Fundamentals of protection practice

This is a collective term which covers all the equipment used for detecting, locating and initiating the removal of a fault from the power system. Relays are extensively used for major protective functions, But the term also covers direct-acting A.C. trips and fuses.

In addition to relays the term includes all accessories such as current and voltage transformers, shunts, D.C. and A.C. wiring and any other devices relating to the protective relays.

In general, the main switchgear, although funda­mentally protective in its function, is excluded from the term 'protective gear', as are also common services, such as the station battery and any other equipment required to secure operation of the circuit breaker.

In order to fulfil the requirements of discriminative protection with the optimum speed for the many different configurations, operating conditions and construction features of power systems, it has been necessary to develop many types of relay which respond to various functions of the power system quantities.

For example, observation simply of the magnitude of the fault current suffices in some cases but measurement of power or impedance may be necessary in others. Relays frequently measure complex functions of the system quantities, which are only readily expressible by mathematical or graphical means.

In many cases it is not feasible to protect against all hazards with any one relay. Use is then made of a combination of different types of relay which individually protect against different risks. Each individual protective arrangement is known as a 'protection system^'; while the whole coordinated combination of relays is called a 'protection scheme'.
 ·     Reliability
The need for a high degree of reliability is discussed in Section 1. Incorrect operation can be attributed to one of the following classifications:

a.        Incorrect design.
b.       Incorrect installation.
c.        Deterioration.
d.       Protection performance
 

 1. Design
This is of the highest importance. The nature of the power system condition which is being guarded against must be thoroughly understood in order to make an adequate design. Comprehensive testing is just as important, and this testing should cover all aspects of the protection, as well as reproducing operational and environmental conditions as closely as possible. For many protective systems, it is necessary to test the complete assembly of relays, current transformers and other ancillary items, and the tests must simulate fault conditions realistically.

 2.     Installation.
     The need for correct installation of protective equipment is obvious, but the complexity of the interconnections of many systems and their relation-ship to the remainder of the station may make.

Difficult the checking of such correctness. Testing is therefore necessary; since it will be difficult to reproduce all fault conditions correctly, these tests must be directed to proving the installation. This is the function of site testing, which should be limited to such simple and direct tests as will prove the correctness of the connections and freedom from damage of the equipment.

No attempt should be made to 'type test' the equipment or to establish complex aspects of its technical performance;
 
 3. Deterioration in service.
  After a piece of equipment has been installed in perfect condition, deterioration may take place which, in time, could interfere with correct function­ing. For example, contacts may become rough or burnt owing to frequent operation, or tarnished owing to atmospheric contamination; coils and other circuits may be open-circuited, auxiliary components may fail, and mechanical parts may become clogged with dirt or corroded to an extent that may interfere with movement.

One of the particular difficulties of protective relays is that the time between operations may be measured in years, during which period defects may have developed unnoticed until revealed by the failure of the protection to respond to a power system fault. For this reason, relays should be given simple basic tests at suitable intervals in order to check that their ability to operate has not deteriorated.

Testing should be carried out without disturbing permanent connections. This can be achieved by the provision of test blocks or switches.

 Draw-out relays inherently provide this facility; a test plug can be inserted between the relay and case contacts giving access to all relay input circuits for injection. When temporary disconnection of panel wiring is necessary, mistakes in correct restoration of con­nections can be avoided by using identity tags on leads and terminals, clip-on leads for injection supplies, and easily visible double-ended clip-on leads where 'jumper connections' are required.

The quality of testing personnel is an essential feature when assessing reliability and considering means for improvement. Staff must be technically competent and adequately trained, as well as self-disciplined to proceed in a deliberate manner, in which each step taken and quantity measured is checked before final acceptance.

Important circuits which are especially vulnerable can be provided with continuous electrical super-vision; such arrangements are commonly applied to circuit breaker trip circuits and to pilot circuits.

4. Protection performance

The performance of the protection applied to large power systems is frequently assessed numerically. For this purpose each system fault is classed as an incident and those which are cleared by the tripping of the correct circuit breakers and only those are classed as 'correct'.

The percentage of correct clearances can then be determined.

This principle of assessment gives an accurate evaluation of the protection of the system as a whole, but it is severe in its judgment of relay performance, in that many relays are called into operation for each system fault, and all must behave correctly for a correct clearance to be recorded.

On this basis, a performance of 94 % is obtainable by standard techniques.

Complete reliability is unlikely ever to be achieved by further improvements in construction. A very big step, however, can be taken by providing duplication of equipment or 'redundancy^'. Two complete sets of equipment are provided, and arranged so that either by itself can carry out the required function. If the risk of an equipment failing is x/unit, the resultant risk, allowing for redundancy, is x². Where x is small the resultant risk (x²) may be negligible.

It has long been the practice to apply duplicate protective systems to bus-bars, both being required to operate to complete a tripping operation, that is, a 'two-out-of-two' arrangement. In other cases, important circuits have been provided with duplicate main protection schemes, either being able to trip independently, that is, a 'one-out-of-two^' arrange­ment. The former arrangement guards against un­wanted operation, the latter against failure to operate.

These two features can be obtained together by adopting a 'two-out-of-three' arrangement in which three basic systems are used and are interconnected so that the operation of any two will complete the tripping function.

