EEP-309

EXPERIMENT-6

FAULT CLEARANCE TO MAINTAIN GENERATOR STABILITY

SUBMITTED BY:
DATE: 11-04-2013 Aditya Vashist (2010EE1043) Karanpreet Singh (2010EE1056)

EXPERIMENT-6 AIM:
To study the critical clearing time for symmetric faults at the terminals of a fully configured hydro generator with specific focus on the effect of: a. Level of active power generation b. Impedance of connecting line c. Turbine-generator inertia

SOFTWARE USED: PSCAD

Theory:
Different stages encountered when a fault occurs at a generator:

Subtransient Period: A few cycles from fault initiation when the armature windings and dampers all affect the AC current variation. Current values are very high but decay fast. All electrical protection systems must operate in this time frame. Transient Period: Up to several seconds from fault initiation when armature windings and field windings affect the AC current variation. Current values are lower than in subtransient period but also decay more slowly. All stability and control systems for the generator must operate in this time frame to ensure stability. Steady State: The sustained states after both subtransient and transient currents have decayed to negligible values.

Different types of faults encountered in transmission lines are: Ground fault: Or SLG fault connects a single phase to ground the other phases remaining as they are generating identical current magnitudes in all three sequences. Line Fault: Or LL fault-connects two of the phases remaining as it is Generating identical current magnitudes in positive and negative sequences only. Double ground fault: Or LLG fault connects two of the three phases to ground simultaneously the third phase remaining as it is. Symmetric fault: Connects all the three phases Generating current only in the positive sequence. simultaneously

The stability of a system refers to the ability of a system to return back to its steady state when subjected to a disturbance. As mentioned before, power is generated by synchronous generators that operate in synchronism with the rest of the system. A generator is synchronized with a bus when both of them have same frequency, voltage and phase sequence. We can thus define the power system stability as the ability of the power system to return to steady state without losing synchronism. Usually power system stability is categorized into Steady State, Transient and Dynamic Stability.

Steady State Stability studies are restricted to small and gradual changes in the system operating conditions. In this we basically concentrate on restricting the bus voltages close to their nominal values. We also ensure that phase angles between two buses are not too large and check for the overloading of the power equipment and transmission lines. These checks are usually done using power flow studies. Transient Stability involves the study of the power system following a major disturbance. Following a large disturbance the synchronous alternator the machine power (load) angle changes due to sudden acceleration of the rotor shaft. The objective of the transient stability study is to ascertain whether the load angle returns to a steady value following the clearance of the disturbance. The ability of a power system to maintain stability under continuous small disturbances is investigated under the name of Dynamic Stability (also known as small-signal stability). These small disturbances occur due random fluctuations in loads and generation levels. In an interconnected power system, these random variations can lead catastrophic failure as this may force the rotor angle to increase steadily. Fault Clearing time: The time required by a system of circuit breakers to disconnect a faulted network element and reconnect the same after the fault has been removed. Critical fault clearing time: With reference to a specific fault at a specific location of a network the maximum value of fault clearing time required of the brakers such that the network should not lose transient stability.

SETUP OF THE UTILITY NETWORK:

The setup of the utility network is as shown in the figure below:

The entire setup is rated at 13.8 kV,120 MVA so that its 1.0 p.u. impedance has a base value of 13.82/120=1.587.It consists of a hydro-generator and an infinite bus connected by an inductive line . On the other side of the inductive line is the three phase AC source as an infinite bus which should be set at 13.8kV.

MATHEMATICAL CLEARING TIME:

ANALYSIS

FOR

OBTAINING

THE

Let us consider a three-phase synchronous alternator that is driven by a prime mover. The equation of motion of the machine rotor is given by

d 2 Tm Te Ta dt 2

(1)

where J Tm Te is the total moment of inertia of the rotor mass in kgm2 is the mechanical torque supplied by the prime mover in N-m is the electrical torque output of the alternator in N-m is the angular position of the rotor in rad

Neglecting the losses, the difference between the mechanical and electrical torque gives the net accelerating torque Ta. In the steady state, the electrical torque is equal to the mechanical torque, and hence the accelerating power will be zero. During this period the rotor will move at synchronous speed s in rad/s. The angular position is measured with a stationary reference frame. To represent it with respect to the synchronously rotating frame, we define
st

(2)

where is the angular position in rad with respect to the synchronously rotating reference frame. Taking the time derivative of the above equation we get
d d s dt dt

(9.83

) Defining the angular speed of the rotor as

d dt

we can write (3) as

r s
d dt

(4)

