th
CIRED 20 International Conference on Electricity Distribution
RELIABILITY EVALUATION OF THE CONNECTION OF INDUSTRIAL CUSTOMERS
TO MEDIUM VOLTAGE DISTRIBUTION SYSTEMS
Armin SORG
AREVA Energietechnik GmbH - Germany
armin.sorg@areva-td.com
ABSTRACT
The paper describes measures to enhance the supply
reliability of medium voltage distribution power systems.
The measures are evaluated by quantitative reliability
analyses and compared to the original status without the
measures. The results are described in detail. The necessity
of a qualified interpretation of the results is given.
Examples are given for a real existing medium voltage
system. The measures are concentrated to enhance the
supply reliability of seven selected industrial customers of
special importance.
Further examples are presented for the analysis of the
supply reliability of an industrial customer distribution
system connected to the public distribution system.
INTRODUCTION
Especially for industrial customers the quality of supply is
the main interesting characteristic of electrical energy
besides the price. The supply reliability as one part of the
supply quality is of special importance, due to increasing
requirements resulting from production processes, getting
more and more complex and sensitive against supply
interruptions.
Reliability analyses tools are an adequate measure to handle
a centre part of these quality requirements. They allow the
determination of quantitative results for the quality of
supply of customers or complete power systems. Reliability
analyses tools are of grown importance since some years
and currently they are an important supplement to classical
power system analyses tools like power flow or short circuit
calculation [1-3].
For two power systems different measures are evaluated by
quantitative reliability indices. The first part of the paper
describes examples for the grid connection of industrial
customers to a medium voltage distribution system. The
second part of the paper deals with an industrial customer
distribution system connected to the public distribution
system.
CIRED2009 Session 5 Paper No 0112
Prague, 8-11 June 2009
Paper 0112
Thomas WEBER
AREVA Energietechnik GmbH - Germany
thomas.weber@areva-td.com
MEDIUM VOLTAGE SYSTEM WITH
INDUSTRIAL CUSTOMERS
The investigated system is a real existing system which
supplies a mixed rural/urban area of 220 substations. The
load of the system is about 125 MW. Nearly 60 % of this
load is consumed by only seven big industrial customers.
The focus of the described investigations is to enhance the
supply reliability of these customers.
For medium voltage systems, the frequency of interruption
of customers is directly depending on the line length and
the number of substations of the protection area, the
customer is connected to.
The interruption frequency is not influenced by the location
of the customer within the protection area e.g. if it is
connected by a spur line or to an open operated cable loop.
The interruption frequency directly depends on the location
of the protection devices, which define the protection area.
The protection area includes all devices, that are switched
off simultaneously in case of a short circuit on any device
within the area.
The interruption frequency is depending on the line length
of the protection area, because the frequency of outages of
lines is proportional to the line length.
This relation is shown in Figure 1 based on simulation
results, summarised for the different protection areas.
0,30
Values for different protection areas
Linear approximation
Interruption frequency in 1/a
0,25
0,20
0,15
0,10
0,05
0,00
0 2 4 6 8 10 12
Line length of protection area in km
Figure 1: Frequency of supply interruptions depending on
the line length of the protection area
 th
CIRED 20 International Conference on Electricity Distribution Prague, 8-11 June 2009

Paper 0112

The figure shows a straight rising of the interruption

0,40
frequency with increasing line length of the protection area.
With a change of the switching status of the normal 0,35 Original state
operational state, the protection areas can be changed. Variant

Interruption frequency in 1/a

Based on resulting reduced line length of protection areas, 0,30
reliability indices can move to lower values.
0,25
So the supply reliability can be influenced by change of the
switching state, which includes measures that are easy to 0,20
realize. The following measures are investigated:
0,15
• Change of the location of the sectionalizing point of
open operated cable loops (see Figure 2a), 0,10
• On switching of multiple feeding lines (see Figure 2b)
0,05
Additional measures are possible like rearranging the
busbar connection, but are not described here. 0,00
K1 K2 K3 K4 K5 K6 K7
a) Customer

Figure 3: Frequency of supply interruptions, comparison

of original state and variant
A
Acc. to [1], a break down of the reliability indices acc. to
their affected components is very useful to understand the
origin of the indices and the main influencing equipment.
Figure 4 shows the interruption frequency for industrial
customer K7 acc. to affected components. The main impact
on reliability indices is especially given by three busbars
A (BB1-BB3), one transformer (T1) and several lines (L1-
L5). The figure shows, that for the original state several
lines (L1-L4) have a big impact on the interruption
frequency while they have nearly no impact due to the
b) executed measures for the variant. The reason is, that the
L1 L1 lines L1-L4 are in the same protection area as customer K7
for the original state, while they are shifted to an other
protection area for the variant.

