EP4344978A1 - Dispositif d'estimation - Google Patents

Dispositif d'estimation Download PDF

Info

Publication number
EP4344978A1
EP4344978A1 EP23197677.0A EP23197677A EP4344978A1 EP 4344978 A1 EP4344978 A1 EP 4344978A1 EP 23197677 A EP23197677 A EP 23197677A EP 4344978 A1 EP4344978 A1 EP 4344978A1
Authority
EP
European Patent Office
Prior art keywords
value
vehicle
axle
band
acceleration
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
EP23197677.0A
Other languages
German (de)
English (en)
Inventor
Daniel EBERLE
Markus Knoll
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Siemens Mobility GmbH
Original Assignee
Siemens Mobility GmbH
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Siemens Mobility GmbH filed Critical Siemens Mobility GmbH
Publication of EP4344978A1 publication Critical patent/EP4344978A1/fr
Pending legal-status Critical Current

Links

Images

Classifications

    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B61—RAILWAYS
    • B61L—GUIDING RAILWAY TRAFFIC; ENSURING THE SAFETY OF RAILWAY TRAFFIC
    • B61L25/00—Recording or indicating positions or identities of vehicles or trains or setting of track apparatus
    • B61L25/02—Indicating or recording positions or identities of vehicles or trains
    • B61L25/021—Measuring and recording of train speed
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B61—RAILWAYS
    • B61L—GUIDING RAILWAY TRAFFIC; ENSURING THE SAFETY OF RAILWAY TRAFFIC
    • B61L15/00—Indicators provided on the vehicle or train for signalling purposes
    • B61L15/0072—On-board train data handling