We can therefore conclude that the rotor angular speed is equal to the synchronous speed only when ddt is equal to zero. We can therefore term ddt as the error in speed. Taking derivative of (3), we can then rewrite (1) as
J d 2 Tm Te Ta dt 2

(5)

Multiplying both side of (5) by m we get

J r d 2 Pm Pe Pa dt 2

(6)

where Pm, Pe and Pa respectively are the mechanical, electrical and accelerating power in MW. We now define a normalized inertia constant as
H J s2 Stored kinetic energyat synchronou s speed in mega - joules Generator MVA rating 2S rated

(7)

Substituting (7) in (5) we get

2H Srated

s2

d 2 Pm Pe Pa dt 2

(8)

In steady state, the machine angular speed is equal to the synchronous speed and hence we can replace r in the above equation by s. Note that in (8) Pm, Pe and Pa are given in MW. Therefore dividing them by the generator MVA rating Srated we can get these quantities in per unit. Hence dividing both sides of (8) by Srated we get

2 H d 2 Pm Pe Pa per unit s dt 2

(9)

Equation (9) describes the behaviour of the rotor dynamics and hence is known as the swing equation. The angle is the angle of the internal emf of the generator and it dictates the amount of power that can be transferred. This angle is therefore called the load angle. Consider the single-machine-infinite-bus(SMIB) system shown in Fig. 9.1. In this the reactance X includes the reactance of the transmission line and the synchronous reactance or the transient reactance of the generator. The sending end voltage is then the internal emf of the generator. Let the

(10) sending and receiving end voltages be given by

Fig. 9.1 An SMIB system. We then have (11)

The sending end real power and reactive power are then given by

(12) This is simplified to

Since the line is loss less, the real power dispatched from the sending end is equal to the real power received at the receiving end. We can therefore write

(13)

where Pmax = V1 V2 / X is the maximum power that can be transmitted over the transmission line. The power-angle curve is shown in Fig. 9.2. From this figure we can see that for a given power P0 . There are two possible values of the angle - 0 and max . The angles are given by

Equal Area Criterion for Stability

The real power transmitted over a lossless line is given by Pe = Pm sin Now consider the situation in which the synchronous machine is operating in steady state delivering a power Pe equal to Pm when there is a fault occurs in the system.

Opening up of the circuit breakers in the faulted section subsequently clears the fault. The circuit breakers take about 5/6 cycles to open and the subsequent post-fault transient last for another few cycles. The input power, on the other hand, is supplied by a prime mover that is usually driven by a steam turbine. The time constant of the turbine mass system is of the order of few seconds, while the electrical system time constant is in milliseconds.

Power-angle curve for equal area criterion.

Let us consider the power angle curve shown in Fig. Suppose the system of Fig. 2 is operating in the steady state delivering a power of Pm at an angle of 0 when due to malfunction of the line, circuit breakers open reducing the real power transferred to zero. Since Pm remains constant, the accelerating power Pa becomes equal to Pm. The difference in the power gives rise to the rate of change of stored kinetic energy in the rotor masses. Thus the rotor will accelerate under the constant influence of non-zero accelerating power and hence the load angle will increase. Now suppose the circuit breaker re-closes at an angle c. The power will then revert back to the normal operating curve. At that point, the electrical

power will be more than the mechanical power and the accelerating power will be negative. This will cause the machine decelerate. However, due to the inertia of the rotor masses, the load angle will still keep on increasing. The increase in this angle may eventually stop and the rotor may start decelerating, otherwise the system will lose synchronism. We know that :

Hence multiplying both sides by

and rearranging we get

Multiplying both sides of the above equation by dt and then integrating between two arbitrary angles 0 and c we get

Now suppose the generator is at rest at 0. We then have d / dt = 0. Once a fault occurs, the machine starts accelerating. Once the fault is cleared, the machine keeps on accelerating before it reaches its peak at c , at which point we again have d / dt = 0. Thus the area of accelerating is given from above equation as

Now consider the case when the line is reclosed at c such that the area of acceleration is larger than the area of deceleration, i.e., A1 > A2 . The generator load angle will then cross the point m , beyond which the electrical power will be less than the mechanical power forcing the accelerating power to be positive. The generator will therefore start accelerating before is slows down completely and will eventually become unstable. If, on the other hand, A1 < A2 , i.e., the decelerating area is larger than the accelerating area, the machine will decelerate completely before accelerating again. The rotor inertia will force the subsequent acceleration and deceleration areas to be smaller than the first ones and the machine will eventually attain the steady state. If the two areas are equal, i.e., A1 = A2 , then the accelerating area is equal to decelerating area and this is defines the boundary of the stability limit. The clearing angle c for this mode is called the Critical Clearing Angle and is denoted by cr. We then get from Fig. 1 by substituting c = cr