B B
L2 L2 0,12
Figure 2: Principle of simple topological measures (part of Original State
investigated system, connection of industrial 0,10 Variant
Interruption frequency in 1/a

customers A, B), original state and variant

a) Change of the location of the sectionalizing 0,08
point of open operated cable loops
b) Closing of multiple feeding lines
0,06
Figure 3 shows the reliability of the industrial customers in
terms of the calculated interruption frequency for the 0,04
system in original state and after the performance of the
described measures (variant). 0,02
The results show for the example system and the industrial
customers K1 to K7 the potential of the measures in terms
0,00
of reliability indices. The supply reliability can be enhanced BB1 L1 L2 L3 BB2 T1 BB3 L4 L5
significantly for the considered customers by the realization Component
of the simple switching measures. The frequency of supply
interruption can be reduced down to 55 % of the original Legend: BB: Busbar, L. Line, T: Transformer
values (see Figure 3). Figure 4: Differentiation of interruption frequency of
The results for other reliability indices, where the customer K7 acc. to affected components
interruption frequency is a multiplicative component, like
the unavailability (defined in min/a) changes in the same It has to be mentioned, that other customers may achieve a
manner as the interruption frequency itself. supply reliability for the variant, which is worse than the
original state. Furthermore the change of system

CIRED2009 Session 5 Paper No 0112

 th
CIRED 20 International Conference on Electricity Distribution Prague, 8-11 June 2009

Paper 0112

disconnection points can lead to higher system losses. In the

12
considered case, the overall evaluation of the measures led
to the decision of the system operator, to realise them due to
10
the importance of the regarded industrial customers within

Unavailability in min/a
the whole power system. 8

INDUSTRIAL CUSTOMER DISTRIBUTION 6

SYSTEM
4
The investigated customer system is a medium voltage
system, which is fed by four independent cable connections 2
of the public distribution system to two injecting points.
The customer system consists of five substations with an 0

L1a
L1b
L2a
L2b
L3a
L3b

L4a
L4b

L5a
L5b
L3c

L4c
overall load of 10 MW. The loads are connected to the LV
busbars. Parallel operation of MV/LV-transformers is used. Load
The substations are connected in closed operated cable
loops. The cables are protected by differential protection Figure 5: Interruption frequency and unavailability,
devices. variant: remote controlled switch gear
The customer is associated with the chemical industry and
is therefore very sensitive against supply interruptions. Implementing the remote control for the switch gear, the
The supply reliability of the different loads of the system is unavailability of the loads decreases due to the lower
of special importance, due to the fact that supply execution duration of switching actions, in comparison to
interruptions lead to very high costs because of ruined the original state where the switching actions are performed
products as a consequence of the interruption. The costs rise manually. The changes are relative low. This is true,
extremely, if supply interruptions last longer than a duration because the main part of the unavailability is given by
of e.g. 60 minutes. Therefore first the occurrence of supply outages, which cause long lasting supply interruptions like
interruptions should have a low probability. If supply outages due to multiple earth faults or outages of single
interruptions occur though, they should be cleared in a short busbars, where supply interruptions can not be cleared by
time, e.g. by the execution of switching actions. switching actions. Therefore the implementation of remote
To limit the duration of supply interruptions, two different controlled switch gear does not deliver major reliability
approaches are investigated for the given system: benefits. Nevertheless the remote control leads to a
simplification and a speed-up of normal operational tasks.
• Implementation of remote controlled switch gear in the
medium voltage system Busbar sections
• Bus sectionalizing of the medium voltage busbars
The distribution system contains several substations that are
equipped with single busbars without bus sectionalizing.
Remote controlled switch gear Figure 6 shows the principle of the bus sectionalizing. In
The implementation of remote controlled switch gear case of busbar failures, the supply of the LV-system can be
influences the duration of supply interruptions due to partly realised if the bus-coupler is opened.
shorter durations for clearing outages. The measure of
course does not influence the occurrence of supply BB4a BB4b
interruption. Therefore the interruption frequency is equal
for the interruption frequency for the original state and the
variant with remote controlled switch gear (see Figure 5).

