Docstoc

Engineering

Document Sample
Engineering Powered By Docstoc
					                  Research Programme
                        Engineering
         Aerodynamics in the Open Air
Cross wind issue - Pilot studies: Results
                                                               Pilot Studies
                                                                                Aerodynamics in Open Air




                             Aerodynamics in Open Air (AOA)

    Public Closing Meeting, Paris, Oct. 21-22, 2008

                                                  WP2 “Cross Wind Issue”

                                                       Pilot Studies. Results

  M. Testa
  RFI - Direzione Tecnica
  Direzione Norme Standard
  Sviluppo e Omologazione
Public Closing Meeting – Paris, 22 October 2008. Mario Testa              1
                                                                                                     AOA
                                                               Pilot Studies
                                                                                  Aerodynamics in Open Air

    Main expectations from AOA

               Assess the availability and and practicability of a robust and easy to handle method for cross
             wind risk analysis

                 Focus on new lines and existing lines

                 HSL operated with Class 1 trains, CRL operated with Class 2 trains


    Pilot studies

               Risk analysis performed independent by two or more AOA partners using the different
             approaches under examination

               Assess reliability and robustness of the methods applied, through comparisons of the results
             obtained

                  Spotting the open points remaining that need further research



Public Closing Meeting – Paris, 22 October 2008. Mario Testa              2
                                                                                                         AOA
                                                               Pilot Studies
                                                                                    Aerodynamics in Open Air

    Pilot studies were to prove the Specifications

                 Specifications on how to represent infrastructure characteristics (WP2.2)

                 Specifications on the derivation of the CWC in respect of the infrastructure scenarios (WP2.2)

                 Specifications on the methodologies to perform the meteo studies (WP2.3)

                 Specifications on the methods to estimate the probability of CWC exceedance (WP2.4)


    The Specifications refer to the 4 Methods under assessment

                  AGM: AOA German Method

                  AFM: AOA French Method

                 ITM: Italian Method

                 GBM: Great Britain Method


Public Closing Meeting – Paris, 22 October 2008. Mario Testa              3
                                                                                                          AOA
                                                               Pilot Studies
                                                                               Aerodynamics in Open Air

    High Speed Lines and Conventional Rail Lines considered in the Pilot studies

                 RFI- HS Rome Naples; two samples

                 RFI- HS Rome Naples; complete extension of the line


                 SNCF- HS LN5 Valence - Marseille; one sample


                 SNCF- HS LN6 Paris - Strasbourg; one sample


                 DB- HS Hannover-Würzburg; one section


                 Network Rail- CR GB West Coast Main Line; one section


                 BV- CR Scandinvian West Coast Line; one section


Public Closing Meeting – Paris, 22 October 2008. Mario Testa              4
                                                                                                    AOA
                                                                  Pilot Studies
                                                                                                                       Aerodynamics in Open Air
   Detailed
   work plan                  line owner                       AGM                        AFM                   ITM                                 GBM
                              line name
                                                               DBRFI infra data                                 RFI / POLIMIRFI infra data
                              RFI
                              Rome-Naples                      DBSNCF infra data          SNCFSNCF infra data   RFI / POLIMISNCF infra data         RSSBSNCF infra data
                              samples, 10 km                   RSSBSNCF infra data                              RFI / POLIMIRFI infra data
                                                                                                                by using the AOA transposition of
                                                               SNCF SNCF infra data                             CWCs
                              RFI
                              Rome-Naples                      DBRFI infra data
                              Complete
                              SNCF
                                                               SNCFSNCF infra data
                              LN5                                                         SNCFSNCF infra data
                                                               DBSNCF infra data
                              sample, 17 km
                              SNCF
                                                               SNCFSNCF infra data
                              LN6                                                         SNCFSNCF infra data
                                                               DBSNCF infra data
                              sample, 10 km
                              DB
                                                               DBDB infra data
                              Hannover - Würzburg
                                                               RSSBDB infra data
                              section
                              Network Rail
                                                                                                                                                    RSSBRSSB infra data
                              GB West Coast Main Line          DBRSSBI infra data