We can calculate the critical clearing angle from the ab move equation. Since the critical clearing angle depends on the equality of the areas, this is called the equal area criterion. We are interested in finding out the maximum time that the circuit breakers may take for opening, we should be more concerned about the critical clearing time rather than clearing angle. Furthermore, notice that the clearing angle is independent of the generalized inertia constant H . Hence we can comment that the critical clearing angle in this case is true for any generator that has a d-axis transient reactance of 0.20 per unit. The critical clearing time, however, is dependent on H and will vary as this parameter varies. To obtain a description for the critical clearing time, let us consider the period during which the fault occurs. We then have Pe = 0. Therefore

Integrating the above equation with the initial acceleration being zero we get

Further integration will lead to

Replacing by cr and t by tcr in the above equation, we get the critical clearing time as

OBSERVATIONS AND CALCULATIONS:

In this experiment, we have to find the critical clearing time for symmetric fault and how it depends CCT on various factors. If the critical clearing time a a network is very less the breakers may not get the required time to disconnect once the fault occurs and reconnect after the fault has been removed. Therefore it will be in the state of transient stability. 1. First the effect of initial output power has on the critical clearing time is to be measured. First of all to get proper readings for the critical clearing time we increase the duration of run in small steps executing the simulation every time until you are able to identify the critical clearing time when the hydro generator will just go unstable. Then the Initial output power is modified between 0.1p.u. to 1.2p.u. in steps of 0.1 pu. And the critical clearing time is tabulated to symmetric fault in each case. As it is clear from the readings and the plot the critical clearing time increases as the initial output power is decreased. Initially at 0.8p.u. the critical clearing time is 0.3265 sec and it increases as the power decrease and rises to 0.9231 sec for 0.4 p.u. while for 1.2 p.u. it is 0.2396 sec. Moreover it is observed that if the initial output power is below 0.3 p.u. the generator never goes into the state of transient instability even if the duration of simulation is increased. The results are shown in the plot and table below:
Initial output Critical power Clearing time 0.1 NA 0.2 NA 0.3 NA 0.4 0.9231 0.5 0.6112 0.6 0.4986 0.7 0.3899 0.8 0.3265 0.9 0.2801 1 0.2411 1.1 0.2396 1.2 0.2396

Variation of Critical Clearing Time

1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 1.2

Explanation:

1. If the initial output power is high and the fault occurs at the electrical side the electrical output power drops to very low value. Given time, the governor installed on the turbine side would bring the turbine output power to the new value. However the governor being a slow control system the change takes time. 2. In the interim period the difference between the mechanical and electrical powers lead to large differential torque on the inertia of the turbine-generator that disturbs the combined inertia of the turbine generator and set its speed hunting around the steady state synchronous value.Since the consequent variation of the power angle is nothing but a time integration of the speed change this varies with time too. 3. If the mechanical electrical torque differential initiated by the fault is excessive the consequent of power angle may throw the generator out of synchronism. So if the initial output power is high the drop in power when a fault occurs on the

electrical side would be very large and hence the governor will take more time to bring the input power equal to the output power which will result in high mechanical electrical torque differential that throws the generator into the state of transient instability. So the critical clearing time available to the breakers will be less. 4. If we decrease the initial output power the mechanical electrical torque differential also begins to decrease and hence the governor will find it easy to bring the generator back into synchronism. So this will give some time to the breakers to disconnect and reconnect and hence will result in the increase of critical clearing time. 5. When the initial output power becomes very low the torque differential is also very less so the governor does not take much time to change the mechanical power and hence synchronism is obtained instantly. Hence the generator never reaches the state of transient instability the critical time for these powers is very large hence it is not observed. The critical clearing time for a network is given by:

As the initial power ouput is changed H and s remain constant while Pm and cr-0 change. As the initial output power is decreased Pm decreases which results in increase of critical clearing time which is clear from the plots as well. Now lets see the variation of tcr with cr-0. In the initial case when the output power is Pm=0.8 p.u. H=3.117MI/MVA s=377rad/sec and Xs=0.1p.u. V1V2sin(0)/Xs=Pe This gives the value of 0=4.60

Line Inductance

Critical Clearing

As we know that V1V2sin(0)/Xs=Pe

Therefore we can write Pe=Ksin(0)

Integrating the following equation yields the following final result sin(0)(-2 0)-cos(0)=cos(cr) As 0=4.6o cr=1390 Which yields cr-0.=2.35 rad Putting the above in the formula for tcr yields tcr=0.311sec . Which is very close to the experimentally obtained value of 0.3265 sec. As the initial output power is decreased it is clear from the above equation that value of cr-0 will tend to increase which will result in the increase of critical clearing time.This is what we observe in the graphs obtained. Hence our experimentally obtained values are correct mathematically as well.

