ECTE324/8324 Power Engineering 1 Week 10: Prof. Sarath Perera Phone: 4221 3405 Room: 35-G33

Download as pdf or txt
Download as pdf or txt
You are on page 1of 35

ECTE324/8324 Power Engineering 1

Week 10

Prof. Sarath Perera


Phone: 4221 3405 Room: 35-G33

ECTE324/8324 Power Engineering 1 Week 10 – 1 / 35


Review of Week 9

• Power Electronics

• The Diode Rectifier

• The Chopper

• The Inverter

ECTE324/8324 Power Engineering 1 Week 10 – 2 / 35


Topics for Today

• Power Quality and Reliability

◦ Reliability: availability of supply

◦ Power Quality: purity of the supply

ECTE324/8324 Power Engineering 1 Week 10 – 3 / 35


Reliability of Supply

• Short Interruptions

◦ A short interruption occurs when the supply voltage decreases to less


than 10% of the nominal value for a period not exceeding one minute.

◦ Can be the result of system faults, equipment failure and control


system malfunctions.

◦ The duration of the interruption can be a function of the protection


settings.

◦ Delayed reclosing, for example, can produce short interruptions.

• May be preceded by a voltage sag; indicates that the source of the


interruption is a fault.

ECTE324/8324 Power Engineering 1 Week 10 – 4 / 35


Reliability of Supply

• Long Interruptions

◦ When the voltage goes to zero for more than one minute.

◦ Caused by a fault in the distribution system which is not automatically


cleared.

◦ Failure to clear is due to the nature of the fault, or equipment failure.

• The permanent fault is one which cannot be cleared by normal recloser


operation.

◦ Fallen tree(s)

◦ Faulty transformers, etc

ECTE324/8324 Power Engineering 1 Week 10 – 5 / 35


Redundancy

• Redundancy is defined as:

“Provision of alternative supply in the case of some


permanent faults”

• For example, feeder segmentation:

Customer Customer
CB1 CB1 CB2 CB3
(a) (b)

• This is a key part of Distribution Automation.

ECTE324/8324 Power Engineering 1 Week 10 – 6 / 35


Improving Reliability - Redundancy

Customer Customer
CB1 CB1 CB2 CB3
(a) (b)

• In (a) there is no redundancy; if the CB1 opens, then the customer is left
without power.

• In (b) there is some redundancy designed into the feeder circuit.

◦ The normal operation is CB1 and 2 closed, CB3 open (reduces fault
current).

◦ Idea is that CB1, 2 open while CB3 closes if fault on the left-hand (in
the illustration) line.

ECTE324/8324 Power Engineering 1 Week 10 – 7 / 35


Improving Reliability - Redundancy

Fault Fault
Customer Customer
CB1 CB1 CB2 CB3
(a) (b)

• In (a), the fault is permanent and so the customer will be without power.

• In (b), permanent fault and failure of CB3 will cause long interruption.

• So . . . , experience tells us:

“No redundant system gives perfect reliability because any part has a
non-zero probability of failure.”

• How can the reliability be estimated?

ECTE324/8324 Power Engineering 1 Week 10 – 8 / 35


Estimating Reliability of Equipment

• Reliability is defined as:

The fraction of time that an item is available


and able to do its intended task

• From this notion, there are several other measures that help us define
reliability an other useful metrics:

◦ Outage rate

◦ Repair time

◦ Unavailability rate

◦ Availability rate

ECTE324/8324 Power Engineering 1 Week 10 – 9 / 35


Single Item Reliability

• Outage Rate

◦ The outage rate, O , is the probability of failure in a period of 1 year.

◦ For example, a utility has 100 transformers. Over a 5 year period,


there are 7 failures. What is the outage rate?

◦ O = probability of failure/year = 7/100/5 = 0.014

• Utilities keep statistics of all their equipment so that they can derive these
probabilities.

• Statistics (of all sorts; not just failure rates) are kept on all similar items of
equipment for their lifetime and beyond.

• Statistics provide a mechanism for probabilistic modelling of system


reliability.
ECTE324/8324 Power Engineering 1 Week 10 – 10 / 35
Single Item Reliability

• Repair Time

◦ The repair time, R, is the average time from item failure to its return to
service. (Includes time for the removal of the device from system.)

• Unavailability Rate

◦ The unavailability rate U , is the fraction of time the item is not able to
perform intended service.
U = OR

◦ i.e. Unavailability Rate = Outage Rate × Repair Time

ECTE324/8324 Power Engineering 1 Week 10 – 11 / 35


Unavailability Rate Example

A utility has 1500 transformers and over a 7 year period, there are 153 failures. Each
transformer requires 2 weeks (on average) to repair. What is the unavailability rate?