                              BV                               Bombardier
                              Scandinavian West Coast Line     DBBT_KTH infra data



Public Closing Meeting – Paris, 22 October 2008. Mario Testa                          5
                                                                                                                                                                  AOA
                                                               Pilot Studies
                                                                                           Aerodynamics in Open Air

          Results from the pilot studies

                     in spite of the fact that the initial Specifications gave a potentially unified set of procedures,
                     some differences became apparent

                     the differences are ranked in two main classes

                     a)        differences arising from the particular way each procedure was executed in practice:
                               these differences are not correlated to the technical background of the calculations,
                               but to the different understandings in doing detailed subsets of calculations

                     b)        differences that are inherent in the ultimate technical background relating to each
                               procedure

                     therefore improvements were made on the specifications and the corrected ones have been
                     introduced in the latest draft of EN 14067-6

                     but other differences could not always be reconciliated and therefore open points were left




Public Closing Meeting – Paris, 22 October 2008. Mario Testa              6
                                                                                                                  AOA
                                                               Pilot Studies
                                                                                          Aerodynamics in Open Air

    … results from the pilot studies


            a)       differences arising from the particular way each procedure was executed in practice

                     -         how the basic infrastructure databases, drawings, maps, etc., are read, interpreted
                               and transformed
                     -         how the uncompensated acceleration is taken into account in the transition curves
                     -         how finely the infrastructure heights above the terrain are categorized
                     -         how the infrastructure and the meteo reference points along the line are associated
                               with each other




Public Closing Meeting – Paris, 22 October 2008. Mario Testa              7
                                                                                                                AOA
                                                               Pilot Studies
                                                                                           Aerodynamics in Open Air

   … results from the pilot studies


            b)       differences that are inherent in the ultimate technical background relating to each procedure

                     -         the way the viaducts are considered: it depends on whether or not the over-speed
                               factor due to the blockage effect of the viaduct deck is considered
                     -         the way the embankments are considered: either through the modification of the
                               aerodynamic coefficients or through the over-speed the embankment causes to the
                               onset air flow
                     -         the way the flat ground scenario is considered: whether or not the flat ground is
                               considered with a ballast and rail configuration
                     -         how the complex meteo are executed: it involves the level of complexity of the routine
                               solving the NS equations and the type of output to feed the CWC exceedance
                               calculations


Public Closing Meeting – Paris, 22 October 2008. Mario Testa              8
                                                                                                                   AOA
                                                               Pilot Studies
                                                                                        Aerodynamics in Open Air

    Convergence of the AGM

      The AGM has been submitted to the widest set of applications by many AOA partners, in the
    domain of Class 1 and Class 2 lines

                 Rome-Naples sample1                                   DB, RSSB, SNCF
                 Rome-Naples sample2                                   DB, RSSB, SNCF
                 LN5 Valence-Marseille sample                          DB, SNCF
                 LN6 Paris-Strasbourg sample                           DB, SNCF
                 Hannover-Würzburg                                     DB, RSSB
                 Swedish West Coast Line                               DB, BT

        All the results obtained show a satisfactory convergence and reproducibility of this method

       The most important discrepancy between RSSB and DB results can be traced to differences in the
    treatment of transition curves

       The largest deviation between the results of SNCF and DB are mostly related to differences in the
    line database



Public Closing Meeting – Paris, 22 October 2008. Mario Testa              9
                                                                                                             AOA
                                                                                                      Pilot Studies
                                                                                                                                                                                                 Aerodynamics in Open Air

                                 Convergence of the AGM

                                                                                        AGM Method applied by SNCF
                                                                                        AGM Method applied by DB
                                      Probability of CWC Exceedance                     AGM Method applied by RSSB
 Probability of CWC Exceedance




                                                                                                                                                                                                                        AGM Method applied by SNCF
                                                                                                                                                             Probability of CWC Exceedance                              AGM Method applied by DB