Definitions

  • the invention is based on the object of specifying an estimation device for estimating a movement value describing a movement quantity of a rail vehicle, which takes into account the above problems of the complex overall situation.
  • the estimation device comprises a trained artificial intelligence system which calculates a reference value based on measured values, and the trained system is provided with a correction device which checks the reference value and outputs it as the movement value of the estimator if the reference value passes a plausibility check, and corrects it to form the movement value of the estimator if the reference value does not pass the plausibility check.
  • a significant advantage of the estimating device according to the invention is that it is designed in at least two stages.
  • a first stage is formed by a trained artificial intelligence system that calculates the movement value based on measured values.
  • the training of artificial intelligence systems is now generally known, so reference should be made to the relevant specialist literature.
  • the training of the first stage or the artificial intelligence system is preferably carried out on the basis of measured values from previous measurement steps or on the basis of simulation results.
  • the accuracy or security of the movement value generated by the first stage may, under certain circumstances, be uncertain or even error-prone beyond a permissible level, for example if a previously insufficiently trained wheel-rail adhesion situation is combined with an intervention of a sliding and Anti-skid control occurs.
  • the second stage is provided in the form of the correction device, which checks the movement values of the first stage for plausibility. The second stage thus makes it possible to bring the entire system or the estimation device to a high level of security without much effort.
  • the correction device takes into account at least two axle speed values, each of which indicates the axle speed of an associated axle of the rail vehicle.
  • the movement quantity is the vehicle speed and the reference value of the trained system and the movement value of the estimation device are vehicle speed values.
  • the correction device considers the vehicle speed value of the trained system to be implausible if the vehicle speed value is smaller than the highest existing axle speed value during a braking process and is greater than the smallest existing axle speed value during an acceleration process.
  • the correction device considers the vehicle speed value of the trained system to be implausible if the current vehicle speed value during a braking process is smaller than an updated maximum axle speed, the fall speed of which is limited by a predetermined maximum vehicle deceleration value, the correction device in this case the vehicle speed value of the Trained system is then preferably corrected to this limited axis speed.
  • the correction device considers the vehicle speed value of the trained system to be implausible if the current vehicle speed value during an acceleration process is greater than a continued minimum axle speed, the rate of increase of which is limited by a predetermined maximum vehicle acceleration value, wherein the correction device In this case, the vehicle speed value of the trained system is then preferably corrected to this limited axle speed.
  • the correction device considers the vehicle speed value of the trained system to be implausible if the current vehicle speed value during a braking process is smaller than an updated highest axle speed, the updated highest axle speed taking into account a predetermined vehicle deceleration maximum value and the last highest axle speed value , which was classified as plausible, is calculated; The correction device then preferably corrects the vehicle speed value of the trained system to this updated highest axle speed.
  • the correction device considers the vehicle speed value of the trained system to be implausible if the current vehicle speed value of the trained system is greater than an updated minimum axle speed during an acceleration process, the updated minimum axle speed taking into account a predetermined vehicle acceleration maximum value and the last smallest axis speed value that was considered plausible is calculated; The correction device then preferably corrects the vehicle speed value of the trained system to this updated minimum axle speed.
  • the correction device preferably processes an input signal that describes the wheel grip on the rail, For example, it shows whether the wheel grip reaches or exceeds a specified minimum grip level.
  • the correction device takes the wheel grip on the rail into account and determines an axle-related band using the largest and smallest existing axle speed values. If the conditions are good, it considers the vehicle speed value of the trained system to be implausible and sets this to a band value (e.g B. Limit of the band or value within the band) of the axle-related band is corrected if the vehicle speed value is outside this axle-related band.
  • a band value e.g B. Limit of the band or value within the band
  • the largest axis speed value preferably forms the maximum value or the upper band limit of the axis-related band;
  • the smallest axis speed value preferably forms the minimum value or the lower band limit of the axis-related band.
  • the correction device preferably corrects the vehicle speed value of the trained system to the highest available axle speed value, i.e. to the maximum band value, if the vehicle speed value is outside the axle-related band during a braking operation.
  • the correction device corrects the vehicle speed value to the middle of the axle-related band if the vehicle speed value of the trained system - regardless of the vehicle condition (accelerating, braking, rolling) - lies outside the axle-related band; Because under good conditions, all axis speeds largely correspond to the real ones (apart from measurement errors and noise). Vehicle speed, so that the axle-related band is very narrow.
  • the correction device preferably corrects the vehicle speed value of the trained system to the smallest available axle speed value, i.e. to the band minimum value, if the vehicle speed value is outside the axle-related band during an acceleration process.
  • the correction device corrects the vehicle speed value of the trained system taking into account at least one of the axle speed values which relates to an axle which is force-free for a predetermined minimum period of time.
  • the correction device in addition to receiving at least one axis speed value relating to an axis that is force-free for a predetermined minimum period of time, the correction device can preferably generate a control signal with which the axis is switched force-free for the predetermined minimum period of time.
  • the correction device determines a vehicle-related brake band when braking by updating the last vehicle speed value, which was considered plausible in the presence of good adhesion conditions or was determined by the correction device through correction, taking into account a vehicle deceleration maximum value and a vehicle deceleration minimum value .
  • the correction device preferably corrects the vehicle speed value of the trained system by When braking, the vehicle speed value is reduced to the maximum value of the brake band if the vehicle speed value exceeds the maximum value of the brake band, and increased to the minimum value of the brake band if the vehicle speed value is below the minimum value of the brake band.
  • the correction device determines a vehicle-related acceleration band when accelerating by taking into account the last vehicle speed value, which was considered plausible in the presence of good detention conditions or was determined by the correction device through correction, taking into account a vehicle acceleration maximum value and a vehicle acceleration minimum value continues.
  • the correction device preferably corrects the vehicle speed value of the trained system by reducing the vehicle speed value to the maximum value of the acceleration band when accelerating, if the vehicle speed value exceeds the maximum value of the acceleration band, and increasing it to the minimum value of the acceleration band, if the vehicle speed value falls below the minimum value of the acceleration band.
  • the correction device calculates the minimum vehicle deceleration value and the minimum vehicle acceleration value preferably as a function of at least one acceleration measurement value.
  • the movement size vehicle acceleration and the reference value of the learned system and the motion value of the estimator are vehicle acceleration values.
  • the correction device preferentially considers the vehicle acceleration value of the trained system to be plausible if the vehicle acceleration value lies within a vehicle-related band.
  • the correction device limits the amount of the vehicle acceleration value of the trained system - at least in poor conditions - preferably downwards to a minimum value (lower band limit) of the predetermined vehicle-related band and upwards to a maximum value (upper band limit) of the vehicle-related band if the vehicle acceleration value is outside the vehicle-related band lies.
  • the minimum and maximum values of the vehicle-related band preferably depend on whether the vehicle is accelerating or decelerating.
