GO5 - Operation Planning and Control
Grid
Operation
Part 5
Operation
Planning and Control
Power system operation planning
is done on the previous day for implementing it on the next day. All required data like
forecasted load, generation availability, network status, etc are collected
along with schedule changes.
Operation planning, monitoring
and control is a continuous task because operation planning has to be modified
in real-time to accommodate unexpected changes in generation, demand, network, etc.
Power system operation planning
and control is based on three fundamental criteria
1. 1. System security.
System security remains a top priority
while system operation planning, monitoring, and control. Therefore any
occurrences that jeopardize system security are set right immediately by
suitable measures. Practically 100% security cannot be guaranteed but the security
level has to be enough that changes of major occurrence are little.
2. Economic Dispatch.
This is
accomplished in two ways.
· Merit order power injection in the system. Power required for the system demand is drawn on the cheapest source first basis.
· Merit order power dispatch. In case of exigency power supply is restricted to users with the least revenue and that minimum persons are affected with the least inconvenience. However, there may be exemption/relaxation to specific users for the minimum requirement.
3. Power Quality
General power quality refers to factors like Sag, Swell, Break, Flicker, and Harmonics, etc. This may be an individual requirement by specific users and is beyond the scope system operator. But power quality is different in system operation. The following three parameters are monitored and controlled.
( a)
Normal
Frequency.
Frequency is system base parameter as frequency is the same throughout the synchronous grid. Variation in frequency due to changes in demand in any area can be adjusted by pick/drop of generation in any area. Network losses and power angle will increase while power flow over a long distance. Frequency is regulated by active power management.
(b) Normal Voltage Profile.
Voltage is a node base parameter and hence may be different at each node. Voltage profile refers to the voltages of all the nodes in the system. Maintaining of voltage profile is monitoring and managing voltage throughout the system. It requires node-wise reactive balance. Hence the task is somewhat difficult. Also contrary to active power, there are several sources and sinks of reactive power. The flow of reactive power is proportional to the voltage difference of nodes. Q=dv. Transformer taps do not generate or absorb reactive power but are useful to reshuffle reactive power flows. Voltage is regulated by reactive power management.
( c) Reliability of Supply
The uninterrupted power supply is feasible by network security. Therefore due care is required about system security during planning, monitoring, and control of system operation. There should not be total supply failure to any primary distribution center. Maintaining power supply beyond this is taken up by the distribution authority.
Power
system security has a threat from the following.
A.
Abnormal
System Frequency
All the generators have design frequency limits.
Operation beyond these limits may mechanically damage the machine. The upper-frequency limit may be around 51.50 Hz and the lower frequency limit is around
47.50 Hz. Therefore all generators in the system are protected by high
frequency and low frequency relays to isolate them from the system during
abnormal system frequency. There may be slight variation in the tripping frequency
setting of different units based on size and manufacturer. Tripping of any
generator due to low frequency aggravates system condition by a further drop of frequency.
Consequently, other generators may trip leading to cascade tripping of all generators
one after another.
System operation is planned that frequency
remains within normal range. But during the operation frequency may vary due to
deviation in the generation, demand, and network on account of various unexpected causes.
System operators continuously monitor the frequency and take timely corrective actions
whenever required. System operators are
extensively trained for this purpose and also have guidelines for corrective actions.
The guidelines have an action plan according to the severity of abnormalities as
under.
· Severe Deviation.
· Routine Adjustments.
· Critical Condition
In addition to these, there is an automatic backup scheme to take care of when there is no time for the operator’s action. This may be also useful in case of delay in operator action.
B.
Abnormal
Network Loading.
Power flows through various system elements like lines, transformers, bus bars, breakers, isolators, current transformers, etc. Among these lines and
transformers are individually protected by overcurrent relays. These relays isolate
the concern element when the current is more than the setting. The system operator has to be
aware of the overcurrent relay setting and actual current through lines and
transformers. Sometimes tripping of critically loaded line or transformer may cause
separation of weakly linked areas due to tripping of all parallel paths one
after other. This is known as network
cascading. Therefore system operator has to be vigilant and act quickly for corrective
measures for critically loaded elements.
Load current
is proportional to VA the complex of active and reactive power. So it is possible
to reduce load current by control of active or reactive power flow. However, preference is always for reactive power control.
Some
hints for managing overload.
Consider
line configuration as in the figure. Active power is flowing from station S to
station R and is critically loaded.
Reactive power flow is proportional to voltage difference. The flow of Q ยต (Vs
–Vr). Check for reactive power flow in the direction of active power.