                                                                                                                      Probability of CWC Exceedance
                                   72,8   73,3   73,8   74,3    74,8       75,3      75,8     76,3    76,8     77,3

                                                               Track Position [km]




                                                                                                                                                           6,5   7,5   8,5   9,5 10,5 11,5 12,5 13,5 14,5 15,5 16,5 17,5 18,5 19,5 20,5 21,5 22,5 23,5
                                                                                                                                                                                                 Track Position [km]




Public Closing Meeting – Paris, 22 October 2008. Mario Testa                                                                                          10
                                                                                                                                                                                                                                            AOA
                                                                                                        Pilot Studies
                                                                                                                                                                                             Aerodynamics in Open Air

                Convergence of the AGM

                                                                                             AGM Method applied by SNCF
                                      Probability of CWC Exceedance                          AGM Method applied by DB
   Probability of CWC Exceedance




                                                                                                                                                                                                                   AGM applied by RSSB
                                                                                                                                                                 Probability of CWC exceedance
                                                                                                                                                                                                                   AGM applied by DB




                                                                                                                            Probability of CWC Exceedance
                                   167,9   168,9   169,9   170,9   171,9   172,9   173,9    174,9   175,9   176,9   177,9
                                                                      Track Position [km]




                                                                                                                                                                 149    159    169     179       189       199     209    219      229
                                                                                                                                                                                             Track Position [km]




Public Closing Meeting – Paris, 22 October 2008. Mario Testa                                                                                                11
                                                                                                                                                                                                                                       AOA
                                                               Pilot Studies
                                                                               Aerodynamics in Open Air

    Convergence between AGM and AFM

         The AGM and AFM are different in the meteo approaches used.

       The AGM is based on a wind map approach while in the AFM a complex numerical simulation of
    the atmospheric wind field is undertaken

      The wind distributions along the line are not directly comparable and thus the comparisons have
    been made only through the final results of the probability exceedance along the line

         No detailed comparison of intermediate meteo results has been undertaken

      Differences are existing on how the infrastructure scenarios are dealt with (overspeed effect on
    embankments and viaducts, flat ground situation)




Public Closing Meeting – Paris, 22 October 2008. Mario Testa             12
                                                                                                    AOA
                                                                                                 Pilot Studies
                                                                                                                                                                                    Aerodynamics in Open Air

   Convergence between AGM and AFM

      The general trend of the CWC exceedance probabilities is reproduced fairly well by both methods.
   The largest deviations between the methods apply to the low embankment and viaduct scenarios
   because of the differences in taking into account the over-speeding effect


                                                                                                 AFM Method
                                                                                                                                                                                                                     AFM Method
                                     Probability of CWC Exceedance                               AGM Method
                                                                                                                                                        Probability of CWC Exceedance                                AGM Method
                                                                                                                                                            scaled by each Mean
     Probability of CWC Exceedance




                                                                                                                     Prob. of CWC Exceed. / Mean
                                                                   Viaducts
                                                                                                                                                                                     Viaducts




                                     72,8   73,3   73,8   74,3   74,8      75,3        75,8   76,3   76,8     77,3                                      72,8   73,3   73,8   74,3    74,8     75,3     75,8   76,3      76,8      77,3
                                                                 Track Position [km]                                                                                                 Track Position [km]




Public Closing Meeting – Paris, 22 October 2008. Mario Testa                                                                                       13
                                                                                                                                                                                                                               AOA
                                                                                                         Pilot Studies
                                                                                                                                                                                                        Aerodynamics in Open Air

                                Convergence between AGM and AFM

                                    For some lines, the differences are more significant. This could be explained by the local meteo
                                features that are better captured by the numerical simulations than by the wind map



                                                                                                              AFM Method                                                                                                                    AFM Method
                                         Probability of CWC Exceedance                                        AGM Method                                                                                                                    AGM Method
                                                                                                                                                                     Probability of CWC Exceedance
                                                                                                                                                                         scaled by each Mean