  • the minimum value of the vehicle-related band corresponds to an expected minimum deceleration of the rail vehicle when braking and to an expected minimum acceleration of the rail vehicle when accelerating;
  • the maximum value of the vehicle-related band corresponds to an expected maximum deceleration of the rail vehicle when braking and to an expected maximum acceleration of the rail vehicle when accelerating.
  • the correction device takes into account at least two axle speed values, each of which indicates the axle speed of an assigned axle of the rail vehicle, and calculates axle acceleration values using the axle speed values, the largest calculated axle acceleration value and the The smallest calculated axis acceleration value forms an axis-related acceleration band.
  • the correction device preferably sees the vehicle acceleration value of the trained system as plausible if the vehicle acceleration value is in the axle-related acceleration band. Otherwise, if the vehicle acceleration value is not in the axle-related acceleration band and is therefore implausible, it corrects the vehicle acceleration value of the trained system preferably to a band value (limit or value within the band) of the axle-related band.
  • the correction device for the plausibility check preferably takes into account an intermediate band which is increased from the axle-related acceleration band to the vehicle-related band taking into account a predetermined maximum jerk value.
  • the correction device corrects the vehicle acceleration value of the trained system taking into account at least one of the axle acceleration values that relates to an axle that is force-free for a predetermined minimum period of time.
  • the correction device can preferably generate a control signal with which the axle is switched to force-free for the predetermined minimum period of time.
  • the invention also relates to a method for estimating a movement quantity of a rail vehicle descriptive movement value.
  • a reference value is calculated using a trained artificial intelligence system on the basis of measured values, the reference value is left unchanged and output as a movement value if the reference value passes a plausibility check, and the reference value is corrected to form the movement value if the reference value does not pass the plausibility check.
  • the invention also relates to a rail vehicle.
  • the rail vehicle has an estimation device - as described above - and/or is designed to carry out a method as described above.
  • the invention also relates to a computer program product.
  • the computer program product is provided so that, when executed by a computing device, the computing device causes the computing device to carry out a method as described above and/or to form an estimation device as described above with the computing device.
  • the Figure 1 shows an exemplary embodiment of an estimating device 10 according to the invention for determining a movement value describing the speed of a rail vehicle in the form of a speed estimated value Vfinal.
  • the estimation device comprises a trained artificial intelligence system 11 which calculates a vehicle speed value Vki using measured values E as a reference value. Downstream of the trained artificial intelligence system 11 is a correction device 12 which subjects the vehicle speed value Vki of the trained system 11 to a plausibility check, taking into account measured values M, such as axle speed measured values and/or axle acceleration measured values.
  • the measured values E, which the trained system 11 processes, and the measured values M, which the correction device 12 processes, can be identical; alternatively, the measured values E and the measured values M can differ.
  • the correction device 12 If the plausibility test is passed, the correction device 12 outputs the vehicle speed value Vki of the trained system 11 as the final movement value in the form of the estimated speed value Vfinal of the estimating device 10. If the vehicle speed value Vki of the trained system 11 does not pass the plausibility check, the correction device 12 corrects the vehicle speed value Vki and outputs the corrected vehicle speed value as the estimated speed value Vfinal of the estimating device 10.
  • the Figure 2 shows an advantageous technical realization of the estimation device 10 according to Figure 1
  • the estimation device 10 is formed by a computing device 100 and a memory 110.
  • a software module 111 is stored in the memory 110, which forms the trained system 11 when executed by the computing device 100.
  • a software module 112 stored in the memory 110 forms the correction device 12 when executed by the computing device 100.
  • the Figure 3 explains an advantageous operation of the correction device 12 using an exemplary braking process.
  • the correction device 12 processes M axis speed values V1, V2, V3 and V4 as measured values, which indicate the measured axis speeds of four axes of the rail vehicle.
  • the correction device 12 determines a speed band 200, which is limited by a band minimum value 201 and a band maximum value 202.
  • the output speed value Vout can be the largest of the four axis speed values V1, V2, V3 and V4.
  • the band minimum value 201 of the speed band 200 is initially calculated by updating the initial speed value Vout taking into account a maximum vehicle deceleration value assumed to be realistic for the rail vehicle.
  • the band maximum value 202 of the speed band 200 is calculated by updating the initial speed value Vout taking into account a minimum vehicle deceleration value assumed to be realistic for the rail vehicle.
  • axle speed values V1 to V4 are below the speed band 200 in the first time period, for example because there is slippage between the wheel and the rail due to poor adhesion; However, this plays no role for the speed band 200 until time t1.
  • the axle speed value V1 of the fastest axle exceeds the band minimum value 201.
  • the rail vehicle is actually faster than would be the case with maximum vehicle deceleration; the rail vehicle is therefore not braked as quickly as is assumed to be the maximum possible.
  • the correction device 12 now modifies the speed band 200 by raising the band minimum value 201 to the axis speed value V1 of the fastest axis.
  • the correction device 12 Since in the second time period between the times t1 and t2 the vehicle speed value Vki of the trained system 11 is still in the speed band 200, the correction device 12 continues to regard the vehicle speed value Vki as plausible and outputs it as the speed estimate Vfinal of the estimation device 10.
  • the vehicle speed value Vki reaches the band minimum value 201. Since the correction device 12 assumes that the speed of the rail vehicle cannot be less than the highest axle speed value V1 or the band minimum value 201, the correction device 12 now corrects the vehicle speed value Vki of the trained system 11 by raising it to the band minimum value 201. In the third time period between times t2 and t3, the band minimum value 201 is thus output as the speed estimate value Vfinal of the estimation device 10.
  • the Figure 4 shows an exemplary embodiment of an estimating device 10 according to the invention, in which a trained artificial intelligence system 11 calculates a vehicle acceleration value aki based on measured values E as a reference value. Downstream of the trained system 11 is a correction device 12, which subjects the vehicle acceleration value aki to a plausibility check, taking into account measured values M such as axle speed measurement values and / or axle acceleration measurement values, and corrects the vehicle acceleration value aki if it is considered implausible, and otherwise uncorrected as the vehicle acceleration estimate afinal der Estimator 10 outputs.
  • a correction device 12 Downstream of the trained system 11 calculates a vehicle acceleration value aki based on measured values E as a reference value.
  • a correction device 12 Downstream of the trained system 11 is a correction device 12, which subjects the vehicle acceleration value aki to a plausibility check, taking into account measured values M such as axle speed measurement values and / or axle acceleration measurement values, and corrects the vehicle acceleration value aki if it is considered implausible, and otherwise uncorrect
  • the correction device 12 preferably sees the vehicle acceleration value aki as plausible if the vehicle acceleration value aki is in an axle-related acceleration band. If it is not in this axle-related acceleration band, the correction device 12 corrects the vehicle acceleration value aki, for example by correcting it to a band value of the axle-related acceleration band.
  • the correction device 12 preferably considers the vehicle acceleration value aki to be plausible if it lies in a vehicle-related acceleration band. If it is not in the vehicle-related acceleration band, the correction device 12 corrects the vehicle acceleration value aki by correcting it, for example, to a band value of the vehicle-related acceleration band.
  • the correction device 12 for the plausibility check preferably takes into account an intermediate band, which is increased from the axle-related acceleration band to the vehicle-related band, taking into account a predetermined maximum jerk value.
  • the Figure 5 shows an embodiment of a rail vehicle 500 according to the invention, which is equipped with an estimation device 10, for example an estimation device according to Figure 1, 2 or 4 , is equipped.
  • the estimation device is preferably integrated in a vehicle control unit 510 of the rail vehicle 500.