Such schemes have already been used to a limited extent and application of the principle will undoubtedly increase. Probability theory suggests that if a power network were protected throughout on this basis, a protection performance of 99.98 % should be attainable.

This performance figure requires that the separate protection systems be completely independent; any common factors, such as, for instance, common current transformers or tripping batteries, will reduce the overall performance to a certain extent.

·       Selectivity.
Protection is arranged in zones, which should cover the power system completely, leaving no part unprotected. When a fault occurs the protection is required to select and trip only the nearest circuit breakers. This property of selective tripping is also called 'discrimination' and is achieved by two general methods:

1.  Time graded systems.

Protective systems in successive zones are arranged to operate in times which are graded through the sequence of equipments so that upon the occurrence of a fault, although a number of protective equip­ments respond, only those relevant to the faulty zone complete the tripping function. The others make incomplete operations and then reset.

  2.     Unit systems.
It is possible to design protective systems which respond only to fault conditions lying within a clearly defined zone. This 'unit protection' or 'restricted Protection' can be applied throughout a power system and, since it does not involve time grading, can be relatively fast in operation.

Unit protection is usually achieved by means of a comparison of quantities at the boundaries of the zone. Certain protective systems derive their 'restricted^' property from the configuration of the power system and may also be classed as unit protection.

Whichever method is used, it must be kept in mind that selectivity is not merely a matter of relay design.

It is a function of the correct co-ordination of current transformers and relays with a suitable choice of relay settings, taking into account the possible range of such variables as fault currents, maximum load current, system impedances and so on, where appropriate.

 ·     Zones of protection
Ideally, the zones of protection should overlap across the circuit breaker as shown in Figure 2, the circuit breaker being included in both zones.

For practical physical reasons, this ideal is not always achieved, accommodation for current trans-formers being in some cases available only on one side of the circuit breakers, as in Figure 3. This leaves a section between the current transformers and the circuit breaker A within which a fault is not cleared by the operation of the protection that responds. In Figure 3 a fault at F would cause the bus-bar protection to operate and open the circuit breaker but the fault would continue to be fed through the feeder.
 

Figure 3 Location of current transformers on circuit side of the circuit breaker.

The feeder protection, if of the unit type, would not operate, since the fault is outside its zone. This problem is dealt. With by some form of zone exten­sion, to operate when opening the circuit breaker does not fully interrupt the flow of fault current. A time delay is incurred in fault clearance, although by restricting this operation to occasions when the bus-bar protection is operated the time delay can be reduced.

Figure 4 Overlapping zones of protection systems.

 
The point of connection of the protection with the power system usually defines the zone and cor­responds to the location of the current transformers. The protection may be of the unit type, in which case the boundary will be a clearly defined and closed loop. Figure 4 illustrates a typical arrange­ment of overlapping zones.

Alternatively, the zone may be unrestricted; the start will be defined but the extent will depend on measurement of the system quantities and will therefore be subject to variation, owing to changes in system conditions and measurement errors.

·     Stability.

This term, applied to protection as distinct from power networks, refers to the ability of the system to remain inert to all load conditions and faults external to the relevant zone. It is essentially a term which is applicable to unit systems; the term 'discrimination' is the equivalent expression applicable to non-unit systems.

·     Speed.
The function of automatic protection is to isolate faults from the power system in a very much shorter time than could be achieved manually, even with a great deal of personal supervision. The object is to safeguard continuity of supply by removing each disturbance before it leads to widespread loss of synchronism, which would necessitate the shutting down of plant.

Loading the system produces phase displacements between the voltages at different points and therefore increases the probability that synchronism will be lost when the system is disturbed by a fault. The shorter the time a fault is allowed to remain in the system, the greater can be the loading of the system. Figure 1.5 shows typical relations between system loading and fault clearance times for various types of fault.

It will be noted that phase faults have a more marked effect on the stability of the system than does a simple earth fault and therefore require faster clearance.

It is not enough to maintain stability; unnecessary consequential damage must also be avoided. The destructive power of a fault arc carrying a high current is very great; it can burn through copper conductors or weld together core laminations in a transformer or machine in a very short time. Even away from the fault arc itself, heavy fault currents can cause damage to plant if they continue for more than a few seconds

 Figure 5 Typical values of power that can be
 transmitted as a function of fault clearance time.

It will be seen that protective gear must operate as quickly as possible; speed, however, must be weighed against economy.

For this reason, distribu­tion circuits for which the requirements for fast operation are not very severe are usually protected by time-graded systems, but generating plant and EHV systems require protective gear of the highest attainable speed; the only limiting factor will be the necessity for correct operation.
·     Sensitivity
Sensitivity is a term frequently used when referring to the minimum operating current of a complete protective system. A protective system is said to be sensitive if the primary operating current is low.

When the term is applied to an individual relay, it does not refer to a current or voltage setting but to the volt-ampere consumption at the minimum operating current.

A given type of relay element can usually be wound for a wide range of setting currents; the coil will have an impedance which is inversely proportional to the square of the setting current value, so that the volt-ampere product at any setting is constant.          

This is the true measure of the input requirements of the relay, and so also of the sensitivity. Relay power factor has some significance in the matter of transient performance.

For D.C. relays the VA input also represents power consumption, and the burden is therefore frequently quoted in watts.