ECTE324/8324 Power Engineering 1 Week 10 – 12 / 35


Single Item Reliability

• Availability Rate (complement of unavailability)

◦ Availability Rate, A, is the fraction of time equipment is able to


perform intended service

A = (1 − U ) = (1 − OR)

◦ This is the same as Reliability.

◦ i.e. The fraction of time that an item is available and able to do its
intended task.

ECTE324/8324 Power Engineering 1 Week 10 – 13 / 35


Reliability Example

A power system is 99.9% reliable (‘three nines’). How many hours of interruption are to be
expected each year?

ECTE324/8324 Power Engineering 1 Week 10 – 14 / 35


Another Reliability Example

Some IT industries hope for no more than 5 minutes lost time per year. What reliability do
they require?

ECTE324/8324 Power Engineering 1 Week 10 – 15 / 35


Combinations of Equipment

• Systems connected in series:

◦ Service is only available if both items are available.


Systems in series: service is only available if both items are available
U1 A1
U2

system
A2 available

O1,R1 O2,R2

• Total Availability = product of separate availabilities.

• In general, for n devices in series, Rel = Rel1 × · · · × Reln

• Reliability is always less than that for the most unreliable piece of
equipment.

ECTE324/8324 Power Engineering 1 Week 10 – 16 / 35


Series Reliability Example

A transformer, circuit breaker and line have reliabilities of 99.9 %, 99.99% and 99.8%. How
many minutes per year is each device unavailable? What is the overall reliability? For how
many minutes per year is the system unavailable?

ECTE324/8324 Power Engineering 1 Week 10 – 17 / 35


Systems in Parallel

• The service isunavailable


unavailable only if both items are unavailable.
U1 A1
U2

O1,R1

A2 system not
available
O2,R2

• Total unavailability = product of separate unavailabilities, or

(1 − Rel) = (1 − Rel1 )(1 − Rel2 )

giving
Rel = Rel1 + Rel2 − Rel1 Rel2

• Difficult to generalise to n devices in parallel using Reliabilities. Better to


use Unavailability.
ECTE324/8324 Power Engineering 1 Week 10 – 18 / 35
Parallel Reliability Example

Device one is unavailable for 5 minutes a year, device 2 for 15 minutes a year. For what time
will the parallel system be unavailable?

ECTE324/8324 Power Engineering 1 Week 10 – 19 / 35


Another Parallel Reliability Example

Device 1 is unavailable for 1 hour a year. What should be the Unavailable rate for Device 2
to give a system which us unavailable for only 5 minutes a year?

ECTE324/8324 Power Engineering 1 Week 10 – 20 / 35


IT Industry Reliability Requirements

• Microprocessor-based controls and computer networks at present require


‘6 nines’ reliability or no more than 32 seconds of lost time annually.
Future requirements are estimated to be ‘9 nines’ (less than 32 ms
unavailability/year) which is uneconomic from the supply system.

• Normal power system reliability is 99.9-99.999% (9 hours to 5 minutes).


This is uneconomic to be met from the normal power system and has to
be met by other means.

ECTE324/8324 Power Engineering 1 Week 10 – 21 / 35


Increase Reliability of the Supply System

• Utility devised solution.

• This is expensive and only feasible in local parts of the power system.
Only economic if many sensitive customers come together to share the
cost and benefits. Custom power (or premium power) is the setting up of
a very reliable power system for just a few customers closely situated in a
‘Premium Power Park’. Some components are:

◦ Multiple transmission lines to substation.

◦ Solid state switches which can quickly disconnect lines with disturbed
voltage waveforms and connect healthy lines.

◦ Dynamic voltage restorers to hold up any missing part of the voltage.


waveform

ECTE324/8324 Power Engineering 1 Week 10 – 22 / 35


Increase Reliability within one Plant

• Customer devised solution.

• This is cost-effective in the case of some processes (IT, manufacturing


with high loss of production costs). This requires some form of energy
storage to be applied to the whole plant or to the critical parts.

• Power Conditioning

ECTE324/8324 Power Engineering 1 Week 10 – 23 / 35


Power Quality

• What is Power Quality?

◦ Power Quality is the study of the sources, effects and control of


disturbances which propagate via the electric power supply.

• Cause: When lightning strikes a power line

• Disturbance: a HV transient is injected and propagates to nearby loads

• Effect: damage to electronic equipment, loss of data in digital equipment

ECTE324/8324 Power Engineering 1 Week 10 – 24 / 35


Why should we care about Power Quality?

• Modern equipment is electronic/digital and more susceptible (IT, variable


speed drives, etc).

• Modern equipment generates more disturbances in the supply!

◦ SMPS generate harmonics

◦ air-conditioners cause annoying light flicker

• Customers becoming more ‘rights’ conscious and may litigate if there is


loss of equipment, production or data.

• State regulators are requiring distributors to be more concerned.