Landscapes

  • Engineering & Computer Science (AREA)
  • Mechanical Engineering (AREA)
  • Regulating Braking Force (AREA)
EP23197677.0A 2022-09-29 2023-09-15 Dispositif d'estimation Pending EP4344978A1 (fr)

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
DE102022210315.7A DE102022210315A1 (de) 2022-09-29 2022-09-29 Schätzeinrichtung

Publications (1)

Publication Number Publication Date
EP4344978A1 true EP4344978A1 (fr) 2024-04-03

Family

ID=88092929

Family Applications (1)

Application Number Title Priority Date Filing Date
EP23197677.0A Pending EP4344978A1 (fr) 2022-09-29 2023-09-15 Dispositif d'estimation

Country Status (2)

Country Link
EP (1) EP4344978A1 (fr)
DE (1) DE102022210315A1 (fr)

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE102013201594A1 (de) * 2013-01-31 2014-09-11 Siemens Aktiengesellschaft Verfahren und Vorrichtung zur Genauigkeitsbewertung von Geschwindigkeitswerten, insbesondere eines Schienenfahrzeuges
DE102016005248A1 (de) * 2016-04-29 2017-11-02 Knorr-Bremse Systeme für Schienenfahrzeuge GmbH Vorrichtung und Verfahren zur Optimierung der Kaftschlussausnutzung zwischen Rad und Schiene
EP3663769A1 (fr) * 2018-12-09 2020-06-10 MagSens Unternehmergesellschaft haftungsbeschränkt Dispositif de détermination la vitesse de véhicules

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE4104775A1 (de) 1991-02-13 1992-08-20 Knorr Bremse Berlin Ag Verfahren zur ermittlung der geschwindigkeit eines fahrzeuges mit schlupfgeregelten raedern
DE102013210361A1 (de) 2013-06-04 2014-12-04 Siemens Aktiengesellschaft Verfahren zur Ermittlung zumindest einer Geschwindigkeit bei einem Schienenfahrzeug
DE102019211944A1 (de) 2019-08-08 2021-02-11 Knorr-Bremse Systeme für Schienenfahrzeuge GmbH Verfahren zur Bestimmung einer lokalen Wagengeschwindigkeit eines Wagens
IT202000005671A1 (it) 2020-03-17 2021-09-17 Faiveley Transport Italia Spa Procedimento per la stima di una velocità longitudinale di almeno un veicolo ferroviario

