Network integration of wind power - Wind Power Integration: Connection and System Operational Aspects - page 93

often specified for generating units connected to distribution systems is based upon
two issues. The first is a tendency to specify the requirement at the point of con-
nection rather than at the generator terminals; the second is that generating reactive
power is of little use at distribution voltages because reactive power tends to be
trapped in a small part of the system by the reactive impedance of the transformer
connecting the distribution system to the higher voltage networks. Generating reac-
tive power therefore tends to produce very high voltages in small parts of the dis-
tribution system. It is common with traditional generation to augment the variation
available from alternator field current adjustment with a wide tap range on the gen-
erator transformer to 'force' reactive power in one direction or the other. Put another
way, the generator terminal voltage can be kept within limits while allowing the set to
contribute or absorb a large amount of reactive power by adjusting the tap changer on
the transformer.
In the case of wind farms, the generation/absorption of reactive power could be
by switched devices, e.g. reactors and capacitors or power electronic devices, or by
control of converter firing as in the case of DFIGs. In the case of mechanically
switched devices, these will be incapable of dealing with a rapid change in the
network requirement for reactive power resulting from a change in topology. Large
wind farms may also require a transformer with a wide tap range (see Appendix 2,
Clause CC.S2.5.4).
The voltage control range and facilities available from generation connected
at these high voltages is well regulated by grid codes (see Appendix 2, Clause
CC.S2.5.3) and at lower voltages by distribution codes. The practice has been to
apply the appropriate requirement to each conventional generating unit (rather than to
a generating station as a whole). However, in the case of a wind farm, it may be
effective to treat the entire generation node as providing the network voltage reg-
ulating capability. (In Appendix 2 the term Wind Farm Power Station (WFPS), or in
the case of WFPS over 10 MW then a controllable WFPS, is used to refer to the entire
wind farm and it can be seen that most objectives are applied to a WFPS as opposed
to a single wind turbine generator.) This allows innovation in how the functional
requirement will be achieved. Where the farm is very large it may be desirable to be
able to control sections of the wind farm so that the failure of one control element
does not affect the overall capability. These sections would then be treated as wind
farm generating units and the utility requirements applied to them individually. It is
clear that no generating unit (whether renewable sourced or otherwise) should be of a
size greater than is able to be covered by the spinning reserve carried on the system.
(Spinning reserve here means the primary and secondary operating reserve levels
calculated to ensure that load is not automatically shed within the time taken to
synchronise fast-start generators available to the system operator.)
Network grid codes (or transmission access codes as they are sometimes
called) were prepared as part of the arrangements for liberalisation or open access.
They relied largely upon earlier standards. Most have since undergone considerable
modification to accommodate new technologies. In this context, appraisals have
been undertaken with equipment producers to determine whether the power factor
range demanded of different technology generators is appropriate. In many places,