Solution EXERCISE N° 3

IFA-2000 France-UK connection

1-1 R is neglected. Give the relationship between Vc1 and Vc2 (average values of vc1(t) and vc2(t)
respectively)
𝑑𝐼
𝑣𝑐1(𝑡) − 𝑅. 𝐼(𝑡) − 𝐿 ∙ + 𝑣𝑐2(𝑡) = 0
𝑑𝑡
𝑑𝐼
R is neglected, and the average value of 𝐿 ∙ is zero. Thus:
𝑑𝑡
Vc1 = -Vc2
Then deduce a relationship between 1 and 2.

3.U . 2 3.U . 2
Since Vc1  cos 1 and Vc2  cos  2 with same phase to phase voltage on the French
 
and UK sides, this leads to
𝛼1 = 𝜋 − 𝛼2
1-2 Operating with 1 = 30°on the French side and with I = 1 850 A,
Calculate 2

𝛼2 = 150°
Compute the exchanged power. What is the direction of the power flow?

I is supposed constant, therefore, P = Vc1 . I = - Vc2 . I (both are in generator convention, so one must be
negative since power is transferred from converter 1 to converter 2 or vice versa)
P = 255MW
Show that the connexion is reversible despite the unidirectional current flow.

The connexion is reversible since Vc1 and Vc2 can change sign. P1 is positive (and P2 is negative) when 1
<90° (and 2 > 90°). P1 becomes negative (and P2 positive) when 1 > 90° (and 2 < 90°).

1-3 The frequencies f1 and f2 are equal to 50 Hz. Is it necessary to have the same frequencies in the
same way as it is the case for 2 alternators in parallel on the same network ? (synchronous
connection).
No since there is a DC link between the two rectifiers so the two rectifiers can see different frequencies on the
AC side
2 – R is no longer neglected and taken equal to R = 3.1 ,, but L is big enough to neglect I current ripple.
2-1 What is the relationship between Vc1, Vc2 and I?
 𝑑𝐼
𝑣𝑐1(𝑡) − 𝑅. 𝐼(𝑡) − 𝐿 ∙ + 𝑣𝑐2(𝑡) = 0
𝑑𝑡
So
𝑉𝑐1 − 𝑅. 𝐼 + 𝑉𝑐2 = 0

What is the new relationship between 1 and 2. ? It is unnecessary to write it as : 1 = …

𝜋
cos(𝛼1 ) + cos(𝛼2 ) = 𝑅. 𝐼.
3. 𝑈. √2
2-2 Operating with 2 = 150°and I = 1 850 A,
Compute 1.

cos(1) = -0.902, 1 = 25,6° (a bit less than 30° to generate a higher voltage on side 1 to anticipate the resistive
voltage drop)

Compute the powers on the French and on the English side.

P1 = Vc1.I = 265.9 MW
P2 = Vc2.I = -255.3 MW
Is the difference predictable and computable?

The difference is R.I², losses in the DC line

Verification: |P1|-|P2| = 10.6 MW, and R.I² = 10.6 MW
3.

3-1 Explain why : V’c1 = V’’c1.

3.U . 2 3.U . 2
Vc1 '  cos 1 , Vc1 ''  cos 1 , with same U and same 1. The phase shift between the
 
2 primary voltages doesn't change anything since both delay angles 1 are defined with respect to thyristor
voltage.
3-2 Assuming 1 = 30°, plot v’c1(t), and compute v’c1(max) and v’c1(min) respectively its
v'
maximum and minimum value. Compute as well V' c1 , the relative difference between these 2
values.
From this figure, Vc1'(max) is the peak of phase to phase voltage
Vc1'(max) = 166,9 kV
For the min value, it occurs 60° after the peak (since the frequency of the full bridge rectifier is 6 times the
grid frequency, so 360°/6 = 60° period)
𝑉𝑐1 ′ (max) = 𝑈. √2. sin⁡(90°) = 166.9 kV
𝑉𝑐1′ (𝑚𝑖𝑛) = 𝑈. √2. sin⁡(90° + 60°) = 83,4 kV
3-3 On the same graph, plot v’’c1(t), and deduce vc1(t). [kV]

The value of vc1(t) is obtained by summing these two voltages. It is not displayed in this figure, it will be
displayed in the following one. The frequency is twice the frequency of vc1' and vc1'' thus 12 times the grid
frequency.

v
What is the period of vc1(t) ? What is its relative ripple: ?
Vc1
The frequency is twice the frequency of vc1' and vc1'' thus 12 times the grid frequency.

v v'
What is the relationship between and (question 3-2) ?
Vc1 V' c1
v' 166.9  83.4
  0.6 (V'c1 = 138kV with 1 = 30°)
V'c1 138
To compute the ratio with two bridge delayed by 30°, the most simple is to measure it on the graph. This
ripple can also be computed mathematically (but this is not really useful):
We obtain Vc1(max) as follows:
𝑉𝑐1 ′ (max) = 𝑈. √2. sin⁡(90°) = 166.9 kV. At this time, 𝑉𝑐1 ′′ = 𝑈. √2. sin⁡(90° − 30°) = 144.5 kV (since
voltage of Vc2'' is lagging 30°). Thus Vc1(max) = 311.4 kV
We obtain Vc1(min) as follows:
𝑉𝑐1 ′ (min) = 𝑈. √2. sin⁡(90° + 60°) = 83.4 kV. At this time, 𝑉𝑐1 ′′ = 𝑈. √2. sin⁡(90° + 60° − 30°) = 144.5 kV
(since voltage of Vc2'' is lagging 30°). Thus Vc1(min) = 227.9 kV
Therefore
v 311.4  227.9
  0.3 (Vc1 = 276kV with 1 = 30°) – This is half voltage ripple (what can be seen on the
Vc1 276
figure below comparing 2.v'c1 and vc1

4–
Compute the active power flow P as well as the reactive one Q.

One rectifier is delivering with 1 = 30° an active power of

P1 = Vc1 . I (with the average voltage Vc1 = Vc1' + Vc1'' and I = 1850 A)
6.U . 2
Vc1  cos 1 since it is twice Vc1'

P1 = 511 MW
Since 4 converters are used, we have:
Ptot = 2042 MW (so IFA 2000, 2000 being MW)
For reactive power, we consider that the first harmonic of the current is lagging by 1 the voltage, thus
each individual cell (vc' and vc'') is ABSORBING Q = 3.V.I1stHarm_RMS.sin()
𝐼𝑐.√3.√2
We remind that 𝐼 =
1stHarm_RMS 𝜋
118𝑘𝑉 1850.√3.√2
Thus 𝑄𝑡𝑜𝑡 = 8 ∙ 3 ∙ √3
∙
𝜋
∙ sin(30°) = 1179⁡𝑀𝑉𝐴𝑟
cos() = Pto/Stot = 0.866, to be compared with power factor (which considers the harmonics)
3
Fp = 0,827 ( f   cos( ) ). This is not exactly the same due to harmonics effects.

Are these powers supplied or absorbed by the French network?

With 1 = 30°, the French side is delivering Active power to UK. Reactive Power is ABSORBED

What happens to P and Q on the English side?

With 1 = 30°, the UK side is receiving Active power from France. Reactive Power is ABSORBED to UK
side