Inference:
The critical clearing time is inversely proportional to the initial output power. So high initial power gives less time to the brakers to disconnect and reconnect once a fault occurs.

2. Initial power output is kept constant at 0.8 p.u. with line reactance varying from 0.1 p.u. to 0.6 p.u. with the difference of 0.05 p.u. for each case tabulate the critical clearing time for symmetric faults as shown below :-

0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 0.55 0.6

time 0.3125 0.2719 0.2521 0.2241 0.2127 0.2348 0.2253 0.143 0.1407 0.2405 0.0671

Variation of Critical Clearing Time With Line Inductance

0.35 0.3 0.25 0.2 0.15 0.1 0.05 0 Critical Clearing time

The variation in this case is not uniform but it can be seen from the plot that as the inductance increases the critical clearing time decreases. Explanation for the variation: The critical clearing time as shown in case 1

With change in line reactance there is no effect on H,s or Pm. The variation is

due to change in cr-0. The output power is given by the formula Pe =V1V2sin(0)/Xs Now in this case Pe is fixed at 0.8p.u. As we change Xs 0 changes which results in change of cr-0 hence such a variation is observed. Suppose when Xs=0.3p.u. Sin(0)=0.24

Using this value yields Thus Giving tcr=0.2229sec which Experimentally obtained value = 0.2127 sec.

0=13.90 cr=84.50 cr-0=1.21

Thus it is clear that as Xs increases cr-0 decrease which results in decrease of tcr. Inference: The critical clearing time varies inversely as reactance. 3. Initial power output is kept constant at 0.8 p.u. with line reactance also fixed at 0.1 p.u. Inertia constant is modified between 0 to 5 MJ/MVA with the difference of 0.5MJ/MVA. For each case tabulate the critical clearing time for symmetric faults as shown below :-

Inertia Constant 1 1.5 2 2.5 3 3.5 4 4.5 5

Critical Clearing time 0.189 0.228 0.258 0.283 0.308 0.337 0.347 0.381 0.397

variation of Critical Clearing Time With Inertia Constant

0.45 0.4 0.35 0.3 0.25 0.2 0.15 0.1 0.05 0 Inertia Constant 1 1.5 2 2.5 3 3.5 4 4.5 Critical Clearing time

It is observed that the critical clearing time keeps on increasing as we increase the inertial constant.

Explanation of such variation: The critical clearing time given by

Now if other factors are not changed then tcr will remain directly proportional to which is also clear from the variation obtained in the following graph. Inference: The critical clearing time is directly proportional to the square root of inertia constant.

CONCLUSION AND DISCUSSIONS:

Factors affecting transient stability:1. Moment of Inertia: The generators with higher inertial constant are more stable as compared to those having low inertia constant.The critical clearing time varies directly as the square root of moment of inertia. 2. Location of the fault: If the fault is farther from the generator the damage to the system will be lesser than in the case when it occurs at the generator. 3. Line Inductance: More the inductance of a line lesser will be the critical clearing time. Hence the system will have more instability. 4. Type of fault: A three phase symmetric fault is more severe than a single line to ground fault and hence the system becomes more unstable in case of a symmetric fault as compared to a single line to ground fault. 5. Initial Output Power: More is the initial output power more lesser will be the critical clearing time and

the network will be less stable.

Some methods to improve transient stability:The transient stability can be improved by: (i) using either machines of high inertia or connecting the synchronous motors to heavy fly wheels (ii) increasing the system voltage (iii) reducing the transfer reactance (iv) use of AVRs and excitation systems at the time of faults can also help to reduce voltage swings. (v) using high speed circuit breakers (the best method) (vi) using high neutral grounding impedance. In order to improve transient stability voltage regulators employed on the lines should be quick acting ones and the governors attached to the turbines driving alternators should be high speed ones so as to adjust the generator input quickly as per demand of the load.