

`US 6,487,478 B1
`
`3
`FIG. 2 is a process flow chart illustrating the process
`executed by the on-board monitor of FIG. 1.
`DETAILED DESCRIPTION OF THE
`PREFERRED EMBODIMENTS
`
`Before describing in detail the particular on-board moni-
`tor in accordance with the present invention, it should be
`observed that the present invention resides primarily in a
`novel combination of processing steps and hardware related
`to an on-board monitor
`for a railroad locomotive.
`Accordingly, these processing steps and hardware compo-
`nents have been represented by conventional processes and
`clements in the drawings, showing only those specific
`details that are pertinent to the present invention, so as not
`to obscure the disclosure with structural details that will be
`readily apparent to those skilled in the art having the benefit
`of the description herein.
`The on-board monitor of the present invention resides in
`a locomotive and serves as a platform to gather data from
`operational locomotive control systems. This data provides
`important locomotive performance and status information
`that can be analyzed to identify active faults, predict incipi-
`ent failures and provide timely information about existing
`operating conditions. Once a failure has occurred, the data
`gathered by the on-board monitor of the present invention
`can also commanded from the monitoring and diagnostic
`center to collect additional information so that the locomo-
`tive experts can isolate the nature of the fault and develop
`the necessary repair recommendations.
`The on-board monitor
`is a signal acquisition,
`conditioning, data processing and logging instrument that
`provides status information to a monitoring and diagnostic
`service center. As will be discussed further below,
`the
`on-board monitor also has the capability to initiate a call to
`the monitoring and diagnostic service center whenever an
`incipient failure or existing condition ofsignificant severity
`is discovered. In the absence of such a severe problem, the
`locomotive status information is logged and periodically
`transferred to the monitoring and diagnostic service center
`via a wireless communications link.
`
`Turning to the FIG. 1, there is shown an on-board monitor
`10 constructed according to the teachings of the present
`invention. The on-board monitor 10 interfaces with several
`locomotive control systems that are employed by the loco-
`motive operator to control the locomotive. ‘he on-board
`monitor 10 is a passive device, Le., it does not interfere with
`operation of the various locomotive systems and processes
`the collected data independent of these systems.
`As is well known to those skilled in the art, within a
`locomotive, an auxiliary equipment controller 14, an exci-
`tation controller 16, and a cab controller 18 provide feed-
`back signals to and are in turn controlled by an integrated
`function controller 20. The integrated function controller 20
`is bidirectionally connected to an integrated function con-
`troller interface 22, which in turn is bidirectionally con-
`nected to a main processor 24 of the on-board monitor 10.
`The locomotivealso includes a propulsion system controller
`26, which interfaces bidirectionally with the main processor
`24 through a propulsion system controller interface 28. The
`propulsion system controller 26 controls the performance of
`the locomotive propulsion system, represented by a single
`block in the FIG. 1 and bearing reference character 27.
`‘The on-board monitor 10 includes a power supply 30 that
`derives its power from the locomotive power bus. In one
`embodiment, the on-board monitor 10 operates on a nominal
`
`10
`
`1s
`
`30
`
`40
`
`45
`
`50
`
`55
`
`60
`
`65
`
`4
`the on-board monitor 10 executes a safe power down
`process, causing all openfiles and ongoing tasks to close in
`an orderly fashion. After these operations are complete, the
`on-board monitor 10 powers down. The time required to
`perform this power-downtask is dependent on the number
`and type of tasks that are open when the power drops below
`the threshold value.
`the excitation
`The auxiliary equipment controller 14,
`controller 16,
`the cab controller 18 and the propulsion
`system controller 26 interface with transducers (not shown
`in the Figure) located within the respective controlled sub-
`systems of the locomotive and in turn control these sub-
`systems. These transducers measure,
`for
`instance,
`temperature, pressure, voltage, current, speed, and time
`intervals between specific events, and provide this informa-
`tion to the respective controller so that the locomotive can be
`properly operated, either by automatic or manual adjust-
`ments to the controlled systems.
`The main processor 24 of the on-board monitor 10 has
`two interfaces with a wireless communications device 40.
`Thefirst interface with the wireless communication device
`40 provides a communications link to a monitoring and
`diagnostic service center 42 via an antenna 44. The com-
`munications link carries data, commands, and configuration
`information between the monitoring and diagnostic service
`center 42 and the on-board monitor 10. The second interface
`with the wireless communication device 40 provides the
`main processor 24 with device status information concern-
`ing the wireless communication device 40 and the antenna
`44, and further provides signal strength information as
`related to an active or proposed communications link with
`the monitoring and diagnostic service center. The on-board
`monitor 10, through the main processor 24, also communi-
`cations with the global positioning satellite system (GPS),
`through a GPS receiver 46. The GPS system provides
`locomotive location and time of day information.