0,14

0,12
Interruption frequency in 1/a

L4a L4b L4c

0,10

0,08
BB4a BB4b
0,06

0,04

0,02 Original state

Variant I
0,00
L1a
L1b
L2a
L2b
L3a
L3b

L4a
L4b

L5a
L5b
L3c

L4c

L4a L4b L4c

Load
Figure 6: Single bus bar (original state) and bus
sectionalizing (variant)

CIRED2009 Session 5 Paper No 0112

 th
CIRED 20 International Conference on Electricity Distribution Prague, 8-11 June 2009

Paper 0112

Figure 7 shows the results of the investigations for the short time by the execution of switching actions. Depending
original state and for the sectionalizing of different single on the transformer size and the installed load, the supplied
busbars (Variants IIa, IIb) or all busbars (Variant IIc) of the transformers can take over LV-loads by closing the LV-
system. Busbar 2 (feeding L2a and L2b) is already in the buscoupler.
original state equipped with sectionalized busses.
The results show, that the interruption frequency is a little CONCLUSIONS
higher for the variants for these loads, which are fed by the
sectionalized busbars. This effect is caused by the outage of The described investigations show, that the quantitative
the bus bar sectionalizing itself for the variants (the number reliability calculation is an adequate method to evaluate the
of switchgear panel rises). This change of the interruption supply reliability of customers and loads. Especially the
frequency is low compared to the effect of the bus possibility to compare indices for different variants makes
sectionalizing on the unavailability. this method a worthwhile addition to the conventional
planning methods.
The contribution describes the application of the method for
0,14 two real existing distribution systems. Special focus is laid
Interruption frequency in 1/a

0,12 on the consideration of the supply reliability of industrial

customers.
0,10 The first case considers the connection of industrial
0,08
customers to the public medium voltage distribution
system. The investigations show, that the supply reliability
0,06 can be enhanced significantly for the considered customers
by the realization of simple measures.
0,04 Original state The second part of the investigations deals with measures
Variant IIa
0,02 Variant IIb within a medium voltage distribution system of an
Variant IIc industrial customer. Even if the reliability level is very high
0,00 of the given original system, further big improvements can
L1a
L1b
L2a
L2b
L3a
L3b

L4a
L4b

L5a
L5b
L3c

L4c

be regarded by the execution of further investment

Load measures like the bus sectionalizing.

12 REFERENCES
10
[1] A. Sorg, Th. Weber, 2001, “Application Examples of
Unavailability in min/a

Reliability Analyses According to the Requirements of

8 a Liberalised Electricity Market”, Proceedings of the
1st Balkan Power Conference (BPC), Bled, Slovenia,
6 pp. 141-146
4
[2] FGH, 2000, “ZUBER, Program for Calculation of the
Supply Reliability in Electric Power Systems,” Manual
2 of Version 3.05, Forschungsgemeinschaft für
elektrische Anlagen und Stromwirtschaft e. V.,
0 Mannheim / Germany
L1a
L1b
L2a
L2b
L3a
L3b

L4a
L4b

L5a
L5b
L3c

L4c

[3] J. Backes, O. Bertoldi, H.-J. Haubrich, H.-J. Koglin,

Load A. Montebaur, M. Schwan, A. Sorg, W. H. Wellßow,
Legend: IIa: sectionalizing of BB 3, IIb: sectionalizing of M. Zdrallek, U. Zimmermann, 2000 “Service
BB4, IIc: sectionalizing of BB1,3,4,5 reliability in a competitive market – tools, criteria and
Figure 7: Interruption frequency and unavailability, new approaches for monetary evaluation and risk
variant: Bus sectionalizing of the medium management” Contribution 37/38/39-106 of the 38th
voltage busbars Cigré Session, Paris

The unavailability of the loads, that are equipped with

sectionalized busses decreases by 50 % and more,
compared to the single busbar-connection.
The reliability standard of the considered system is very
high. Therefore the outage of the bus bar has a main effect
on the reliability indices. In case of the outage of a single
busbar, the connected loads suffer supply interruptions,
which last as long, as the outage duration of the busbar lasts
(e.g. duration for repair or replacement).
If the busbar is sectionalized, the supply interruption can be
cleared by sectionalizing the affected busbar section and re-
supply the loads via the busbar section that is not affected
by the outage. The supply interruption can be cleared in a

CIRED2009 Session 5 Paper No 0112