· Switch off reactive
power compensation device at exporting station S.
·
Reduce voltage at exporting
station S by tap changer. Importing station R will draw more reactive power
from other sources.
·
Take in service standby
parallel path (1)
·
Switch off the line
importing power at exporting station S (2)
·
Transfer some load of importing
station R by switching offline to a station having an alternate source (3)
·
Drop generation at the
station towards station S and/or pick generation at the station towards
station R.
The above alternatives are not always
effective. The system operator decides the appropriate step according to the situation.Sometimes the situation is so complex
that a quick decision not feasible. But at the same time, it is very essential to
avert tripping of the line by operation of overcurrent relay. One quick but rude solution
is to increase the current trip setting of the relay. This is not a bona fide way to
protect overload but is the only alternative for timely action to avoid tripping. After
doing so the line will not trip on over current protection. But it requires continuous
monitoring of current and should not increase much else there may be various problems
due to overheating. Avoid this during a very hot environment because it will
aggravate the heating problem. This is just an interim arrangement still some appropriate
solution to relieve overload is available and implemented. Efforts should be to control
loading and revert to the normal overcurrent setting as early as possible. There
were incidences of failure due to overheating, heavy sparking, melting and
burning of conductor, contacts and bursting of current transformer, etc.
This is a temporary solution for the overloaded line only. In this condition possibility of melting/burning of
jumper or conductor may be due to very high current for a longer time. But it can
be rectified and restored easily in a short time at least cost. But upgrading of
overcurrent relay for transformer is dangerous and should not be done. Because
fault occurred due to high current in transformer may require repair at works. The
task may be difficult, time-consuming, and costly also. But most incontinence is
to manage without transformer till it gets repaired and received back.
Computer-based contingency
analysis is possible to get the solution in the matter. But it may not be suitable
for real-time application due to its inherent limitations. It may require all
real-time data for analyses. Feeding real-time correct data is a task itself. The program takes time due to the complex analysis by recurring interactions.
During real-time operation, there may be reduction/tripping of generation, tripping of lines and
transformer, etc. Rather than just noting down such the event for record only, the operator should evaluate the impact of each event on system operation and
particularly on system security. Unawareness in the matter may lead to an untoward
event as recorded.
C.
Abnormal
Stability Condition
The network may fail due to overrun of steady-state or dynamic angular
stability limit. But generally modern power systems are having multiple links and
have a rare chance of disturbance on this account. But have to be vigilant in the
matter of operation planning and control. Heavy reactive demand without local
compensation at remote load center may subject to voltage collapse resulting in load throw-off when voltage stability limit is
crossed. Of course, these have more relevance with system development.
The power system is planned and
operated with the above objectives with a strategy depending upon system conditions
as under.
A.
Estimated
availability above peak demand
When generation availability is more than
sufficient, system operation is referred to as surplus management. Generation
chases the demand as per the basic principle of system operation. In this
condition, there is no need for any restriction on any type of power user.
However optimum generation scheduling is required for minimum power cost. It is
prepared on a merit order base. The least-cost source has first priority and higher
the cost will follow in scheduling. The generation schedule is prepared to meet
forecasted demand at normal frequency.
System demand is varying over the day with maximum
load in the evening and minimum load after midnight. There is a 25% to 35% drop in
demand from peak to off-peak depending upon load pattern and season. So
cheapest generation is continued in service during off-peak and generation picked
up / brought back in service on least-cost first bases as per the rise in
demand. Similarly, generation is dropped/withdrawn in reverse order means the costliest first base after peak hours of the day. The result is demand at any time
is met by generation from the least cost sources. In some cases, generation from the highest
cost source may require to stop during the off-peak period to match the drop in
demand. But kinetic and heat energy is wasted during stopping and again energy
is wasted during the restart process. Therefore it is techno-economic consideration
to decided whether to stop the costliest generator or continue in service with technical
minimum generation and adjust balance load drop on the next costly source.
There may be a situation such that all cheapest
generations are away from the load centers. So system losses will be higher
than normal. Here also it is techno-economic consideration to decided as to
continue cheaper generation with extra loss or to switch over to little costly
generation near the load center. Such considerations are for economic dispatch.
Sometimes costlier generation has to be scheduled for maintaining voltage
profile and system stability. Economic consideration is overlooked while
security issues. Generators not scheduled to operate remains on standby as a reserve
for service in exigency. All these are considered while planning system
operation. Computerized application for unit commitment and generation
scheduling is useful for operation planning on the above line.
