Professional Documents
Culture Documents
E. Gulski1* , P. Cichecki1, J.J. Smit1, P.P. Seitz 2, B. Quak 2, F. Petzold 3 and Frank de Vries 4
1
Delft University of Technology, Mekelweg 4, 2628CD Delft, The Netherlands
2
Seitz Instruments AG, Mellingerstrasse 12, CH-5443 Niederrohrdorf, Switzerland
3
SebaKMT, Dr.-Herbert-Iannstr. 6, 96148 Baunach, Germany
4
Nuon Tecno. Voltastraat 2, 1817 DD Alkmaar, The Netherlands
*Email: e.gulski@tudelft.nl
number of failures
Therefore in this contribution based on field application
Installed length
(a) (b)
of advanced diagnostics a systematic approach for
condition assessment of HV power cables is discussed.
2 HV INSULATION DIAGNOSTICS
Pipe type oil -filled/ aged 10 -50 years Pipe type oil -filled/ aged 10 -50 years
Number of leakages
The insulation failures in a cable network may be
Installed length
(c) (d)
caused by lower dielectric strength due to aging
processes and by internal defects in the insulation
system. It is known that unlike voltage testing,
measurements of the dielectric may give an absolute
Fig. 2: Failure statistics of two types of oil-filled power indicator for the quality level of the cable insulation. For
cables. It follows from these examples that the service age HV power cables different on-site
of the whole cable system: inspections/diagnostics are available [1,3,12,13]. In
(a-b) is not sufficient to be used as indicator for replacement particular with regard to information as provided,
decisions, conclusions about short- and long -term condition can be
(c-d) is not direct related to the failure statistics of a sub- made (figure 3).
system. The results of these measurements may have a direct
relation to the average qualitative level of the insulation
weighting factor for at the moment of measurement and can thus be applied
long term assessment as a trend- or fingerprint measurement during future
a b inspections. It follows from figure 3 that partial
1
discharge diagnosis and dielectric diagnosis may play
important role in short- and long -term condition
c, f assessment, figure 4 .
0,5
HV Power Cable
Integral insulation degradation
e,g,d h Thermal aging
weighting factor for Insulation defects degradation
short term assessment
0,5 1 PD diagnosis Dielectric losses diagnosis
Fig. 3: Contribution of different on-site diagnostics to PD inception voltage PDIV Tan @ 0 U0 ..2.0 U 0
tan @ 0.5 U
short and/or long term condition assessment [3, 13] PD extinction voltage PDEV 0 to 2U 0
0
Dielectric diagnosis
Dielectric losses
two designated voltages so called tan or tip-up is
important for condition assessment. The loss tangent is Partial discharges: PDIV, PD level, PD
measured as function of voltage to check the quality of location
1000
50kV 900
Partial Discharges
800 0,3xUo
700
0,7xUo
Ontladingsniveau (pC)
600
1,0xUo
500
1,3xUo
400
300 1,7xUo
200 2,0xUo
100
L1 Tangens delta
L 2 versus testspanning L3
Fase
100
100kV 10
L1
Tg d (*E-4)
L2
L3
1
0.9 1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2.0 2.1
Dielectric losses
0.1
T e stsp a nn i ng (. . x U o)
4 PRACTICAL EXAMPLES
Performing on-site testing/diagnosis of installed HV
power cable circuit may have different purposes. In
(b) particular the following goals could be of interest:
Fig. 7: Examples of DAC voltages as applied to a 220kV a) verify that a new circuit installation or repaired
power cable showing PD activity in a termination: (a) PD circuit does not contain workmanship problems
patterns at different voltage levels, (b) Mapping of PD which can lead to localised PD,
location in the cable up to 250kV. b) to asses the actual condition of a cable system as a
losses, see figure 6. Due to the fact that by using this part of asset management program e.g.
method: - to support maintenance and replacement
- the frequency of damped AC voltages is in the range of management or
power frequency of acceptable HV test systems, - as a means to support the operational decisions e.g.
- a number of power cycles is applied to the cable about the load profile as acceptable for a particular
sample provide ignition of PD sources in similar ways cable connection.
as compared to operating conditions, In the following figures several examples of measuring
the PD activity can be on-site measured with multiple results as obtained from such diagnostic tests are shown.
undisturbed sinusoidal voltage cycles, see figure 7. The In figure 9 PD measurements as made during after-
PD activity signals, ignited during one or more laying test of 6.4km long, 50kV XLPE insulated cable
oscillating voltage waves, are detected by the system, section are shown. Based on the PD detection and
which can process the signals for several purposes. localisation it could be concluded that up to 2.0xU 0
- A phase-resolved PD pattern can be resolved from phases L1 and L2 are PD-free and that at 1.4xU 0 in the
multiple DAC sequences. In this way, patterns can be phase L3 PD activity of 180 pC (corona) has been
obtained which are similar to those recognized under observed in the cable termination.
50(60) Hz conditions [2], In figure 10 PD diagnosis of 6km long, 50kV a paper-
- Single PD pulses can be analyzed for original location mass-insulated cable section. Due to the fact that this
by using travelling wave analysis. Statistical type of insulation is not PD-free the goal of this test was
evaluation of PD signals obtained after several
It follows from these values that with regard to PD
(b) activity in function of the test voltage up to 1.7xU0
phase L3 shows increased PD activity in the cable
(a)
termination. Having experience norms for this type of
cable terminatio n could be used to support the decision
about maintenance/replacement of this particular
component.
(c)
L1
1 L2
L3
0,1
(e) 40 50 60 70 75 80
Voltage [kV peak]
(a)
R NC : A M K 6 6 K V to A D M I R A L T Y R D W E S T 2 3 0 k V
L1
0,1 L2
Fig. 9: Example of PD measurements on new 6.4km long L3