                                                                                                                                                                                                                                    2.5m embankments
Probability of CWC Exceedance




                                                                                                                                     Prob. of CWC Exceed. / Mean
                                                                                                                                                                                                          Viaducts


                                   6,5    7,5   8,5   9,5   10,5 11,5 12,5 13,5 14,5 15,5 16,5 17,5 18,5 19,5 20,5 21,5 22,5 23,5                                  6,5   7,5   8,5   9,5   10,5 11,5 12,5 13,5 14,5 15,5 16,5 17,5 18,5 19,5 20,5 21,5 22,5 23,5
                                                                              Track Position [km]                                                                                                           Track Position [km]




Public Closing Meeting – Paris, 22 October 2008. Mario Testa                                                                        14
                                                                                                                                                                                                                                                   AOA
                                                                                             Pilot Studies
                                                                                                                                                                            Aerodynamics in Open Air

    Convergence between ITM and AFM

       The meteo approaches used in the ITM and AFM are similar, but have different levels of complexity
    in solving the NS equations. One more source of discrepancies consisted in the way the exceedance
    probability was calculated. In the ITM the intersection of the probabilistic distribution of the wind with
    the probabilistic distribution of the wheel unloading (stochastic CWC) is considered. In the AFM,
    instantaneous wind speeds are generated and compared with the deterministic CWC values (Monte
    Carlo approach)                                                 ITM applied by POLIMI/RFI

                                                                                                                                         AFM applied by SNCF
                                       Probability of CWC exceedance




                                                                                                                             V-14/10 m
                                                                                               V-9 m

                                                                                                                                                          C-2 m
                                                                                                                                         E1-12.5 m
                                                                                                  E1-7.5 m   E1-7.5 m



                                                                                                                                                 V-6/9m
                                                                                                       E1-2.5 m

                                                                                  E1-7.5 m


                                                                                                                                                     E1-7.5m   E1-7.5m



                                                                       72.00   73.00          74.00           75.00             76.00                 77.00              78.00

                                                                                                       Track position [km]




Public Closing Meeting – Paris, 22 October 2008. Mario Testa                                                            15
                                                                                                                                                                                                 AOA
                                                                                                   Pilot Studies
                                                                                                                                                                                Aerodynamics in Open Air

    Convergence between ITM and AFM

        Intermediate steps of comparison of the outcomes of the meteo approaches were performed. The
    differences were identified in the transfer coefficients of the wind speed from the meteo reference
    stations to the line, rather than in the statistical treatment of the meteo data at the reference meteo
    stations


                                                      Sample1: k mean wind speed                         max_AFM
                                                                                                                                                                      Sample1: k peak wind speed                        max_AFM
                                                   envelopes of minima and maxima                        max_ITM                                                   envelopes of minima and maxima                       max_ITM
                                          over all direction sectors at reference meteo stations         min_AFM                                          over all direction sectors at reference meteo stations        min_AFM
                             1,20                                                                                                             2,00
                                                                                                         min_ITM                                                                                                        min_ITM
                                                                                                                                              1,80
                             1,00
                                                                                                                                              1,60
     tranfer coefficient k




                                                                                                                   tran sfer co efficient k
                             0,80                                                                                                             1,40
                                                                                                                                              1,20
                             0,60                                                                                                             1,00
                                                                                                                                              0,80
                             0,40
                                                                                                                                              0,60

                             0,20                                                                                                             0,40
                                                                                                                                              0,20
                             0,00                                                                                                             0,00
                                    72   73         74             75               76     77       78                                               72   73         74           75            76       77        78
                                                           track pos ition [k m ]
                                                                                                                                                                          track position [km]




Public Closing Meeting – Paris, 22 October 2008. Mario Testa                                                                                   16
                                                                                                                                                                                                                        AOA
                                                                                               Pilot Studies
                                                                                                                                                                   Aerodynamics in Open Air

    Convergence between ITM and AFM

       For the CWC derivation, the ITM uses a stochastic approach and not the deterministic one used in
    the AFM and AGM. The different approach to the over speed effect in the embankment scenario has
    played an important role