`The main processor 24 includes a processor 50, a flash
`memory 52, and a RAM memory 54. These elements are
`connected in the conventional manner as is well knownto
`those in the art,
`to form a computer system. The main
`processor 24 also includes a master clock/GPSinterface 56,
`which keeps clock time for the on-board monitor 10,
`in
`response to time signals received by the GPS receiver 46.
`Flash memory is non-volatile and can therefore retain data
`when power is removed from the on-board monitor 10.
`The main processor 24 and its constituent elements pro-
`vide the primary functionality for the on-board monitor 10.
`This functionality includes employing various parametric
`sensors to gather data from the integrated function controller
`20 and the propulsion system controller 26, processing and
`storing that data, and finally downloading the data to the
`monitoring and diagnostic service center 42 either automati-
`cally or upon receipt of instructions therefrom. The data
`collected according to the teachings of the present invention
`can include any parameters indicative of operation of the
`railroad locomotive. or example, voltages, currents,
`temperatures, pressures,
`fluid levels,
`fluid flow rates,
`weights, forces, relative time and time intervals and the
`position of operator selectable devices. In short, any mea-
`surable parameters indicative of system or subsystem per-
`formance can he collected by the on-board monitor 10. In
`the absence of specific download commands, the informa-
`tion collected by the on-board monitor 10 is downloaded to
`the monitoring and diagnostic service center on a regular
`and periodic basis. Additionally,
`in response to certain
`parametric or fault-related data parameters and conditions
`
`Page 5 of 10
`
`
`Page 5 of 10
`
`

`

`US 6,487,478 B1
`
`5
`10 automatically contacts the monitoring and diagnostic
`service center 42 for transferring the relevant data or for
`receiving instructions (for example, commands to collect
`additional data or
`the same data more frequently).
`Alternatively,
`the on-board monitor 10 may contact
`the
`monitoring and diagnostic service center 42 (e.g., send a
`flag) and request that the center immediately call the loco-
`motive (specifically the on-board monitor 10) for the pur-
`pose of downloading the relevant data from the on-board
`monitor 10 to the monitoring and diagnostic service center
`42. In response to the flag, the monitoring and diagnostic
`service centers transmit an acknowledgment signal to the
`on-board monitor 10, and the on-board monitor 10 then
`sends the relevant data.
`
`In one embodiment, the on-board monitor 10 synchro-
`nizes the current time with the time provided by the global
`positioning system (via the GPS receiver 46) every five
`seconds. Internal
`time is kept
`in the master clock/GPS
`interface 56. In the event the on-board monitor LO cannot
`
`10
`
`1s
`
`25
`
`30
`
`6
`it an identifier as to whether the value is to be converted to
`allernalive units.
`
`the collection and transmittal of
`As discussed above,
`parametric data is the primary function of the on-board
`monitor 10. This data file is downloaded by the on-board
`monitor LO to the monitoring and diagnostic service center
`42, where it is analyzed to detect active and incipient faults
`and used to generate repair recommendations.
`The time interval between gathering these performance
`statistics is a configurable item, as defined in the configu-
`ration file for
`the specific parametric data.
`In one
`embodiment,
`the interval between data gathering is one
`hour. The monitoring and diagnostic service center 42
`controls the rate at which the operational parameters are
`gathered by changing the time interval value in the upload
`start-up file.
`The uploadstart-up file also defines certain fault condi-
`tions as determined by specified parametric data valucs or
`performance anomalies. An active fault log file and a reset
`fault log file of the on-board monitor 10 store information
`concerning the faults, and link this information to the
`geographical position of the locomotive at the onset of the
`fault, through the GPS receiver 46. Two files are required so
`that a record can be kept of the occurrence time and thereset
`time of the fault. The active fault log contains an identifi-
`cation of the specific fault, the time at which it occurred, the
`time the fault was reset, and certain locomotive operating
`conditions when the fault occurred. The reset fault
`log
`contains an identification of the fault, the time at which it
`occurred, and the time at whichit wasreset by the respective
`locomotive controller.
`
`the integrated function
`To create the active fault log,
`controller 20 or the propulsion systemcontroller interface 28
`sends a message to the on-board monitor 10 whenevera fault
`occurs or is reset. Upon receipt, the on-board monitor 10
`determinesif the fault is a new one, in which case it gathers
`information including an identification of the fault, the GPS
`location upon onset, and locomotive performance data that
`is pertinent to the fault and the time of occurrence. This
`informationis stored in the active fault log. If the event is a
`fault reset, the on-board monitor 10 gathers the fault iden-
`tification information, reset time and the geographical loca-
`tion at reset, which is stored in the reset fault file. The
`on-board monitor 10 also tracks the number of active and
`reset faults, the most recent fault, and the time at which the
`most recent fault occurred.