B.
Estimated
availability equal average demand
(1)
System availability is equal to average demand.
So demand is suppressed by not allowing some load during the period when demand
is more than availability. These loads are allowed during the period when
demand is lower than availability. Ideally, load restricted during high demand
periods is supplied during low demand periods to flatten the demand curve. Shifting
of load from high demand to low demand period is also known as load staggering.
Practically load shifting is done such that the final demand curve has minimum ups/downs.
Preparing a load staggering schedule is a task. The load has to be withdrawn in steps as per rising in demand and has to be released
in steps as per the drop in demand. Therefore load earmark for shedding is
divided into area-wise parts for steps as above. The numbers of parts will be less in
scattered load areas and more parts in the dense load area. This arrangement
facilitates steps for load out/in. Also avoids wide frequency variation when all
loads are out/in at a time. Also avoids voltage rise due to total load shedding
in any area. Also, power rush from one area to another area and overload network
can be avoided by load shedding throughout the system. Load to be staggered
is selected such that minimum persons are affected with the least inconvenience. Most
important is the equitable treatments to all the loads staggered.
Generation scheduling is done as per the revised
demand curve modified due to a staggering schedule. Generation scheduling is
simple as all generators have to operate full day full load with only small
variations. The normal daily load factor of the system may be around 80% to 85%.
But it is expected nearly 100% in this mode of operation. We have observed the power system operating at 97% to 99% daily load factor.
(2)
Load above availability is to be staggered in
steps as in the red zone in the figure. The area where the load is below availability is the green zone when the load is resumed in steps. As availability is equal to average
demand, both the zone area may be almost the same and the final demand curve is nearly flat.
Red zone area is expected to be smaller than
green zone area when availability is above the average demand but less than the peak. Load shed during red zone is supplied any time during green zone as
usual. Even after the release of load as above in the green zone, the demand remains
lower than availability. Consequently, the final demand curve is partially flat but
has a small drop for some period. Generation scheduling is required for this
period as per merit order.
(3)
Red zone area is expected to be larger than
green zone area when availability is below the average demand. So there is no
scope for release of all load shed during the red zone. Only part of the load shed
is possible to release in the green zone. Therefore final demand curve will be almost flat
and there is no need for generation scheduling because all the generation has to
be full all the time.
C.
Estimated
availability bellow minimum demand
Power system management is crucial when
availability is below the average demand and refers to power crisis management.
Contrary to operating standard wherein generation follows the demand, here demand
has to follow the available generation. So there is no need for generation
scheduling because all the generation has to be full all the time. There is the only red zone where demand is curtailed by load shedding as per generation
available. There is no green zone when the load can be released. However, the problem
can be mitigated by imposing the required load shedding of different loads at different times for different durations.
Elasticity of Demand
During operation planning load shedding schedule
is also prepared whenever necessary. Load shedding is disliked by consumers due
to obvious reasons. But it is more inconvenient when unexpected power failure
and don’t know when to resume. Numbers of affected consumers may contact the authority
to know when the power supply is to resume. So load shedding schedule is made
available to users. Consumers adjust their work to use the power before the supply is
put off. Similarly, they are waiting for the power to resume and start using it as
soon as power is available. Therefore system demand increases before and after
the load shedding period. This is like the bloat of the balloon around the point of
pressing. This effect is known as the Elasticity of
Demand. Therefore while preparing load shedding schedule, the start time
is a little earlier and power resume time is a little late than expected as per the demand curve. This is to compensate above effect and void operational problems on
this account.
Recurrence of load.
It appears that energy supplied in the system will reduce
due to load shedding because some load is not delivered for some period. Apparently, energy reduction expected is in load shed M multiply by hours of shedding H. (M×H).
But some of the load recurs as per the elasticity of
demand. It is estimated that about 40% of energy is consumed during recurring load as
above. Hence net energy drop expected is just 0.6 × MH only. Of course this 40%
recurred energy is an average based on observations but actual may differ depending
upon time, duration, and type of the load shed.
Control center monitor and control
system frequency regulating active power on merit order. But there is meager awareness
for voltage profile and reactive power management. Also, operational data and
events are recorded. Recorded data is useful for reference in future planning. But
its importance is to envisage its impact on system security and timely
appropriate action.
Load above availability is to be staggered in steps as in the red zone in the figure. The area where the load is below availability is the green zone when the load is resumed in steps. As availability is equal to average demand, both the zone area may be almost the same and the final demand curve is nearly flat.