ECTE324/8324 Power Engineering 1 Week 10 – 25 / 35


Is there a way of classifying PQ disturbances?

The ‘normal’ waveform has three parameters:

• Frequency: 50 Hz

• RMS value: 230 V rms (line-neutral) in LV systems

• Waveform: sinusoidal

Broadly, PQ disturbances can be classified as:

• Frequency disturbances – very unusual and not considered further here

• Voltage disturbances – rms value affected

• Waveform disturbances – nonsinusoidal

ECTE324/8324 Power Engineering 1 Week 10 – 26 / 35


The Dirty Dozen Disturbance Types

Sadly, there’s really only 10. . .

• Voltage outside of normal range (too high or low)

• Unbalance (unequal voltages across the phases)

• Sag (short term reduction in V to 10-90%)

• Momentary interruption (short term reduction into 0-10%)

• Swell (short term increase in V above 110%)

• Fluctuations (continual small changes in V within the normal range)

ECTE324/8324 Power Engineering 1 Week 10 – 27 / 35


Harmonic disturbance types.

3.1
7 3.2
8 3.3
9

• Harmonics

• Oscillatory transients

• Impulsive transients

• Notching (not that common)


10

ECTE324/8324 Power Engineering 1 Week 10 – 28 / 35


How are these disturbance types caused?

Disturbance Type Cause


1 Voltage outside normal range Uncontrolled load growth; tap-changer maloperation 2,4,5
2 Unbalance Unequal loads on 3 phases 4
3 Sag Fault, DOL motor start 1,3,4,5
4 Momentary interruption Recloser operation 1,2,3,5
5 Swell Sudden load removal, intermix 1,2,4
6 Flicker Pulsating load (air-conditioner, arc furnace) 1,4,5
7 Harmonics Power electronic loads 1,4,5
8 Oscillatory transients Capacitor switching on 1,2,4,5
9 Impulsive transients Lightning strikes 1,3,5
10 Notching Low power supply inductance controlled rectifiers 10

ECTE324/8324 Power Engineering 1 Week 10 – 29 / 35


Location and propagation of PQ disturbances

transmission distribution

(4)
(3) lightning distorting
load

(1)

(2)
(5)
wiring
capacitor defects sensitive
(3) accidents
switching load

1 Other interconnected suppliers generating 2 From normal operations within the


disturbances supply system
3 Accidents (faults) in the supply system (e.g. 4 Nearby customers
lightning, trees, animal, car impacts)
5 Within the affected premises

ECTE324/8324 Power Engineering 1 Week 10 – 30 / 35


How is equipment affected by PQ events?

Disturbance Type Effects


1 Voltage outside normal range Motor overheating; light globe loss of life; TV picture distortion
2 Unbalance Motor overheating
3 Sag Maloperation/shut-down; clock reset; loss of hold-up time
(SMPS, VSD); contactors drop out
4 Momentary interruption Maloperation/shut-down; clock reset; temporary loss of HID lig-
hts; loss of hold-up time (SMPS, VSD); contactors drop out
5 Swell Maloperation, Capacitor Failure
6 Flicker Annoyance (especially at close to 8 Hz)
7 Harmonics Overheat motors, capacitors; clock errors, tariff meter errors;
high neutral currents
8 Oscillatory transients VSD overvoltage trip
9 Impulsive transients Damage to electronics

ECTE324/8324 Power Engineering 1 Week 10 – 31 / 35


How is it possible to mitigate PQ issues?

• It is not possible to globally remove all PQ issues.

• Control can be oriented to reducing disturbance at the:

◦ source

◦ propagation path, or

◦ victim (affected equipment)

• Source (Network)

◦ A good automatic power system protection scheme .

◦ Good design, construction and maintenance (tap-changers, loading,


balancing loads, vegetation trimming).

ECTE324/8324 Power Engineering 1 Week 10 – 32 / 35


More PQ mitigation strategies.

• Propagation path: use power conditioning devices.

• Mitigation at the load itself:

◦ Soft-starters or VSDs

◦ Standards for the connection of difficult loads such as


air-conditioners, power electronic loads etc.

• Victim: Equipment should be well designed and take account of


standards for emission and immunity (e.g. CBEMA or ITI curve for
computers)

ECTE324/8324 Power Engineering 1 Week 10 – 33 / 35


The ITI (CBEMA) Curve

• Computer and Business Equipment Manufacturers Association (CBEMA)


and Information Technology Industry (ITI) Council.

• A curve designed to describe the tolerance of equipment to the


magnitude and duration of voltage variations on the power system.

• Originally devised for mainframe computers but now encompasses a


broader range of devices; generally microprocessor based.

ECTE324/8324 Power Engineering 1 Week 10 – 34 / 35


The ITI (CBEMA) Curve

ECTE324/8324 Power Engineering 1 Week 10 – 35 / 35

You might also like