Rolling stock/Supplementary information and regulations/Appendix/Examples of converter unit response to changes in the catenary load

1 Annex 8b

1.1 Examples of converter unit response to changes in the catenary load.

Figur 8b.1
Figur 8b.2: U 60km for active power step in the train of 0 – 2.0 MW
Figur 8b.3:PA 60km for active power step in the train of 0 – 2.0 MW
Figur 8b.4: U 60km for increased power 0 -2.0 MW for linear steepness of 0.5 MW/s
Figur 8b.5
Figur 8b.6

Introduction: The dynamic behaviour of rotating converter units is in the following text illustrated by computer simulations for a 5.8 MVA rotating converter unit which on the single-phase side supplies the train without interconnection with other converter units. The train is simulated with voltage independent power consumtion. Further description of the simulation model, input data etc. is given in annex 8c.

Definitions: Abbreviations used:

U 60km: Pantograph voltage 60km from the converter station when catenary has resistance 0.18 ohm/km and inductive reactance 0.19 ohm/km (16 2/3 Hz)
PA 60km: Angle of pantograph voltage 60km from the converter station when catenary has resistance 0.18 ohm/km and inductive reactance 0.19 ohm/km (16 2/3 Hz). Phase angle is relative to the 50 Hz power supply of the converter station.

Figure 8b.1 shows the converter unit’s frequency response characteristics for active power between the single-phase and the 3-phase side in the 0 – 5 Hz frequency band. I.e. the dynamic, electromechanical connection between a single-phase power disturbance ΔP1 and the response in 3-phase power ΔP3 for the converter unit. It is assumed that the disturbance is sine shaped.

Example of the converter unit’s frequency response characteristic, ΔP3/ ΔP1, for sine shaped change in single-phase power ΔP1.

The frequency response characteristic for this converter unit has a distinctive resonance crest at approximately 1.5 Hz with an amplification of 9 pu.

The voltage oscillations shown in the figures below arise due to changes in the converter rotor velocity when the converter’s pole angle oscillates.

Maximum oscillation (top – bottom) varies with the power step size as follows:

Power step in Train Max. oscillation
U 60km PA 60 km
0 – 3.50 MW  ≈690 V  ≈13°
0 – 2.00 MW  ≈270 V (Shown in figure 8b.2) ≈7° (Shown in figure 8b.3)
0 – 0.25 MW  ≈25 V

Comment:

Sudden step changes in active power cause heavy oscillation in the converter rotor position (pole angle).

Even small power steps cause noticeable oscillation.

Whether the power step is caused by load decrease or load increase has little significance for the oscillation’s magnitude.

Power change of linear steepness – ramp (active power)

Maximum oscillation (top – bottom) varies with the power step (ramp) steepness as follows:

Power change in the train 0 – 2.0MW Max. oscillation U 60km
Step 270 V (Shown in figure 8b.2)
1,5 MW/s linear steepness 14 V
0,5 MW/s linear steepness 5 V (Shown in figure 8b.4)

Comment:

The oscillation magnitude due to change in active power decreases significantly when the power change take place as a ramp of limited steepness. Whether the oscillation is caused by load decrease or load increase has little significance for the oscillation’s magnitude.

Figure 4b.5 shows U 60km for a train that in the beginning consume 4.0 MW and then an unsymmetrical saw-tooth shaped change in load with maximum unfavourable frequency (1.6 Hz) and flank steepness of -1.5 MW/s and + 0.5 MW/s respectively. Repeating power change in each saw-tooth period is 234.375 kW with 4.0 MW as maximum power. Maximum oscillation in catenary voltage (top – bottom) is as follows:

Periodic power change 1.6 Hz: Max. oscillation U 60km
234.375 kW 990 V as shown in figure 8b.5

Comparatively, a repetitive symmetric load change having the same change in each saw-tooth will have flank steepness of +0.75 MW/s and -0.75 MW/s and cause about 5 % larger oscillation in the catenary voltage.

Comment:

Even small power changes will after a few oscillation periods cause considerable oscillation if they are repeated at an unfavourable frequency. If being able to allow rapid repeating power changes in one power direction (e.g. power reduction when activating wheel skid or glide protection) is technically desirable, the converter rotor oscillation can to some extent be limited by implementing stricter restrictions for power changes in the opposite direction.

Reactive power changes

Figure 8b.6 shows U 60km for a converter having 2.0 MW preload and voltage independent load for steps of 0 – 1.0 Mvar inductive power.

Maximum deviation in catenary voltage will be:

Reactive power step: Trancient reduction U 60km
0 - 1,0 Mvar ca. 1000 V (Shown in figure 8b.6)

Comment:

Steps in reactive power in themselves cause only marginal oscillation in the converter unit’s pole angle. However, due to converter unit and catenary inductance, a big initial catenary voltage variation occurs. Indirectly, steps in reactive power may cause some oscillation due to the resulting changes to active losses in the system.


Step reductions of reactive power will also cause brief powerful variations in catenary voltage, but causing the voltage to rise rather than drop.