`
`As a further check, the on-board monitor 10 periodically
`requests fault status information from the integrated function
`controller 20 (and the propulsion system controller interface
`28) to ensure that both units show the same number of
`generated faults. Typically, these check requests are made
`every hour and both active faults and reset faults are tracked.
`If the fault status numbers between the on-board monitor 10
`
`and the integrated function controller 20 (or the propulsion
`system controller interface 28) differ, the on-board monitor
`10 issues a fault data request to the integrated function
`controller 20 for information on the most recent fault. If
`status informationstill does not match, the on-board monitor
`10 continues to scroll back throughthe list of faults until the
`stored fault and time of its occurrence match the most
`recently requested fault data, or until the scrolling process
`reaches the end of the list.
`
`Accritical faults file is a list uploaded from the monitoring
`and diagnostic service center 42 to the on-board monitor 10.
`The listed faults are those that are of sufficient severity to
`
`the 4
`synchronize with the global positioning system at
`required interval (for example, because the GPS receiver 46
`cannot close a link with a GPSsatellite), the master clock/
`GPSinterface 56 continues a zero-based time count until the
`next synchronization occurs. Also, in the event that the GPS
`time is not available when the on-board monitor 10 is
`powered-up, the main processor 24 establishes a zero-based
`time. The current time, as stored in the master clock/GPS
`interface 56 is used to time stamp the data and performance
`parameters collected by the on-board monitor 10.
`The on-board monitor 10 uses configurable software files
`to control various aspects of its operation, including iden-
`tification of the information to be gathered and the manner
`of storing, processing and uploading that
`information.
`Generally, these files include various startup and configu-
`rationfiles that are exchanged betweenthe on-board monitor
`10 and the monitoring and diagnostic service center 42. The
`data within these files performs several functions including,
`ensuring that the locomotive road number in which the
`on-board monitor 10 is installed matches the configuration at
`the monitoring and diagnostic service center 42. In this way,
`information downloaded from a specific on-board monitor
`10 will be properly identified as providing performance
`information for a specific locomotive road number. Addi-
`tional information included within these configuration files
`is the software versions of the various software programs
`running on the on-board monitor 10, the integrated function
`controller 20 and the propulsion system controller 26.
`Additionally, the configurationfiles identify the parametric
`operational information to be collected and the statistical
`analysis to be conducted on the retrieved data. Further
`attributes of the configuration files will be discussed herein
`below asrelated to the data collection process.
`The parametric data file, which is a file periodically
`downloaded by the on-board monitor 10 to the monitoring
`and diagnostic service center 42, contains various locomo-
`tive performancestatistics and operational parameters col-
`lected from the integrated function controller 20 and the
`propulsion system controller 26. In one embodiment, these
`performancestatistics include:
`total locomotive operating
`time, time spent at idle, time spent at each operating level
`(throttle notch position or dynamic brake step position), and
`horsepower output. Parametric data collected and transmit-
`ted to the monitoring and diagnostic service center includes
`temperature, pressure, voltage, and current of the various
`systems and subsystems of the locomotive. All parametric
`data gathered is identified by the locomotive controller
`
`40
`
`45
`
`50
`
`55
`
`60
`
`65
`
`Page 6 of 10
`
`
`Page 6 of 10
`
`

`

`US 6,487,478 B1
`
`8
`7
`Each anomalytrigger equation has an associated configu-
`monitoring and diagnostic service center 42 whenever that
`fault occurs. Each fault onthe list is associated with a timer
`rationfile that stores the following information: the trigger
`value, which is configurable, and in one embodimentis one
`identification (an integer to identify each specific trigger
`hour. When the on-board monitor 10 initiates a call to the
`equation), the length of time that the trigger equation must
`monitoring and diagnostic service center 42 due to the
`be true before data anomaly-driven collection begins, the
`
`occurrence of suchacritical fault, the timer begins its count number of data samples to be collected, the trigger count
`down process. Another call home for that particular fault
`down time, a list of data parameters to be collected when the
`cannot be initiated until the counter has expired. If another
`trigger is true, and a specific trigger equation. An exemplary
`instantiation of that specific critical fault occurs while the
`trigger equation maybe: (the engine RPM is greater than
`timeris in its count down mode, the fault is entered into the
`value onc) AND (the locomotive speed is greater than valuc
`fault log, but a call home is not initiated.