                                                         1.00E-04
                                                                                                           Determinstic CWC’s
                                                                                                           Stochastic CWC
                                                         1.00E-05
                         Probability of CWC exceedance




                                                         1.00E-06


                                                         1.00E-07


                                                         1.00E-08
                                                                    E1 12.5
                                                                                                                             E1 12.5
                                                                                              E1 7.5                                            E1 2.5    E1 7.5
                                                                                         V9                  V 9-14.5                  V 6-12
                                                         1.00E-09
                                                                        E1 7.5
                                                                                                                                           E1 7.5   C-4
                                                         1.00E-10


                                                         1.00E-11
                                                                72.82            73.82          74.82           75.82                  76.82
                                                                                               Track position [km]

Public Closing Meeting – Paris, 22 October 2008. Mario Testa                                                            17
                                                                                                                                                                                        AOA
                                                                                                      Pilot Studies
                                                                                                                                                                                        Aerodynamics in Open Air
    Convergence between GBM and AGM

         There are a number of factors which could contribute to the substantial discrepancy between CWC
        based results (AGM) and results based on the stability and aerodynamic load calculations used for
        the GBM:
               the use of low turbulence vs atmospheric boundary layer wind tunnel tests
               the wind “speed up” model for embankments
               the MBS vehicle dynamics modelling used for determining the CWC
               differences between the Class 390 vehicle and the reference Class 1 vehicle

                                                                                                                                                         GBM
                                                                                                                                                         GBM +CWCs
                                                                                                                                                         AGM + cat. heights
                                             Probability per curve transit

                                                                             1

                                                                                 3

                                                                                     5

                                                                                         7

                                                                                             9

                                                                                                 11

                                                                                                      13

                                                                                                           15

                                                                                                                17

                                                                                                                     19

                                                                                                                          21

                                                                                                                               23

                                                                                                                                    25

                                                                                                                                          27

                                                                                                                                               29

                                                                                                                                                    31

                                                                                                                                                         33

                                                                                                                                                              35

                                                                                                                                                                   37

                                                                                                                                                                        39

                                                                                                                                                                              41

                                                                                                                                                                                   43
                                                                                                                      Curve number




Public Closing Meeting – Paris, 22 October 2008. Mario Testa                                                                             18
                                                                                                                                                                                                             AOA
                                                                                                        Pilot Studies
                                                                                                                                                                     Aerodynamics in Open Air
    Convergence between GBM and AGM

        The GBM + CFD overturning calculations are accepted in Great Britain as reliable. The simplified
        GBM is somewhat less accurate, as it does not take account of large scale sheltering and funnelling
        effects, which have the potential to cause errors of an order of magnitude in predicted probability

         Methods using the TSI CWCs, including the AGM, should be regarded as conservative relative to
        the GBM. This is because the use of the CWCs rather than the GB stability criterion results in
        probability predictions of the order of a thousand times greater than those determined using the
        GBM + CFD                                                                    1.E-04
                                                                                                                                                        GBM + CWCs
                                                                                     1.E-05                                                             GBM

                                                                                     1.E-06

                                                                                     1.E-07
                                                     probability per curve transit




                                                                                     1.E-08

                                                                                     1.E-09

                                                                                     1.E-10

                                                                                     1.E-11

                                                                                     1.E-12

                                                                                     1.E-13

                                                                                     1.E-14   v iaducts circle d

                                                                                     1.E-15
                                                                                            7

                                                                                            6

                                                                                            4

                                                                                            4

                                                                                            3

                                                                                            3

                                                                                            8
                                                                                           22

                                                                                           42

                                                                                           61

                                                                                           17

                                                                                           55

                                                                                           04

                                                                                           70

                                                                                           00

                                                                                           67

                                                                                           02

                                                                                           47

                                                                                           90

                                                                                           18

                                                                                           48

                                                                                           82
                                                                                          .7

                                                                                          .9

                                                                                          .5

                                                                                          .9

                                                                                          .2

                                                                                          .7

                                                                                          .2
                                                                                         6.