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE102013201594A1 (de) * 2013-01-31 2014-09-11 Siemens Aktiengesellschaft Verfahren und Vorrichtung zur Genauigkeitsbewertung von Geschwindigkeitswerten, insbesondere eines Schienenfahrzeuges
DE102016005248A1 (de) * 2016-04-29 2017-11-02 Knorr-Bremse Systeme für Schienenfahrzeuge GmbH Vorrichtung und Verfahren zur Optimierung der Kaftschlussausnutzung zwischen Rad und Schiene
EP3663769A1 (fr) * 2018-12-09 2020-06-10 MagSens Unternehmergesellschaft haftungsbeschränkt Dispositif de détermination la vitesse de véhicules

Also Published As

Publication number Publication date
DE102022210315A1 (de) 2024-04-04

Similar Documents

Publication Publication Date Title
EP1149006B1 (fr) Procede et dispositif de surveillance de detecteurs, en particulier pour un systeme esp pour vehicules
EP0859960B1 (fr) Dispositif et procede de surveillance de capteurs dans un vehicule
DE69935471T2 (de) Gierwinkelregelverfahren für Kraftfahrzeuge
WO2002055356A1 (fr) Dispositif de controle de moyens detecteurs agences dans un vehicule
WO2009149862A1 (fr) Procédé de surveillance d'au moins un paramètre système influencé par le comportement de fonctionnement de véhicules ou de trains
DE19939872B4 (de) Verfahren und Vorrichtung zur Sensorüberwachung insbesondere für ein ESP-System für Fahrzeuge
DE69204172T2 (de) Einrichtung zur Berechnung der abgeschätzten Fahrzeugbeschleunigung und Geschwindigkeit.
DE102018112846A1 (de) Steuereinrichtung und Verfahren für die Ansteuerung eines Aktuators zur Betätigung von Bremsmitteln eines Fahrzeuges, insbesondere eines Schienenfahrzeuges
EP3983777B1 (fr) Procédé et dispositif de contrôle pour contrôler un véhicule
WO2020057874A9 (fr) Procédé pour la détermination de sauts et/ou de points d'inflexion dans une caractéristique d'actionnement d'une unité d'actionnement, module d'évaluation et véhicule
DE102007029605A1 (de) Verfahren und Vorrichtung zum Beeinflussen der Querdynamik eines Fahrzeugs
DE102016201205A1 (de) Fahrerassistenzsystem für ein Kraftfahrzeug zur Regelung der Längsdynamik
EP0925484B1 (fr) Procede et dispositif pour le controle de detecteurs dans un vehicule
EP4344978A1 (fr) Dispositif d'estimation
WO2019166325A1 (fr) Procédé pour fournir un paramètre de correction pour corriger des valeurs d'entrée pour l'estimation du coefficient de frottement pour un véhicule et procédé pour corriger des valeurs d'entrée pour l'estimation du coefficient de frottement pour un véhicule et procédé pour commander un véhicule
DE102018218968B4 (de) Verfahren zur Längsdynamikregelung eines Kraftfahrzeugs
EP3789264A1 (fr) Procédé et dispositif par la detection de glissement et véhicule ferroviaire
DE102006040912A1 (de) Verfahren zur Verbesserung eines indirekt messenden Reifendrucküberwachungssystems
DE19542295B4 (de) Bremssystem für ein Fahrzeug
WO2020048714A1 (fr) Structure de régulateur distribuée pour obtenir des propriétés de régulateur optimisées et une durée de vie de soupape accrue
DE102017009031A1 (de) Verfahren zur objektiven Bewertung eines Querregelverhaltens eines aktiven Spurhaltesystems eines Fahrzeugs
DE102023101380B3 (de) Verfahren und Vorrichtung zur computerimplementierten Regelung einer Bremsanlage eines Schienenfahrzeugs
DE102022210316A1 (de) Reduktionsberechnungsverfahren
EP2210787B1 (fr) Système de freinage électropneumatique doté d'une commande sans signal de charge d'essieu
WO2021115837A1 (fr) Procédé et véhicule pour transférer des données de diagnostic du véhicule à une station d'évaluation, et système d'évaluation

Legal Events

Date Code Title Description
PUAI Public reference made under article 153(3) epc to a published international application that has entered the european phase

Free format text: ORIGINAL CODE: 0009012

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: THE APPLICATION HAS BEEN PUBLISHED

AK Designated contracting states

Kind code of ref document: A1

Designated state(s): AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC ME MK MT NL NO PL PT RO RS SE SI SK SM TR

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: REQUEST FOR EXAMINATION WAS MADE