`two) AND(the oil temperature is greater than value three)
`Whenever a new fault
`is identified by the on-board
`OR(the engine RPM is greater than value one) AND (the
`monitor 10, it is compared toalist of critical faults. If the
`locomotive speed is greater than value two) AND (the water
`new fault is a critical fault, the critical fault timer is checked
`temperature is greater than value four). Using exemplary
`to determine whether it is clear. If the timer is clear, the
`parametric values, mathematically this may be expressedas:
`on-board monitor 10 initiates a call back to the monitoring
`(1FC1234>1040)4+(1FC1235>45)+(LFC1236>170)|
`and diagnostic service center 42.
`(IFC1234>1040)+(IFC1235>45)+(IFC1237>200).
`The on-board monitor 10 also includes a signal strength
`When the on-board monitor 10 collects performance
`file for plotting wireless communication satellite signal
`parameters in response to a satisfied trigger equation, the
`strength against the geographical location (as determined
`performance data is segregated into data blocks, with one
`from the GPS receiver 46). The data in this file can later be
`data block for each monitored device. The following statis-
`analyzed to gain a better understanding of anysituations
`tics are then calculated for each data block: the maximum
`where the on-board monitor 10 was unable to close a
`and mioimum values, the mean value of the data block, the
`communications link with the monitoring and diagnostic
`standard deviation of the data block, and the median value
`service center 42 or where an active link dropped out of
`of the data block. The medianis defined as the middle value
`25
`service. ‘his file also includes the time the signal strength
`of n data points when arranged in increasing order andnis
`sample was gathered. The collection rate for the signal
`odd; and is defined as the mean of the two middle values of
`strength information is set in a configuration file for the
`n data point when arrangedin increasing order andnis even.
`signal strength file. ‘l'ypically, this time is set at every ten
`The last two statistical values collected are the first value in
`minutes.
`the data block and the last value in the data block, when n
`data points are arranged in increasing order bytime of
`collection.
`
`10
`
`30
`
`In addition to the predetermined data collection process
`discussed above, the on-board monitor 10 executes a special
`data collection process in response to certain performance
`anomalies as determined by certain anomalytrigger equa-
`tions. ‘he anomaly trigger equations are developed at the
`monitoring and diagnostic service center 42 and uploaded to
`the on-board monitor 10. These trigger equations define
`locomotive performance conditions (limits or ranges for
`relevant operational parameters) requiring unique data col-
`Icction actions and also specify the specific data to be
`collected. As discussed above,
`the on-board monitor 10
`periodically gathcrs paramctric information from the inte-
`grated function controller 20 and the propulsion system
`controller 26. In addition to storing this information in the
`parametric data file, the gathered data is also examined to
`determine whether it satisfies any of the anomaly trigger
`equations. When a trigger condition is satisfied, additional
`performance data is gathered from the integrated function
`controller 20 and the propulsion system controller 26, as
`directed by the instructions for that anomalytrigger equa-
`tion. This additional data gathering process continues as
`long as the trigger condition remainstrue. In response to an
`anomalytrigger, the on-board monitor 10 maybeinstructed,
`as set forth in the anomaly trigger information, to gather
`certain non-numeric data such as locomotive control settings
`or callect certain information that is not otherwise collected
`in the absence of the occurrence of the anomaly. The
`information gathered during an anomaly event may also be
`aggregated bycalculating variousstatistical metrics for the
`data, as discussed below. The results of the statistical cal-
`culations are written to the anomalystatistics file, along with
`the GPS location information, the numeric and non-numeric
`data, and an identification of the associated anomalytrigger.
`Whenever an anomaly equation trigger is satisfied, a
`trigger count down timer is also activated.
`‘That
`trigger
`equation is not permitted to trigger again until the timer has
`expired. The value set in the count downtimeris defined in
`
`The on-board monitor 10 also includes the ability to
`collect data based on customized data trigger equations. The
`custom data trigger equations can be individualized for a
`specific railroad or a specific locomotive. These equations
`can also be customized based on operating conditions of the
`locomotive. For example, during the winter season, custom
`data equations can be used to collect temperature statistics
`that may not be needed during the summer months. Any
`performance parameter that is accessible to the on-board
`monitor 10, through the integrated function controller 20 or
`the propulsion system controller 26, can be the subject of a
`custom data trigger equation. A call-home feature can also
`be associated with a custom trigger equation, allowing the
`on-board monitor 10 to call home when the data gathering
`for a custom data trigger equation has been completed or if
`a parameter is beyond a threshold limit. The monitoring and
`diagnostic service center 42 defines the number of samples
`to be gathered and the time interval over which the data
`collection occurs, for each custom data trigger equation, via
`the configuration file. Also, the call-home feature can be
`turned on or off on command from the monitoring and
`diagnostic service center 42.
`Specifically,
`the customized data trigger equation file
`uploaded from the monitoring and diagnostic service center
`42 to the on-board monitor 10 includes: an identification
`integer used t