                                                                                         6.

                                                                                         6.

                                                                                         7.

                                                                                         7.

                                                                                         8.

                                                                                         8.

                                                                                         9.

                                                                                         9.

                                                                                         0.

                                                                                         0.

                                                                                         0.

                                                                                         1.

                                                                                         1.

                                                                                         1.
                                                                                        72

                                                                                        73

                                                                                        74

                                                                                        75

                                                                                        76

                                                                                        76

                                                                                        77
                                                                                       13

                                                                                       13

                                                                                       13

                                                                                       13

                                                                                       13

                                                                                       13

                                                                                       13

                                                                                       13

                                                                                       13

                                                                                       14

                                                                                       14

                                                                                       14

                                                                                       14

                                                                                       14

                                                                                       14
                                                                                                                   curve sequence in "down" direction




Public Closing Meeting – Paris, 22 October 2008. Mario Testa                                                                        19
                                                                                                                                                                                          AOA
                                                               Pilot Studies
                                                                                 Aerodynamics in Open Air

          Open points left

                     Over-speed effect on viaducts
                     -    In one type of approach (AFM, ITM, GBM) the over-speeding effect exists, but does
                          not lead to corresponding increases in the aerodynamic force/moment coefficients
                          acting on the vehicle.
                     -    In the other approach (AGM) the over-speeding effect also exists, but affects the
                          aerodynamic force and moment coefficients acting on the vehicle. This approach
                          delivers more conservative evaluations i.e. higher calculated risks on viaducts.

                     Over-speeding effect on embankments
                     -    The AGM, ITM and basic GBM way of considering embankments consists of taking
                          into account the embankment by the means of the accelerating effect it produces on
                          the normal component of the unperturbed upstream airflow.
                     -    In the AFM and modified GBM the accelerating effect provided by presence of the
                          embankment on the wind speed upstream is accounted for in the CWCs and not in the
                          wind distribution on the line.




Public Closing Meeting – Paris, 22 October 2008. Mario Testa             20
                                                                                                      AOA
                                                               Pilot Studies
                                                                                    Aerodynamics in Open Air

          …. open points left

                     Flat ground situation
                     -     The AGM and ITM take into account the flat ground CWCs
                     -     The GBM refers to a GB ballast and rail scenario
                     -     The AFM considers the flat ground situation as being equivalent to that of a very low
                           embankment. The AFM approach to the flat ground situation was recognized to be too
                           conservative and some mitigating considerations were developed.

                     Handling the complex meteo studies
                     -    The ways in which the complex meteo methods are dealt with and the probabilities of
                          exceedance are handled are intrinsically cross-linked. In fact, in the complex meteo
                          methods, the meteo outputs have forms that are somehow “subservient” to the
                          probabilistic post-processing of the data.
                     -    The convergence of those methods needs deeper investigations on the
                          correspondence between the different probabilistic distribution types




Public Closing Meeting – Paris, 22 October 2008. Mario Testa             21
                                                                                                          AOA
                                                               Pilot Studies
                                                                               Aerodynamics in Open Air




Public Closing Meeting – Paris, 22 October 2008. Mario Testa             22
                                                                                                    AOA
                                                               Pilot Studies
                                                                                Aerodynamics in Open Air




                             Aerodynamics in Open Air (AOA)

    Public Closing Meeting, Paris, Oct. 21-22, 2008

                                                  WP2 “Cross Wind Issue”

                                                       Pilot Studies. Results

  M. Testa
  RFI - Direzione Tecnica
  Direzione Norme Standard
  Sviluppo e Omologazione
Public Closing Meeting – Paris, 22 October 2008. Mario Testa             23
                                                                                                     AOA
RSSB Research Programme
Block 2 Angel Square
1 Torrens Street                                                  research@rssb.co.uk
London
EC1V 1NY                  www.rssb.co.uk/research/rail_industry_research_programme.asp

				
DOCUMENT INFO